+ All Categories
Home > Documents > THESIS DOCTOR OF PHILOSOPHY (PhD)

THESIS DOCTOR OF PHILOSOPHY (PhD)

Date post: 09-Dec-2021
Category:
Upload: others
View: 8 times
Download: 0 times
Share this document with a friend
103
University of Veterinary Medicine Hannover Department of Small Animal Medicine and Surgery Central venous catheter induced thrombosis in dogs occurrence under an antithrombotic treatment regimen and possible indicators THESIS Submitted in partial fulfilment of the requirements for the degree DOCTOR OF PHILOSOPHY (PhD) awarded by the University of Veterinary Medicine Hannover by José Mauricio Pereira Mora (Cartago, Costa Rica) Hannover, Germany 2020
Transcript
Page 1: THESIS DOCTOR OF PHILOSOPHY (PhD)

University of Veterinary Medicine Hannover

Department of Small Animal Medicine and Surgery

Central venous catheter induced thrombosis in dogs ndash occurrence

under an antithrombotic treatment regimen and possible indicators

THESIS

Submitted in partial fulfilment of the requirements for the degree

DOCTOR OF PHILOSOPHY

(PhD)

awarded by the University of Veterinary Medicine Hannover

by

Joseacute Mauricio Pereira Mora

(Cartago Costa Rica)

Hannover Germany 2020

Supervisor Prof Dr Reinhard Mischke

Supervision Group Prof Dr Reinhard Mischke

Prof Dr Manfred Kietzmann

PD Dr Mario von Depka Prondzinski

1st Evaluation Prof Dr Reinhard Mischke

Department of Small Animal Medicine and Surgery

University of Veterinary Medicine Hannover Germany

Prof Dr Manfred Kietzmann

Institute for Pharmacology Toxicology and Pharmacy

University of Veterinary Medicine Hannover Germany

PD Dr Mario von Depka Prondzinski

Werlhof Institute Medical Care Center

Hannover Germany

2nd Evaluation

Prof Dr Andreas Moritz

Clinical Pathophysiology and Clinical Laboratory Diagnostics

Justus-Liebig-University Giessen Germany

Date of final exam November 9th 2020

Parts of the thesis have been published as full paper previously in journals

Pereira JM Rohn K Mischke R 2020 Reference intervals for rotational

thromboelastometry measurements using the ROTEMreg delta device in dogs Res Vet

Sci 130 26ndash32

Parts of the thesis have been presented in abstracts related to the poster and oral

presentations at congresses

Pereira JM Mischke R 2017 Reference intervals for thromboelastometric

measurements using the ROTEMreg delta device in dogs In 10th Graduate School

DaysHannover Graduate School for Veterinary Pathobiology Neuroinfectology and

Translational Medicine 01ndash02122017 Bad Salzdetfurth Germany Stiftung

Tieraumlrztliche Hochschule Hannover P 48 P 08

Pereira JM Mischke R 2018 Reference intervals for thromboelastometric

measurements using the ROTEMreg delta device in dogs In 26 Jahrestagung der DVG-

Fachgruppe Innere Medizin und klinische Labordiagnostik 02ndash030218 Hannover

Germany Tierarztl Prax Ausg K Kleintiere Heimtiere 46(02) P18-A16

(DOI 101055s-0038-1647663)

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2018

Central venous catheter associated thrombosis in dogs In 11th Graduate School

DaysHannover Graduate School for Veterinary Pathobiology Neuroinfectology and

Translational Medicine 3011ndash021217 Hannover Germany Stiftung Tieraumlrztliche

Hochschule Hannover V07

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2019

Central venous catheter associated thrombosis in dogs In 27 Jahrestagung der DVG-

Fachgruppe Innere Medizin und klinische Labordiagnostik 01ndash020219 Munich

Germany Tierarztl Prax Ausg K Kleintiere Heimtiere 47(02) P14-139 (DOI

101055s-0039-1679119)

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs In 35th

World Veterinary Association Congress 27 ndash300319 San Joseacute Costa Rica

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs Abstracts

of the ESVCP-ECVCP Annual Congress 25ndash280919 Arnhem The Netherlands Vet

Clin Pathol 48 813 P 57

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2020

Central venous catheter associated thrombosis in dogs In 20 Annual ECVIM-CA

Congress 03ndash050920 Barcelona Spain J Vet Int Med (in press)

Sponsorship Joseacute Mauricio Pereira received a scholarship from the Universidad Nacional

de Costa Rica (reference Code JB-C-1107-2016)

Table of contents

Table of contents

Chapter 1 Introduction 1

Epidemiology of CVC related thrombosis 1

Pathogenesis of CVC related thrombosis 2

Diagnosis of CVC related thrombosis 3

Heparin thromboprophylaxis for CVC related thrombosis 5

Prediction of CVC related thrombosis 7

Aims of the study 8

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs 9

ABSTRACT 11

1 Introduction 12

2 Materials and methods 14

3 Results 20

4 Discussion 26

5 Conclusions 31

6 Acknowledgments 31

7 References 32

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes 37

ABSTRACT 40

1 Introduction 42

2 Material and methods 44

3 Results 54

Table of contents

4 Discussion 62

5 Conclusions 67

6 Acknowledgments 67

7 References 68

Chapter 4 General discussion 72

Incidence of CVC related thrombosis 72

Heparin thromboprophylaxis for CVC related thrombosis 76

Prediction of CVC related thrombosis 77

Conclusion 78

Chapter 5 Summary 79

Chapter 6 Zusammenfassung 82

Chapter 7 References 86

Acknowledgements 91

List of abbreviations

i

List of abbreviations

CV coefficient of variation

CVC central venous catheter

EJV external jugular vein

ETP endogenous thrombin potential

HDL high-density lipoprotein

IV intravenous

LDL low-density lipoprotein

LMWH low molecular weight heparin

ROTEM rotational elastometry

SC subcutaneous

TGA thrombin generation assay

UFH unfractionated heparin

Chapters 2 and 3 have their own abbreviations included

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 2: THESIS DOCTOR OF PHILOSOPHY (PhD)

Supervisor Prof Dr Reinhard Mischke

Supervision Group Prof Dr Reinhard Mischke

Prof Dr Manfred Kietzmann

PD Dr Mario von Depka Prondzinski

1st Evaluation Prof Dr Reinhard Mischke

Department of Small Animal Medicine and Surgery

University of Veterinary Medicine Hannover Germany

Prof Dr Manfred Kietzmann

Institute for Pharmacology Toxicology and Pharmacy

University of Veterinary Medicine Hannover Germany

PD Dr Mario von Depka Prondzinski

Werlhof Institute Medical Care Center

Hannover Germany

2nd Evaluation

Prof Dr Andreas Moritz

Clinical Pathophysiology and Clinical Laboratory Diagnostics

Justus-Liebig-University Giessen Germany

Date of final exam November 9th 2020

Parts of the thesis have been published as full paper previously in journals

Pereira JM Rohn K Mischke R 2020 Reference intervals for rotational

thromboelastometry measurements using the ROTEMreg delta device in dogs Res Vet

Sci 130 26ndash32

Parts of the thesis have been presented in abstracts related to the poster and oral

presentations at congresses

Pereira JM Mischke R 2017 Reference intervals for thromboelastometric

measurements using the ROTEMreg delta device in dogs In 10th Graduate School

DaysHannover Graduate School for Veterinary Pathobiology Neuroinfectology and

Translational Medicine 01ndash02122017 Bad Salzdetfurth Germany Stiftung

Tieraumlrztliche Hochschule Hannover P 48 P 08

Pereira JM Mischke R 2018 Reference intervals for thromboelastometric

measurements using the ROTEMreg delta device in dogs In 26 Jahrestagung der DVG-

Fachgruppe Innere Medizin und klinische Labordiagnostik 02ndash030218 Hannover

Germany Tierarztl Prax Ausg K Kleintiere Heimtiere 46(02) P18-A16

(DOI 101055s-0038-1647663)

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2018

Central venous catheter associated thrombosis in dogs In 11th Graduate School

DaysHannover Graduate School for Veterinary Pathobiology Neuroinfectology and

Translational Medicine 3011ndash021217 Hannover Germany Stiftung Tieraumlrztliche

Hochschule Hannover V07

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2019

Central venous catheter associated thrombosis in dogs In 27 Jahrestagung der DVG-

Fachgruppe Innere Medizin und klinische Labordiagnostik 01ndash020219 Munich

Germany Tierarztl Prax Ausg K Kleintiere Heimtiere 47(02) P14-139 (DOI

101055s-0039-1679119)

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs In 35th

World Veterinary Association Congress 27 ndash300319 San Joseacute Costa Rica

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs Abstracts

of the ESVCP-ECVCP Annual Congress 25ndash280919 Arnhem The Netherlands Vet

Clin Pathol 48 813 P 57

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2020

Central venous catheter associated thrombosis in dogs In 20 Annual ECVIM-CA

Congress 03ndash050920 Barcelona Spain J Vet Int Med (in press)

Sponsorship Joseacute Mauricio Pereira received a scholarship from the Universidad Nacional

de Costa Rica (reference Code JB-C-1107-2016)

Table of contents

Table of contents

Chapter 1 Introduction 1

Epidemiology of CVC related thrombosis 1

Pathogenesis of CVC related thrombosis 2

Diagnosis of CVC related thrombosis 3

Heparin thromboprophylaxis for CVC related thrombosis 5

Prediction of CVC related thrombosis 7

Aims of the study 8

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs 9

ABSTRACT 11

1 Introduction 12

2 Materials and methods 14

3 Results 20

4 Discussion 26

5 Conclusions 31

6 Acknowledgments 31

7 References 32

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes 37

ABSTRACT 40

1 Introduction 42

2 Material and methods 44

3 Results 54

Table of contents

4 Discussion 62

5 Conclusions 67

6 Acknowledgments 67

7 References 68

Chapter 4 General discussion 72

Incidence of CVC related thrombosis 72

Heparin thromboprophylaxis for CVC related thrombosis 76

Prediction of CVC related thrombosis 77

Conclusion 78

Chapter 5 Summary 79

Chapter 6 Zusammenfassung 82

Chapter 7 References 86

Acknowledgements 91

List of abbreviations

i

List of abbreviations

CV coefficient of variation

CVC central venous catheter

EJV external jugular vein

ETP endogenous thrombin potential

HDL high-density lipoprotein

IV intravenous

LDL low-density lipoprotein

LMWH low molecular weight heparin

ROTEM rotational elastometry

SC subcutaneous

TGA thrombin generation assay

UFH unfractionated heparin

Chapters 2 and 3 have their own abbreviations included

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 3: THESIS DOCTOR OF PHILOSOPHY (PhD)

Parts of the thesis have been published as full paper previously in journals

Pereira JM Rohn K Mischke R 2020 Reference intervals for rotational

thromboelastometry measurements using the ROTEMreg delta device in dogs Res Vet

Sci 130 26ndash32

Parts of the thesis have been presented in abstracts related to the poster and oral

presentations at congresses

Pereira JM Mischke R 2017 Reference intervals for thromboelastometric

measurements using the ROTEMreg delta device in dogs In 10th Graduate School

DaysHannover Graduate School for Veterinary Pathobiology Neuroinfectology and

Translational Medicine 01ndash02122017 Bad Salzdetfurth Germany Stiftung

Tieraumlrztliche Hochschule Hannover P 48 P 08

Pereira JM Mischke R 2018 Reference intervals for thromboelastometric

measurements using the ROTEMreg delta device in dogs In 26 Jahrestagung der DVG-

Fachgruppe Innere Medizin und klinische Labordiagnostik 02ndash030218 Hannover

Germany Tierarztl Prax Ausg K Kleintiere Heimtiere 46(02) P18-A16

(DOI 101055s-0038-1647663)

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2018

Central venous catheter associated thrombosis in dogs In 11th Graduate School

DaysHannover Graduate School for Veterinary Pathobiology Neuroinfectology and

Translational Medicine 3011ndash021217 Hannover Germany Stiftung Tieraumlrztliche

Hochschule Hannover V07

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2019

Central venous catheter associated thrombosis in dogs In 27 Jahrestagung der DVG-

Fachgruppe Innere Medizin und klinische Labordiagnostik 01ndash020219 Munich

Germany Tierarztl Prax Ausg K Kleintiere Heimtiere 47(02) P14-139 (DOI

101055s-0039-1679119)

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs In 35th

World Veterinary Association Congress 27 ndash300319 San Joseacute Costa Rica

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs Abstracts

of the ESVCP-ECVCP Annual Congress 25ndash280919 Arnhem The Netherlands Vet

Clin Pathol 48 813 P 57

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2020

Central venous catheter associated thrombosis in dogs In 20 Annual ECVIM-CA

Congress 03ndash050920 Barcelona Spain J Vet Int Med (in press)

Sponsorship Joseacute Mauricio Pereira received a scholarship from the Universidad Nacional

de Costa Rica (reference Code JB-C-1107-2016)

Table of contents

Table of contents

Chapter 1 Introduction 1

Epidemiology of CVC related thrombosis 1

Pathogenesis of CVC related thrombosis 2

Diagnosis of CVC related thrombosis 3

Heparin thromboprophylaxis for CVC related thrombosis 5

Prediction of CVC related thrombosis 7

Aims of the study 8

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs 9

ABSTRACT 11

1 Introduction 12

2 Materials and methods 14

3 Results 20

4 Discussion 26

5 Conclusions 31

6 Acknowledgments 31

7 References 32

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes 37

ABSTRACT 40

1 Introduction 42

2 Material and methods 44

3 Results 54

Table of contents

4 Discussion 62

5 Conclusions 67

6 Acknowledgments 67

7 References 68

Chapter 4 General discussion 72

Incidence of CVC related thrombosis 72

Heparin thromboprophylaxis for CVC related thrombosis 76

Prediction of CVC related thrombosis 77

Conclusion 78

Chapter 5 Summary 79

Chapter 6 Zusammenfassung 82

Chapter 7 References 86

Acknowledgements 91

List of abbreviations

i

List of abbreviations

CV coefficient of variation

CVC central venous catheter

EJV external jugular vein

ETP endogenous thrombin potential

HDL high-density lipoprotein

IV intravenous

LDL low-density lipoprotein

LMWH low molecular weight heparin

ROTEM rotational elastometry

SC subcutaneous

TGA thrombin generation assay

UFH unfractionated heparin

Chapters 2 and 3 have their own abbreviations included

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 4: THESIS DOCTOR OF PHILOSOPHY (PhD)

Translational Medicine 3011ndash021217 Hannover Germany Stiftung Tieraumlrztliche

Hochschule Hannover V07

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2019

Central venous catheter associated thrombosis in dogs In 27 Jahrestagung der DVG-

Fachgruppe Innere Medizin und klinische Labordiagnostik 01ndash020219 Munich

Germany Tierarztl Prax Ausg K Kleintiere Heimtiere 47(02) P14-139 (DOI

101055s-0039-1679119)

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs In 35th

World Veterinary Association Congress 27 ndash300319 San Joseacute Costa Rica

Pereira JM Rohn K Mischke R 2019 Reference intervals for rotational

thromboelastometric measurements using the ROTEMreg delta device in dogs Abstracts

of the ESVCP-ECVCP Annual Congress 25ndash280919 Arnhem The Netherlands Vet

Clin Pathol 48 813 P 57

Pereira JM Hewicker-Trautwein M von Depka Prondzinski M Mischke R 2020

Central venous catheter associated thrombosis in dogs In 20 Annual ECVIM-CA

Congress 03ndash050920 Barcelona Spain J Vet Int Med (in press)

Sponsorship Joseacute Mauricio Pereira received a scholarship from the Universidad Nacional

de Costa Rica (reference Code JB-C-1107-2016)

Table of contents

Table of contents

Chapter 1 Introduction 1

Epidemiology of CVC related thrombosis 1

Pathogenesis of CVC related thrombosis 2

Diagnosis of CVC related thrombosis 3

Heparin thromboprophylaxis for CVC related thrombosis 5

Prediction of CVC related thrombosis 7

Aims of the study 8

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs 9

ABSTRACT 11

1 Introduction 12

2 Materials and methods 14

3 Results 20

4 Discussion 26

5 Conclusions 31

6 Acknowledgments 31

7 References 32

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes 37

ABSTRACT 40

1 Introduction 42

2 Material and methods 44

3 Results 54

Table of contents

4 Discussion 62

5 Conclusions 67

6 Acknowledgments 67

7 References 68

Chapter 4 General discussion 72

Incidence of CVC related thrombosis 72

Heparin thromboprophylaxis for CVC related thrombosis 76

Prediction of CVC related thrombosis 77

Conclusion 78

Chapter 5 Summary 79

Chapter 6 Zusammenfassung 82

Chapter 7 References 86

Acknowledgements 91

List of abbreviations

i

List of abbreviations

CV coefficient of variation

CVC central venous catheter

EJV external jugular vein

ETP endogenous thrombin potential

HDL high-density lipoprotein

IV intravenous

LDL low-density lipoprotein

LMWH low molecular weight heparin

ROTEM rotational elastometry

SC subcutaneous

TGA thrombin generation assay

UFH unfractionated heparin

Chapters 2 and 3 have their own abbreviations included

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 5: THESIS DOCTOR OF PHILOSOPHY (PhD)

Table of contents

Table of contents

Chapter 1 Introduction 1

Epidemiology of CVC related thrombosis 1

Pathogenesis of CVC related thrombosis 2

Diagnosis of CVC related thrombosis 3

Heparin thromboprophylaxis for CVC related thrombosis 5

Prediction of CVC related thrombosis 7

Aims of the study 8

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs 9

ABSTRACT 11

1 Introduction 12

2 Materials and methods 14

3 Results 20

4 Discussion 26

5 Conclusions 31

6 Acknowledgments 31

7 References 32

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes 37

ABSTRACT 40

1 Introduction 42

2 Material and methods 44

3 Results 54

Table of contents

4 Discussion 62

5 Conclusions 67

6 Acknowledgments 67

7 References 68

Chapter 4 General discussion 72

Incidence of CVC related thrombosis 72

Heparin thromboprophylaxis for CVC related thrombosis 76

Prediction of CVC related thrombosis 77

Conclusion 78

Chapter 5 Summary 79

Chapter 6 Zusammenfassung 82

Chapter 7 References 86

Acknowledgements 91

List of abbreviations

i

List of abbreviations

CV coefficient of variation

CVC central venous catheter

EJV external jugular vein

ETP endogenous thrombin potential

HDL high-density lipoprotein

IV intravenous

LDL low-density lipoprotein

LMWH low molecular weight heparin

ROTEM rotational elastometry

SC subcutaneous

TGA thrombin generation assay

UFH unfractionated heparin

Chapters 2 and 3 have their own abbreviations included

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 6: THESIS DOCTOR OF PHILOSOPHY (PhD)

Table of contents

4 Discussion 62

5 Conclusions 67

6 Acknowledgments 67

7 References 68

Chapter 4 General discussion 72

Incidence of CVC related thrombosis 72

Heparin thromboprophylaxis for CVC related thrombosis 76

Prediction of CVC related thrombosis 77

Conclusion 78

Chapter 5 Summary 79

Chapter 6 Zusammenfassung 82

Chapter 7 References 86

Acknowledgements 91

List of abbreviations

i

List of abbreviations

CV coefficient of variation

CVC central venous catheter

EJV external jugular vein

ETP endogenous thrombin potential

HDL high-density lipoprotein

IV intravenous

LDL low-density lipoprotein

LMWH low molecular weight heparin

ROTEM rotational elastometry

SC subcutaneous

TGA thrombin generation assay

UFH unfractionated heparin

Chapters 2 and 3 have their own abbreviations included

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 7: THESIS DOCTOR OF PHILOSOPHY (PhD)

List of abbreviations

i

List of abbreviations

CV coefficient of variation

CVC central venous catheter

EJV external jugular vein

ETP endogenous thrombin potential

HDL high-density lipoprotein

IV intravenous

LDL low-density lipoprotein

LMWH low molecular weight heparin

ROTEM rotational elastometry

SC subcutaneous

TGA thrombin generation assay

UFH unfractionated heparin

Chapters 2 and 3 have their own abbreviations included

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 8: THESIS DOCTOR OF PHILOSOPHY (PhD)

List of figures and tables

ii

List of figures and tables

Chapter 2

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples 20

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations 20

Table 3 Reference values (median reference interval based on 25 - and 975 -

quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) 23

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using ex-tem reagent stratified by sex and neuter status 24

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg

variables using the in-tem reagent stratified by sex and neutering status 25

Chapter 3

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound 48

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy 53

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy 53

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique 56

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 9: THESIS DOCTOR OF PHILOSOPHY (PhD)

List of figures and tables

iii

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration 57

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection 57

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy 58

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses 60

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy 61

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 10: THESIS DOCTOR OF PHILOSOPHY (PhD)

Chapter 1 Introduction

1

Chapter 1 Introduction

Central venous catheters (CVC) are mainly used at critical care units in human and small

animal patients requiring infusion of large amounts of fluids or blood parenteral nutrition

support routine drug therapy haemodialysis chemotherapy plasmapheresis or

periodical blood sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is

one of the most severe complications (apart from infections and pneumothorax) of CVCs

in humans (Comerlato et al 2017 Evans and Ratchford 2018 Geerts 2014) leading to

patientrsquos distress CVC dysfunction venous stenosis and in some cases to the

development of thromboembolism (Geerts 2014) In the human literature there are

various studies dealing with epidemiologic and prophylactic aspects which are described

below in detail In contrast to the best of the authoracutes knowledge studies about extra

luminal CVC related thrombosis in dogs are rare in the available veterinary literature

Studies are mainly dealing with development of intraluminal CVC thrombi occurring

frequently in dogs and cats (Langston et al 2014) and leading to CVC malfunction (Vose

et al 2019)

Epidemiology of CVC related thrombosis

CVC related thrombosis represents 70ndash80 of all upper extremity thrombosis cases and

approximately 10 of thromboembolism cases in humans (Kucher 2011) The incidence

of CVC related thrombosis reported in humans based in 25 studies ranged widely from

2 to 67 (median = 30 ) (van Rooden et al 2005) The incidence of CVC related

thrombosis detected by colour Doppler ultrasound of the internal jugular vein reported in

3 studies performed in humans was 2 (Harter et al 2002) 30 (Lordick et al 2003)

and 56 (Wu et al 1999)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 11: THESIS DOCTOR OF PHILOSOPHY (PhD)

Chapter 1 Introduction

2

The incidence of CVC related thrombosis in humans can vary between studies according

to the CVC type used entry site of the CVC (jugular femoral subclavian axillary

cephalic vein) and time period with the CVC in place the diagnostic thrombus detection

technique (venography or ultrasonography) and criteria for diagnosis duration of follow

up diseases and presence of predisposing factors (eg previous venous thrombosis

history presence of cancer) and finally prophylactic anticoagulation treatment (Geerts

2014 Liu et al 2015 van Rooden et al 2005)

To the best of the authoracutes knowledge data about the incidence of CVC related

thrombosis detected by colour Doppler ultrasound into the lumen of the jugular vein in

dogs are rare in the available veterinary literature In one study performed in 81 dogs and

12 cats (Adamantos et al 2010) the incidence of vein thrombosis detected by palpation

or ultrasound was 2 and the incidence of intraluminal CVC thrombosis (presumed by

the inability to draw blood back) was 15 Additionally regarding to intraluminal CVC

thrombosis an incidence of 8 was reported in one case control study performed in 24

dogs receiving CVC flushing with heparinised solution (Vose et al 2019) Data about

incidence in other animal species was also found In another case-control study performed

on 37 pigs thrombosis in the caval vein related to CVC placed in the external jugular

vein (EJV) was evaluated by histological analysis however no ultrasound detection was

performed (Kohler and Kirkman 1998) In this study in pigs 94 of patients with CVC

developed thrombi in the vena cava based on sections at the site of the end of the catheter

tip and 05 and 1 cm proximal of the first section (Kohler and Kirkman 1998)

Pathogenesis of CVC related thrombosis

Two types of CVC related thrombosis are distinguished in the human medicine literature

(Geerts 2014 van Rooden et al 2005) First symptomatic thrombosis is defined as

Chapter 1 Introduction

3

thrombosis detected by diagnostic images (ultrasonography venography) when

symptoms and signs (eg localised swelling pain redness) are present (Lee and

Kamphuisen 2012) Second subclinical thrombosis is defined as thrombosis

demonstrated by the previously mentioned diagnostic images without overt symptoms or

signs (Boersma et al 2008) Additionally the CVC related thrombi can be classified as

occlusive thrombus into the CVC lumen peri-catheter related fibrin sheaths and mural

thrombi (Geerts 2014)

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to (1) (partial or complete) occlusion of the catheter lumen and (2)

sometimes if a thrombus formation starts into the lumen of the vein (attached to the vein

wall or to the CVC) a partial or complete obstruction of the blood flow andor embolism

can develop (Langston and Eatroff 2018) Additionally a study performed in humans

(Lordick et al 2003) demonstrated a relationship between CVC related thrombosis and

microbial colonisation of the catheter leading to CVC related systemic infections This

relationship was explained among others by the ease of microorganisms like

Staphylococcus aureus and Staphylococcus epidermidis to adhere to CVC related

thrombin sheaths (Boersma et al 2008)

Diagnosis of CVC related thrombosis

Venous thrombosis is commonly diagnosed in human and small animal medicine

(especially in dogs) with vascular ultrasonography with or without use of the colour

Doppler technique which is normally used to evaluate the blood flow through the blood

Chapter 1 Introduction

4

vessels (Evans and Ratchford 2018 Laurenson et al 2010) According to the literature

in humans cited below the CVC related thrombosis detected by colour Doppler

ultrasound is present if any of the following aspects are observed (1) well defined

thrombus visualised in the lumen of the vein (2) impossibility to perform compression

of the vein by slight pressure with the ultrasound probe andor (3) no detectable blood

flow signal by colour Doppler ultrasound (Bonnet et al 1989 Koumlksoy et al 1995)

Similarly in dogs the CVC related thrombosis is usually diagnosed by colour Doppler

ultrasonography through the detection of the thrombus within the lumen of the vein

andor the obstruction (partial or complete) of the normal blood flow (Bliss et al 2002)

Especially in humans the contrast venography technique is used as gold standard to

confirm thrombosis For this purpose a contrast solution is injected through a peripheral

vein to assess its passage through the bloodstream by x-ray However this diagnostic

technique is normally used only in the case of a strong clinical suspicion of venous

thrombosis that cannot be detected by ultrasonography (Laurenson et al 2010 Langston

and Eatroff 2018) In a study performed in 44 human patients requiring a CVC (Koumlksoy

et al 1995) a sensitivity of 94 and a specificity of 96 were determined for the colour

Doppler ultrasound diagnosis when it was compared with the diagnosis performed with

the contrast venography technique That results make the colour Doppler ultrasound an

accurate diagnosis method to detect CVC related thrombosis in humans (Koumlksoy et al

1995 van Rooden et al 2003)

Advanced diagnostic image techniques like computed tomography or magnetic resonance

tomography are also accurate to perform the diagnosis of CVC related thrombosis in

humans (Whitesell and Steenburg 2014) In dogs multidetector computed tomography

venography was used in one study for the diagnosis of CVC related thrombosis allowing

the identification of catheter related sheaths around the CVC external wall (Wang at al

Chapter 1 Introduction

5

2015) However the high costs of these advanced diagnostic imaging procedures and the

requirement to perform at least sedation or general anesthesia are disadvantages when

compared with ultrasonography

Finally electron microscopy (ELMI) is a diagnostic technique that has been used in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) mainly to perform a microstructural evaluation of thrombi developed into the

lumen of the CVC or in the vein containing the CVC ELMI allows the ex vivo diagnosis

of very small thrombi or sleeve related thrombosis that perhaps cannot be detected by

ultrasonography or venography during the evaluation of patients with CVCs (Xiang et al

2001) However one point to consider is that it is possible that compact thrombi

developed on the CVC can be detached during the CVC removal from the vein and

thereby before its analysis by ELMI (Xiang et al 2001)

Heparin thromboprophylaxis for CVC related thrombosis

Application of CVCs in human medicine is commonly accompanied by prophylactic

anticoagulation in order to decrease the risk of development of partial or complete vein

occlusion andor thromboembolism and thereby to improve the patientrsquos comfort

(Baumann-Kreuziger et al 2015) Apart from vitamin K antagonists (Mismetti et al

2003) unfractionated heparin (UFH) (Abdelfeki et al 2004) and low molecular weight

heparin (LMWH) (Karthaus et al 2006) are anticoagulants that have been used on

studies in humans administrated by intravenous (IV) infusion or by subcutaneous (SC)

injection to prevent thrombi development into the lumen of the veins containing the CVC

(review article Baumann-Kreuziger et al 2015) Different heparin dosages have been

used as thromboprophylactic anticoagulation in humans with CVC (review article Lee

and Kamphuisen 2012) For example in five different case-control studies performed in

Chapter 1 Introduction

6

human haemato-oncological patients UFH at a daily dosage of 100 IUkg given by IV

infusion (Abdelfeki et al 2004) and LMWH at SC daily dosages of 40 mg enoxaparin

(Verso et al 2005) 5000 IU AXa dalteparin (Karthaus et al 2006 De Cicco et al 2009)

and 2850 IU AXa nadroparin (Niers et al 2007) were used Only in part of these studies

it is stated that systemic heparin was combined with flushing of the CVC using

heparinised saline solution to reduce the risk of catheter occlusion (Karthaus et al 2006)

In 3 of the 5 previously mentioned studies no statistically significant difference was found

in the incidence of CVC related thrombosis between patients receiving the

anticoagulation prophylaxis compared with the control groups without systemic

anticoagulation (Verso et al 2005 p = 035 Karthaus et al 2006 p gt 005 Niers et al

2007 p = 049) In contrast in the residual two studies a lower incidence of CVC related

thrombosis was found in the group receiving an anticoagulation prophylaxis compared

with the control groups (Abdelfeki et al 2004 150 vs 126 p = 003 De Cicco et

al 2009 400 vs 526 p = 005)

Obviously there is no consensus in the available human literature regarding the beneficial

use of anticoagulation prophylaxis to prevent CVC related thrombosis and this may

reflect great variations for example according to the study design period trial drug

administration route (Lee and Kamphuisen 2012) and the moment of the initial

anticoagulant dosage administration (Decicco et al 2009)

To the best of the authoracutes knowledge studies about thromboprophylaxis for extraluminal

CVC related thrombosis in dogs cannot be found in the available veterinary literature

However similarly to humans prophylactic anticoagulation with heparinised saline

(administrated through CVC flushing) to prevent CVC occlusion by intraluminal thrombi

was used in one case-control study performed in 24 dogs (Vose et al 2019) In the cited

study 12 dogs received dosages of 3 ml of heparinised saline solution (10 IU UFHml) to

Chapter 1 Introduction

7

flush the catheter every 6 hours during 72 hours whereas 12 control dogs received

flushing with the same volume of non-heparinised saline solution administered at the

same frequency and no statistically significant difference was found

Prediction of CVC related thrombosis

Studies regarding the relationship between laboratory biomarkers and CVC related

thrombosis are available in the human literature (Boersma et al 2016 Cheng et al 2013

Liu et al 2015) however to the best of the authoracutes knowledge literature about laboratory

tests used as predictors for CVC related thrombosis in small animals cannot be found at

all Availability of approved blood tests which can predict the risk of thrombosis

development associated to CVC already at the moment of the CVC placement are

important in order to identify an increased risk of CVC related thrombosis (Boersma et

al 2016) In the study performed by Boersma et al (2016) in 168 human patients without

indication of anticoagulation prophylaxis elevated leukocyte counts high levels of

plasminogen activator inhibitor and of coagulation factor VIII (all measured on blood

samples taken after CVC placement) were associated with an increased incidence of

symptomatic CVC-related thrombosis diagnosed by ultrasonography In another study

(Liu et al 2015) high blood levels of lactate dehydrogenase and low concentrations of

HDL and albumin were associated with the incidence for CVC related thrombosis

diagnosed by ultrasound in 324 senile patients which was in total 617 Additionally

dyslipidemia (high triglycerides low HDL and high LDLHDL ratio) was associated

with CVC related thrombosis (diagnosed by ultrasound and computer tomography

angiography) in human patients undergoing haemodialysis (Cheng et al 2013) Studies

about predictive tests for CVC related thrombosis in dogs were not found in the available

literature to the best of the authoracutes knowledge

Chapter 1 Introduction

8

Aims of the study

The aims of the preliminary methodological study of the present scientific project were

(1) to define reference intervals for viscoelastic analyses of canine haemostasis using the

ROTEMreg delta analyser and (2) to determine the precision (repeatability) of this

method And the aims of the main study of the present scientific project were (1) to

determine the incidence of catheter-induced thrombosis in canine intensive care patients

(2) to evaluate the efficacy of a prophylactic heparin regimen routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

9

Chapter 2 Reference intervals for rotational thromboelastometry measurements

using the ROTEMreg delta device in dogs

Pereira JM Rohn K Mischke R

Contributions of the authors

Pereira JM performed (1) the patients data and blood samples collection (2)

haematology and ROTEMreg analysis (3) generated the reference intervals and

coefficients of variation included in the study and (4) wrote the manuscript

Rohn K performed the statistical analysis

Mischke R performed the clinical chemistry and coagulation tests and was involved in

designing and supervision of the study analysis and discussion of the data and its

statistical analysis and revision edition and correction of the manuscript

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

10

Reference intervals for rotational thromboelastometry measurements using the

ROTEMreg delta device in dogs

Pereira JMa Rohn Kb Mischke Ra

aSmall Animal Clinic University of Veterinary Medicine Hannover Buumlnteweg 9 D-

30559 Hannover Germany

bInstitute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Buumlnteweg 2 30559 Hannover Germany

Corresponding author

E-mail ReinhardMischketiho-hannoverde

Original source Pereira J M Rohn K amp Mischke R 2020 Reference intervals

for rotational thromboelastometry measurements using the ROTEMreg delta device

in dogs Res Vet Sci 130 26ndash32 httpsdoiorg101016jrvsc202001019

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

11

ABSTRACT

Aims of the present study were to define reference intervals for viscoelastic analyses of

canine haemostasis using the ROTEMreg delta analyser and as a secondary aspect to

determine the precision (repeatability) of this method Blood samples from 125 clinically

healthy dogs were included Measurements were performed with commercially available

activating reagents (ex-tem in-tem and kaolin solution) as well as without activation

Additional fourfold measurements were done in 3 of the normal blood samples and in 3

samples with haemostatic alterations to evaluate the precision of the method

Coefficients of Variation (CVs) for most of the ROTEM variables were lt 10 Clot

formation time and maximum clot elasticity showed a wide inter-individual variation in

comparison with alpha angle and maximum clot firmness A multivariate analysis on

various ROTEM parameters revealed particularly a significant influence of neuter status

and a significant interrelationship between the factors sex and neuter status for

measurements with different activating reagents These results reflected the fact that

significant differences occurred only between intact and neutered females but not in

males No or only occasionally significant differences were found between groups of sex

age and size

In conclusion CVs demonstrated that the method delivers repeatable results in canine

citrated whole blood Established reference intervals should deliver valuable orientation

for the evaluation of viscoelastic properties of clotting whole blood in dogs using the

ROTEM delta analyser Neuter status in females appeared to be the most relevant

influencing factor and should be considered for the interpretation of ROTEM delta test

results

Keywords Viscoelastic measurements reference values canine age sex neuter status

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

12

1 Introduction

Heamostasis is a physiologic and dynamic process in which plasma soluble coagulation

factors blood cells and blood vessels are involved Clot formation and possible

subsequent clot lysis as well as vascular tissue repair aim to prevent haemorrhages on the

one hand and thromboses on the other hand (Kol and Borjesson 2010 Versteeg et al

2013) Conventional assays of coagulation such as prothrombin time (PT) activated

partial thromboplastin time (aPTT) and thrombin time TT have limitations especially

cellular elements are not incorporated into the test and thereby these tests reflect only

limitedly in vivo functionality of haemostasis (McMichael and Smith 2011 Crochemore

et al 2018)

The rotational thromboelastometry (ROTEMreg) currently termed as thromboelastometry

(TEM) and thromboelastography (TEG) are methods for the rapid evaluation of the

overall haemostasis capacity through the measurement of viscoelastic properties of

clotting whole blood (Kol and Borjesson 2010 Theusinger et al 2010 Sankarankutty

et al 2012 Versteeg et al 2013) In TEG the cup with the blood sample is rotating

whereas the torsion wire is fixed In ROTEMTEM the cup is fixed whereas the pin is

rotating Changes in torque ie resistance to rotation by the fibrin bridge created between

pin and cup in the clotting blood are detected electromechanically in TEG and optically

in ROTEM (Bollinger et al 2012) and displayed as a graph that indicates the time for the

formation of blood clot and its strength (Bagge et al 2016) The resulting graph reflects

the initiation and propagation of the coagulation system fibrin-platelet interactions and

clot lysis

TEG was first described in 1948 by Hellmut Harten (Hartert 1948) as an in vitro method

to evaluate the clot formation process while resembling the in vivo blood flow The

development of the actually used instrument types TEG and ROTEMregTEM were based

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

13

on that original principle Viscoelastic measurements have been widely used since the

1980s in human medicine to evaluate haemostasis and hypercoagulable states in surgical

patients (eg cardiac and hepatic surgeries) and in general in critical care units (Kol and

Borjesson 2010 Nogami 2016 Wikkelso et al 2016 Crochemore et al 2017) In

veterinary medicine viscoelastic measurements using TEG and ROTEMTEM have been

validated for use in dogs cats and horses and used to study haemostasis in different

diseases (eg infectious diseases liver diseases) and medical conditions (eg

orthopaedic surgery hyperfibrinolysis) and also to investigate the effect of

anticoagulants and other drugs (eg corticosteroids) on haemostasis (Kol and Borjesson

2010 Bruno et al 2015a 2015b Woodman et al 2015 Conversy et al 2017 Sigrist et

al 2018 Turner et al 2019) The methodology seems to be especially suitable to detect

hypercoagulability (Wiinberg et al 2008 Kol and Borjesson 2010 Marschner et al

2018)

In the available veterinary literature information about reference values for TEG were

found using kaolin as activation reagent (Bauer et al 2009) Falco et al (2012)

determined reference values for the ROTEM analyser in dogs using ex-tem in-tem and

fib-tem reagents for selected ROTEM parameters in a study on forty five healthy dogs

However data related to reference intervals obtained in an adequate number of samples

(at least 120) according to the American Society for Veterinary Clinical Pathology

(ASVCP) (Friedrichs et al 2012) and the National Committee for Clinical Laboratory

Standards (NCCLS) (Horn and Pesce 2003) could not be found in the available literature

The aim of this study is to determine reference values for the ROTEMreg delta device in

dogs using citrated whole blood with different clot activator reagents (ex-tem in-tem

kaolin) and without activator

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

14

2 Materials and methods

21 Study design

Reference intervals Single thromboelastometric measurements were performed in 125

clinically healthy dogs using the ROTEMreg analyser and different activating reagents (ex-

tem in-tem and kaolin) as well as without activation Reference values were calculated

and influences of age groups (young adult and senior) sex neuter status and size

(groups small medium and large) were assessed

Precision analysis Additional fourfold measurements using ex-tem or in-tem activating

reagents respectively were performed on 3 of the normal and also in 3 abnormal blood

samples each in order to assess within-run imprecision for these measurements

22 Animals

The clinically healthy dogs with a body weight of at least 5 kg were recruited by launching

a call in the University of Veterinary Medicine Hannover network for students and staff

members Interested dog owners received a free health check of their animals (clinical

examination and blood tests specified below) in combination with the blood collection

In addition individual privately owned patient dogs of the Small Animal Clinic with

undisturbed general health were also included if blood was collected due to another

indication (eg general health check pre-anaesthetic examination test for hereditary

diseases etc) Before taking the blood samples all dogs were fasted for at least 12 h with

free access to water 125 animals for which the examination of the health status (clinical

examination routine haematology and clinical chemistry basis coagulation profile) did

not reveal any deviations from the reference were included in the study whereas 3 of the

128 initially recruited animals had to be excluded due to laboratory abnormalities

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

15

Haematological and clinical chemistry profiles performed in order to confirm the health

status included white and red cell count haematocrit platelet counts and reticulocytes

plasma activities of alanine aminotransferase glutamate dehydrogenase alkaline

phosphatase and pseudocholinesterase as well as concentrations of total bilirubin urea

creatinine glucose total protein albumin cholesterol and ionised calcium A basis

coagulation profile including PT PTT and TT was also performed to exclude significant

abnormalities within the blood coagulation system

The 125 healthy dogs enclosed in the establishment of reference values included 48 mixed

breed dogs and the following pedigree dogs 13 Labrador Retrievers 6 Australian

Shepherds 4 Havanese dogs 3 from each of the following breeds American

Staffordshire Terriers and Cocker Spaniels 2 from each of the following breeds Beagles

Border Collies Dalmatians Elo dogs (breed not recognised by Feacutedeacuteration cynologique

international [FCI]) German Longhaired Pointers German Shorthaired Pointers Giant

Schnauzers and Golden Retrievers 1 from each of the following breeds Belgian

Malinois Bolonka Zwetna Briard Brussels Griffon American English Coonhound

Italian Mastiff Flat Coated Retriever French Bulldog German Shepherd Great Dane

Greater Swiss Mountain Dog Giant Spitz Goldendoodle Hovawart Hungarian Vizsla

Italian Bloodhound Lycanis Wolfdog Miniature American Shepherd Miniature

Schnauzer Old German Shepherd Podenco Poodle Pyrenean Shepherd Rhodesian

Ridgeback Rottweiler Schafpudel (breed not recognised by FCI) Small

Munsterlander Spanish Mastiff Tibet Terrier Weimaraner West Highland White

Terrier and Wirehaired Pointer

These dogs were aged between 6 and 154 months (median = 410 months) and had body

weights ranging from 50 to 707 kg (median = 226 kg) 30 were castrated males 33

intact males 41 spayed females and 21 intact females According to the height at withers

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

16

the dogs were divided into 3 different size groups small (le40 cm n = 22) medium (41ndash

60 cm n = 77) and large (gt60 cm n = 26) Dogs were also divided in three different age

groups young (le 24 months n = 34) adult (25ndash84 months n = 59) and senior dogs (gt84

months n = 32)

The three blood samples with haemostatic abnormalities were taken from two dogs with

complex haemostatic disorders (Disseminated Intravascular Coagulation [DIC])

including reduced platelet count and abnormal PT and APTT and one dog with

thrombocytopenia

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine Hannover Foundation and by the ethics committee of the

responsible authority (Lower Saxony State Office for Consumer Protection and Food

Safety reference number 17A101)

23 Blood sampling

The venipuncture area was shaved and then disinfected with alcohol The saphenous or

cephalic vein was visualised with slight pressure and punctured with disposable cannulas

(diameter 09 or 10 mm) The continuously running blood was directly collected from

the needle into the tubes containing the respective anticoagulants From each dog

approximately 10 ml of blood were collected including approximately 8 ml of citrated

blood for the measurement of haemostasis tests (ROTEM PT aPTT TT) in tubes

containing sodium citrate 313 (Eifelfangoreg GmbH amp Co KG Bad Neuenahr-

Ahrweiler Germany) in a ratio of 1 volume9 volumes of blood In addition

approximately 1 ml of EDTA blood for haematological analysis and approximately 1 ml

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

17

of lithium heparin blood for clinical chemical measurements were collected into

commercial 13 ml tubes (Sarstedt AG amp Co Nuumlmbrecht Germany)

Citrated blood was initially kept at room temperature (20-25 C) in order to avoid errors

within the preanalytical phase (Toulon et al 2017) Two ml of citrated blood were used

for the ROTEMreg analysis the residual citrate blood was centrifuged for 10 min at 3800

rpm and 5 degC then the plasma was separated by careful pipetting and it was centrifuged

a second time under the same conditions to get platelet poor plasma which was stored in

aliquots of 500 microl and 250 microl at -80 degC for coagulation tests

24 Rotational elastometry

Viscoelastic measurements using the ROTEMreg delta for the establishment of reference

intervals were performed approximately 15 min after the blood collection using activating

reagents (liquid reagents) provided by the manufacturer of the instrument (r-ex-temreg in-

temreg) and a kaolin solution (5 g1 as part of the PTT reagent CK Prest [Diagnostica

Stago SAS Asniegraveres sur Seine Cedex France]) In addition non-activated analyses

were performed using isotonic NaCl solution instead of the activating reagent in the fourth

channel of the device Fourfold measurements for precision analysis were performed

simultaneously in the four channels of the device in separate runs

According to the standard test protocol provided by the manufacturer of the ROTEMreg

delta (Werfen GmbHInstrumentation Laboratory Munich Germany) 20 μl of the

reagent for recalcification (star-temreg20) and 20 μl of the activating reagent were pipetted

into a cuvette using the automated pipette provided with the device Immediately

thereafter 300 μl of citrated whole blood (prewarmed to 37 degC for five minutes in the

heating device of the instrument and gently rocked to resuspend possibly sedimented

cells) were added to the test preparation in the cuvette using the automated pipette and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

18

then the cuvette was laced carefully in the ROTEMreg device whereby the rotating pin

entered into the test approach

Different parameters are provided automatically by the ROTEMreg delta device and the

following variables (described according to the Pentapharm GmbH ROTEMreg delta

Manual [2009]) were used in the study

Clotting time (CT) the time (in s) from the beginning of the test (by adding the clot

activator) until the formation of fibrin starts (defined as the moment when the curve

achieves an amplitude of 2 mm) This Parameter facilitates the decision to replace clotting

factors when it is required

Clot formation time (CFT) describes the next phase of clotting by the formation of a

stable clot through the activation of thrombocytes and fibrin formation It is defined as

the time (in s) between the 2 mm amplitude and 20 mm amplitude of the curve

Alpha angle (α angle) is the angle (in degrees) between the middle axis and the tangent

to the clotting curve through the 2 mm amplitude point

Amplitude after ldquoXXrdquo minutes (eg A30 after 30 min) is the amplitude (in mm) at a

certain time after the CT has finished

Maximum clot firmness (MCF) corresponds to the maximum amplitude (in mm) reached

before the clot is dissolved by fibrinolysis This parameter measures the firmness of the

clot and its quality Like CFT and α angle it facilitates the decision for substitution therapy

with platelet concentrate andor fibrinogen especially in humans

Maximum clot elasticity (MCE) is a parameter calculated from MCF and may be superior

at high amplitudes as compared to the MCF

Lysis index at 30 45 and 60 min (LI30 LI45 LI60) represents the fibrinolysis (in

percentage) 30 45 and 60 min after the CT It may facilitate the decision for or against a

therapy with anti-fibrinolytic drugs

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

19

Maximum lysis (ML) describes the degree of fibrinolysis relative to the MCF achieved

during the measurement (percent loss of clot firmness)

25 Statistical methods

Original data obtained from the ROTEMreg delta device software were transferred into the

Microsoftreg Excelreg formal (Microsoft Office Professional Plus 2013) The Statistical

analyses were performed using the SAS Enterprise Guide Version 71 (SAS Institute Inc

Cary NC USA)

Coefficients of Variation (CVs) for selected relevant parameters of the ROTEM delta

analysis using ex-tem and in-tem regents were calculated for the fourfold measurements

on 3 normal and 3 abnormal blood samples in order to assess within-run imprecision CVs

for the results of each sample were calculated by dividing the standard deviation of the

data set by its arithmetic mean Finally the median of the CVs for each group of 3 samples

was calculated The reference values were calculated according to recommendations of

the American College of Veterinary Clinical Pathology (ACVCP) based on non-

parametric methods (25 and 975 quantiles) supplemented by 95 confidence

intervals (Friedrichs et al 2012)

Data were analysed for Standard normal distribution by Kolmogorov Smirnov test and

assessment of residuals of the analysis of variance (ANOVA) Because these tests

indicated almost normally distributed values for the studied ROTEM variables data were

analysed by a fourway ANOVA stratified by the activating reagent (ex-tem in-tem

kaolin or without activation) to test the influence of size age sex and neuter status on

several parameters and taking all interactions into account Post hoc t-tests were

performed when indicated The significance level was set to P lt 05

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

20

3 Results

31 Precision of the method

For all the parameters evaluated in the study the CVs for the fourfold measurement in

healthy dogs were less than 10 (Table 1) In the case of dogs with haemostatic

abnormalities CVs of the vast majority of ROTEM variables were also less than 10

except the variables CFT and ML irrespective whether ex-tem or in-tem reagents were

used (Table 2)

Table 1 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in normal blood samples

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 875 629 262 213 179 634 000 020 037 534

In-temreg 245 596 166 117 135 299 000 000 046 216

Abbreviations CT ndash clotting time CFT ndash clot formation time α ndash alpha angle A30 ndash

amplitude after 30 min MCF ndash maximum clot firmness MCE ndash maximum clot elasticity

LI 30 LI 45 LI 60 ndash Lysis index at 30 45 and 60 min ML ndash maximum lysis

Median values of the measurements of 3 samples

Table 2 Coefficients of variation () for rotational thromboelastometry using the

ROTEMreg delta analyser in canine blood samples with haemostatic alterations

Activator CT (s) CFT (s) α (deg) A30 (mm) MCF (mm) MCE(mm) LI30 () LI45 () LI60 () ML ()

Ex-temreg 668 1478 393 218 268 582 0 018 000 848

In-temreg 228 799 461 219 251 371 0 0 017 283

Median values of the measurements of 3 samples from patients with DIC (n = 2)

thrombocytopenia (n = 1) abbreviations please see legend to Table 1

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

21

32 Reference intervals

Activation by ex-tem reagent (tissue factor) resulted in faster CTs when compared to

activation via the intrinsic pathway using ellagic acid (in-tem) or kaolin reagents whereas

the CFT was much less affected by the activating reagent (Table 3) Especially CFT and

MCE showed a wide inter-individual variation in comparison for example with α angle

and MCF regardless of the used activating reagent (Table 3)

33 Influence of age sex neuter status and size

ANOVA did not reveal statistically significant differences in ROTEM delta parameters

between dog groups of variable size or sex (P gt 05) Age had influence only on CT and

only when the in-tem reagent was used (P = 0169 ANOVA) but not when using the ex-

tem reagent (P = 7105) Post hoc comparisons using t-test of the in-tem results between

age groups revealed shorter CTs in the group of senior dogs when compared with young

dogs (P = 0194) but not between adult dogs and these two groups ANOVA showed

particularly influences of the neuter status and a significant interrelationship between the

factors sex and neuter status on different ROTEM delta parameters using ex-tem and in-

tem respectively as activator (Tables 4 and 5) ANOVA revealed also a significant

influence of neuter status on different ROTEM parameters when kaolin was used as

activating reagent (CFT α angle MCF MCE) whereas non-activated measurements

were not subject to any significant influence of the assessed factors (results not shown)

The interrelationship between sex and neuter status mentioned above reflected the fact

that significant differences between neutered and not neutered dogs were only found in

females The statistical analysis using the t-test to compare intact vs neutered dogs

stratified by sex showed statistically significantly (P lt 05) longer CTs and CFTs and

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

22

lower α angle A30 MCF and MCE values in spayed when compared to intact females

(Tables 4 and 5)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

23

Table 3 Reference values (median reference interval based on 25 - and 975 - quantiles) for thromboelastometric measurements with the ROTEMreg delta analyser in

dogs (n=125 95 confidence intervals in brackets) Abbreviations please see legend to Table 1

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Ex-temreg

Median 450

[430ndash470]

110

[105ndash115]

680

[670ndash690]

600

[580ndash610]

610

[600ndash620]

154

[147ndash164]

100

[960ndash100]

100

[100ndash100]

970

[960ndash980]

220

[210ndash240]

25 340

[270ndash350]

650

[490ndash710]

540

[420ndash560]

460

[410ndash480]

460

[420-490]

870

[720ndash950]

100

[100-100]

870

[800ndash960]

700

[640ndash820]

800

[100ndash110]

975 830

[790ndash990]

220

[201ndash315]

790

[770ndash800]

770

[690ndash770]

730

[700ndash770]

264

[230-334]

100

[100ndash100]

100

[100ndash100]

100

[990ndash100]

590

[530ndash680]

In-tem

Median 190

[187ndash201]

116

[111ndash125]

680

[670ndash690]

570

[560ndash580]

590

[580ndash600]

143

[135ndash149]

100

[100ndash100]

100

[100ndash100]

980

[980ndash980]

180

[170ndash190]

25 135

[112ndash142]

730

[590ndash800]

510

[460ndash570]

450

[410ndash470]

480

[410ndash500]

920

[710ndash101]

100

[950ndash100]

990

[820ndash990]

930

[740ndash930]

500

[100ndash700]

975 283

[268ndash291]

251

[198ndash227]

750

[750ndash780]

660

[660ndash690]

680

[660ndash690]

210

[195ndash227]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

270

[260ndash380]

Activator

Statistical

para-meter

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Kaolin

Median 162

[154ndash173]

870

[820ndash910]

740

[720ndash750]

500

[470ndash530]

550

[530ndash560]

121

[115ndash127]

960

[940ndash980]

840

[790ndash880]

710

[640ndash770]

500

[450ndash540]

25

109

[990ndash115]

470

[440ndash580]

560

[490ndash600]

110

[700ndash230]

390

[380ndash400]

640

[600ndash680]

270

[140ndash480]

110

[800ndash260]

800

[400ndash170]

200

[160ndash240]

975 246

[230ndash337]

214

[185ndash290]

810

[790ndash810]

640

[630ndash690]

650

[650ndash690]

187

[182ndash223]

100

[100ndash100]

990

[980ndash100]

940

[930ndash990]

970

[930ndash100]

Without activation

Median 607

[495ndash662]

221

[194ndash265]

510

[460ndash550]

520

[500ndash530]

540

[520ndash550]

115

[109ndash123]

100

[100ndash100]

100

[100ndash100]

980

[980ndash990]

170

[160ndash190]

25 203

[123ndash215]

860

[550ndash930]

190

[120ndash230]

350

[230ndash370]

370

[260ndash400]

600

[350ndash670]

100

[950ndash100]

980

[890ndash990]

930

[840ndash950]

400

[000ndash500]

975 1356

[1305ndash1721]

811

[688ndash1333]

730

[710ndash790]

660

[640ndash720]

660

[640ndash730]

197

[179ndash265]

100

[100ndash100]

100

[100ndash100]

100

[100ndash100]

290

[280ndash320]

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

24

Table 4 Median values (minimum and maximum in brackets) for different ROTEMreg variables using ex-tem reagent stratified by sex and neuter status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 430

(320ndash530)

960

(590ndash198)

710

(570ndash800)

620

(480ndash730)

640

(510ndash740)

179

(103ndash282)

100

(100ndash100)

100

(980ndash100)

960

(800ndash100)

230

(100ndash640)

Female spayed 460

(270ndash830)

121

(650ndash224)

660

(520ndash770)

580

(430ndash690)

590

(440ndash700)

146

(790ndash230)

100

(960ndash100)

100

(800ndash100)

970

(690ndash100)

240

(100ndash590)

p-value (t-test)

Female

intact vs spayed

00025 00063 00049 00023 00016 00015 na na na na

Male intact 450

(320ndash990)

105

(650ndash315)

690

(420ndash790)

590

(420ndash700)

610

(440ndash710)

154

(770ndash244)

100

(100ndash100)

100

(870ndash100)

970

(650ndash100)

220

(700ndash680)

Male castrated 460

(340ndash930)

111

(490ndash255)

685

(480ndash800)

590

(410ndash770)

600

(420ndash770)

152

(720ndash334)

100

(100ndash100)

100

(830ndash100)

980

(640ndash100)

205

(500ndash590)

p-value (t-test)

Male

Intact vs castrated

08326 05106 04756 01799 01568 02163 na na na na

p-values (ANOVA)

Factor Sex

05428 04847 06612 02206 01887 01630 08489 07893 07689 02758

Factor neuter status 01872 00167 00173 00039 00037 00028 07141 04737 09356 05948

Interrelationship

sexneuter status 00383 00968 00782 00488 00472 00319 04943 08611 05422 00160

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using the ROTEMreg delta device in dogs

25

Table 5 Median values (minimum and maximum in brackets) for different ROTEMreg variables using the in-tem reagent stratified by sex and neutering status

Group

CT

(s)

CFT

(s)

Α

(deg)

A30

(mm)

MCF

(mm)

MCE

(mm)

LI30

()

LI45

()

LI60

()

ML

()

Female intact 186

(126ndash238)

940

(590ndash164)

720

(620ndash780)

610

(490ndash690)

620

(530ndash690)

166

(113ndash227)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

170

(100ndash300)

Female spayed 201

(149ndash291)

1220

(820ndash262)

670

(500ndash740)

550

(450ndash660)

570

(480ndash680)

133

(910ndash214)

100

(950ndash100)

100

(820ndash100)

980

(740ndash100)

190

(500ndash380)

p-value (t-test)

Female

intact vs spayed

na na lt0001 00009 00005 00005 na na na na

Male intact 199

(117ndash283)

122

(730ndash277)

680

(460ndash750)

570

(430ndash660)

590

(450ndash660)

146

(810ndash192)

100

(100ndash100)

100

(990ndash100)

980

(930ndash100)

180

(600ndash270)

Male castrated 181

(112ndash288)

123

(820ndash222)

670

(540ndash740)

565

(410ndash630)

585

(410ndash640)

138

(710ndash180)

100

(100ndash100)

100

(970ndash100)

980

(890ndash100)

180

(800ndash260)

p-value (t-test)

Male

intact vs castrated

na na 08232 01401 01162 00849 na na na na

p-values (ANOVA)

Factor Sex

02105 03317 03370 00877 01258 00510 09564 07832 08117 03638

Factor neuter status 04289 00660 00619 00107 00088 00054 07152 09795 07056 08299

Interrelationship

sexneuter status

04310 00550 00408 00707 01733 00996 05240 05982 08047 04578

na ndash not analysed all significant P-values (lt 005) are highlighted in bold print

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

26

4 Discussion

41 Coefficients of variation

In accordance with the literature CVs below 10 indicate adequate repeatability in intra-

assays evaluations (Steiner et al 2003 Hua et al 2018) According to our results

obtained from the fourfold measurements in citrated whole blood ROTEM delta

measurements on healthy dogs and dogs with haemostatic alterations fulfil these

requirements with only a few exceptions concerning CFT and ML values in abnormal

samples

Our results regarding repeatability are similar to those obtained for several TEM

parameters in previous studies performed in animals In one study the imprecision of the

ROTEM analyser was evaluated by duplicate measurements in 10 blood samples of

healthy dogs and as result of that evaluation a good precision (CVs below 10 ) was

calculated for CT CFT MCF and α angle but in contrary to our results a low

repeatability was found for ML (CVs of 233 [ex-tem] and 285 [in-tem] own values

534 and 216 ) (Falco et al 2012)

In a previous feline study using ex-tem and in-tem as activating reagents for the ROTEM

delta repeatability was assessed similarly to the present study based on fourfold

measurements of blood samples of two healthy individuals (Doumlderlein and Mischke

2015) The results also revealed CVs below 10 for most of the parameters evaluated

The partially low repeatability of ML values may reflect the vulnerability of this variable

which is not only influenced by the complex clot formation process but also by its

subsequent lysis Reproducibility of ex-tem or in-tem activated measurements of the

ROTEM delta analyser which was not investigated in our study was studied in blood

from healthy and unhealthy horses by two different operators using three different

ROTEM devices Results indicated acceptable reproducibility (Junge et al 2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

27

42 Reference intervals

The number of 125 healthy dogs used in the present study to create reference intervals is

regarded adequate according to the guidelines suggested by the American Society of

Veterinary Clinical Pathology (ASVCP) and the National Committee for Clinical

Laboratory Standards (NCCLS) who recommend to include at least 120 individuals

(Friedrichs et al 2012) To the best of our knowledge reference intervals for the most

relevant ROTEMreg delta parameters obtained in an adequate number of samples of whole

citrated canine blood have not been published so far A previous study was performed on

a smaller number of 45 dogs (Falco et al 2012) The median values for reference

intervals obtained in the present study are partially similar to those calculated in the

mentioned study based on 45 healthy dogs (eg ex- tem CT 610 s450 s [Falco et al

2012own results] CFT 150 s 110 s α angle 620deg680deg MCF 530 m 610 mm ML

180 220 in-tem CT 235 s190 s CFT 136 s116 s α angle 660deg680deg MCF

560 mm590 mm ML 30 180 ) However in comparison to our results the

reference intervals in the mentioned study are slightly wider (eg ex-tem CT 29ndash92

s34ndash83 s [Falco et al 2012own results] CFT 540ndash275 s650ndash220 s α angle 470ndash

790deg540ndash790deg MCF 360ndash730 mm460ndash730 mm ML 000ndash560 800ndash590

in-tem CT 126ndash363 s135ndash283 s CFT 470ndash224 s730ndash251 s aα angle 550ndash

810deg510ndash750deg MCF 500-ndash750 mm480ndash680 mm ML 000ndash400 500ndash270 )

although the previous study used the 5 and 95 quantiles to define limits of the

reference range whereas we calculated the commonly used 25 and 975 quantiles

to capture the central 95 of reference values (Friedrichs et al 2012)

Although it is desirable that each laboratory should specifically determine its own

reference intervals the results obtained in this study may provide an orientation for other

laboratories to interpret ROTEM delta parameters analysed in citrated canine blood In

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

28

order to test transference of the reference ranges to another laboratory a validation

procedure with measurement of 20 samples should be performed (Friedrichs et al 2012)

Although intra-individual control values (subject-based reference values) are regarded

superior to population based reference values (Wiinber et al 2007) the latter have to be

routinely used in clinical patients due to lacking individual controls especially for specific

tests such as ROTEM analyses In addition the calculated reference values cannot be

used for the fully automated device ROTEM sigma because - according to an actual

human study - values of part of the parameters do not correlate well between both

instruments (Gillissen et al 2019)

43 Influence of size age sex and neuter status

In previous studies in humans (Attard et al 2013) and animals (Mischke 1994 Borrelli

et al 2017) age has been described as an influencing factor for different haemostasis

parameters We also found an influence of age on the CT with shorter times in senior

dogs (gt 84 months) compared with young dogs (up to 24 months) which was limited to

measurements activated by the in-tem reagent

This age dependency confirms in principle a previous study on 20 healthy dogs

(Barthelemy et al 2015) which found shorter CTs in elderly dogs and results of a human

study where individuals with an age gt 45 years showed shorter CTs compared with

individuals lt 45 years old (Sucher et al 2011) However in the cited studies this effect

was only seen when ex-tem but not when the in-tem reagent was used

Based on the age distribution of our dogs which were primarily recruited to establish

reference intervals we cannot deliver information regarding very young dogs A human

study focusing on ROTEM reference ranges in children demonstrated significant

differences for nearly all ROTEM parameters with the most striking differences in

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

29

subjects aged 0-3 months (Oswald et al 2010) The limited number of dogs with an age

lt 1 year which did not allow to consider these separately in the present study may also

be one reason for the different results in a feline study (Doumlderlein and Mischke 2015) In

the cited study on healthy cats age showed an influence on CT but contrary to our study

the CT was shorter in young individuals (6-12 months) compared with older ones In

addition this was only the case in measurements using kaolin and in non-activated

measurements When compared to humans and cats uni- or multivariate analyses

considering factors age and breed or size based on do populations including different

breeds have the limitation that they can be biased by the variate life expectancy between

small medium and large dog breeds (Willems et al 2017) Apart from age we found a

significant influence only for the neuter status in bitches reflected by a significant

interaction effect between factors neuter status and sex in the ANOVA analysis

Especially for CFT values (eg activated by ex-tem median values of 121 s [spayed] vs

96 s [intact]) this difference seems to be also clinically relevant with respect to the

interpretation of ROTEM results In contrast although also significantly different

absolute and relative differences of CT values were small (eg activated by ex-tem 46 s

[spayed] vs 43 s [intact]) These differences may not only be considered for the result

interpretation of clinical patients but indicate also the necessity for sex-matched patient

groups in scientific studies

The obviously more active haemostasis system in intact females is probably related to an

influence of sex hormones (specifically estrogens) on the coagulation system

Accordingly ex vivo and in vivo studies demonstrated the activating properties of

endogenous estrogens on the haemostasis system In one ex vivo study addition of

endogenous estrogens (estrone [E1] 17szlig-estradiol [E2] and estriol [E3]) to blood samples

from men resulted in shorter mean values of the reaction time in TEG analyses

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

30

corresponding to CT in ROTEM measurements (452-537 s vs 698 s [control]) and of

the TEG variable kinetics (corresponding to the CFT in ROTEM analyses) (Swanepoel

et al 2017) In a further study subcutaneous injection of E2 into male rats induced a

significant increase in activities of coagulation factors X (by 8 ) and XI (by 10-24 )

however did not show any significant effect on PT aPTT and factor VII activity (Franco-

Murillo and Jaimez 2017)

In the cited studies also the effects of the synthetic estrogen ethinylestradiol [EE] were

studied which interestingly seems to have an inhibiting effect as indicated by prolonged

reaction time and kinetics (TEG) compared with the control in the human ex vivo study

and significantly longer PTs and aPTTs and lower factor VII and X activities in the study

on rats As expected we did not find a significant influence of size on ROTEM parameters

in dogs To the best of the authoracutes knowledge size has not been examined as an

influencing factor on viscoelastic measurements or any other haemostatic variable in

dogs Apart from Greyhounds there is limited literature on breed comparisons of

haemostatic variables (Clemons and Meyers 1984 Mischke 1994 Nielsen et al 2007

Nielsen et al 2010 Boyd et al 2018) indicating very limited breed differences for

example for factor VII (Mischke 1994) However these also do not indicate systematic

influence of size and in addition possible breed-specific effects have to be considered

As already stated one limitation of the study is the limited number of young animals

which does not allow to verify reference values for animals in this stage of life In

addition with respect to the main aim of the study possible influencing factors such as

platelet count and haematocrit were not considered in the statistical analysis Finally the

relatively small number of dogs used for the precision analysis which express the fact

that this was only a secondary aspect of the study may be regarded as a limitation At the

same time these considerations provide space for further investigations

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

31

5 Conclusions

Results of the precision analysis revealed acceptable results and thereby confirm that the

ROTEMreg delta analyser delivers repeatable results to measure overall haemostasis in

canine citrated blood samples The established reference values should offer a reliable

basis for the interpretation of thromboelastometric measurements in whole blood using

the device ROTEMreg delta in dogs with suspected haemostasis disorders Neuter status

in female dogs appeared to be a relevant influencing factor on most important ROTEMreg

parameters and should be considered for result interpretation and with respect to

experimental designs

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

32

7 References

1 Attard C van der Straaten T Karlafitis V Monagle P Ignjativic V 2013 Developmental

hemostasis age-specific differences in the levels of hemostatic proteins J Thromb and

Haemost 11 1850ndash1854

2 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and

ROTEMreg A prospective experimental pilot study in healthy volunteers BMC Complemen

Altern Med 17 (16) 64

3 Bartheacutelemy A Rannou B Forterre M Verwaerde P Bonnet-Garin JM Pouzot-Nevoret

C Goy-Thollot I 2015 Differences between coagulation and cytokine profiles in dogs of

different ages Vet J 205 410ndash412

4 Bauer N Eralp O Moritz A 2009 Establishment of reference intervals for kaolin-activated

thromboelastography in dogs including an assessment of the effects of sex and anticoagulat use

J Vet Diagn Investi 21 641ndash648

5 Bollinger D Seeberger MD Tanaka KA 2012 Principles and practice of

thrromboelastography in clinical coagulation management and transfusio practice Transfus Med

Rev 26 1ndash13

6 Borrelli A Botto A Maurella C Falco S Pagani E Miniscaldo B Tarducci A Bruno

B 2017 Thromboelastometric assesment of hemostasis in newborn Piemontese calves J Vet

Diagn Invest 29 293ndash297

7 Boyd CJ Claus MA Raisis AL Hosqood G Sharp CR SmartL 2018

Hypocoagulability and platelet dysfunction are exacerbated by synthetic colloids in a canine

hemorrhagic shock model Front Vet Sci 5 279

8 Bruno B Maurella C Falco S Tarducci A Zanatta R DrsquoAngelo A Borrelli A 2015a

Thromboelastometric evaluation of hemostasis in dogs infected with Leishmania infantum J Vet

Emerg Crit Care (San Antonio) 25 502ndash511

9 Bruno B Maurella C Falco S Tarducci A Zanatta R Gianella P DrsquoAngelo A Piras L

Di Bella A Borrelli A 2015b Assessment of coagulation utilizing thromboelastometry in dogs

undergoing orthopedic surgery J Vet Emerg Crit Care (San Antonio) 25 358ndash363

10 Clemons RM Meyers KM 1984 Acquisition and aggregation of canine blood platelets Basic

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

33

mechanisms of function and differences because of breed origin Am J Vet Res 45 137ndash144

11 Conversy B Blais MC Dunn M Gara-Boivin C del Castillo JRE 2017 Anticoagulant

activity of oral rivaroxaban in healthy dogs Vet J 223 5ndash11

12 Crochemore T Piza FMT Rodriacuteguez RDR Guerra JCC Ferraz LJR Correcirca TD

2017 A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

13 Crochemore T Correcirca TD Lance MD Solomon C Neto AS Guerra JCC Lellis PS

Bernz LM Nunes N Mancio CM Yokoyama APH Silva E 2018 Thromboelastometry

profile in critically ill patients  A single-center retrospective observational study PLoS One 13

e0192965

14 Doumlderlein E Mischke R 2015 Reference intervals for thromboelastometry with the ROTEMreg

delta in cats Res Vet Sci 100 271ndash276

15 Falco S Bruno B Maurella C Bellino C DacuteAngello A Gianella P Tarducci A Zannata

R Borrelli A 2012 In vitro evaluation of canine hemostasis following dilution with

hydroxyethyl starch (13004) via thromboelastometry J Vet Emerg Crit Care (San Antonio)

22 640ndash645

16 Franco-Murillo Y Jaimez R 2017 Inhibitory effect of ethinylestradiol on coagulation factors

in rats Exp Anim 66 107ndash113

17 Friedrichs KR Harr KE Freeman KP Szladovits B Walton RM Barnhart KF Blanco-

Chavez J 2012 ASVCP reference interval guidelines Determination of de novo reference

intervals in veterinary species and other related topics Vet Clin Pathol 41 441ndash453

18 Gillissen A va den Akker T Caram-Deelder C Henriquez DDCA Bloemenkamp

RWM Eikenboom J van der Bom JG de Maat MPM 2019 Comparison of

thromboelastometry by ROTEMreg Delta and ROTEMreg Sigma in women with postpartum

haemorrhage Scand J Clin Lab Invest 79 32ndash38

19 Harttert H 1948 Blutgerinnungstudien mit der Thrombelastographie einem neuen

Untersuchungsverfahren Klin Wochenschr 26 577ndash583

20 Horn P Pesce A 2003 Reference intervals and update Clin Chim Acta 334 5ndash23

21 Hua Yanhun Qiu W Xiao Q Wu Q 2018 Precision (repeatability and reproducibility) of

ocular parameters obtained by the Tomey OA-2000 biometer compared to the IOLMaster in

healthy eyes PLoS One 13 e0193023

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

34

22 Junge H Ringer S Mayer N Schwarzwald C 2016 Assesment of method reliability and

determination of reference intervals for rotational thromboelastometry in horses J Vet Emerg

Crit Care (San Antonio) 26 691ndash703

23 Kol A Borjesson DL 2010 Application of thrombelastographythromboelastometry to

veterinary medicine Vet Clin Pathol 39 405ndash416

24 Marschner CB Wiinberg B Tarnow I Markussen B Kuumlhnel L Bochsen L Kristensen

AT 2018 The influence of inflammation and hematocrit on clot strength in canine

thromboelastographic hypercoagulability J Vet Emerg Crit Care (San Antonio) 28 20ndash30

25 McMichael MA Smith SA 2011 Viscoelastic coagulation testing  technology applications

and limitations Vet Clin Pathol 40 140ndash153

26 Mischke R 1994 Activity of coagulation factors II V VII and X in healthy dogs- dependence

on age sex and breed Berl Muumlnch Tieraumlrztl Wochenschr 107 289ndash294

27 Nielsen A Zois N Pedersen H Olse L Tarnow I 2007 Platelet function in dogs bread

differences and effect of acetylsalicylic acid administration Vet Clin Pathol 36 267ndash273

28 Nielsen L Kjelgaard‐Hansen M Jensen A Kristensen A 2010 Breed‐specific variation of

hematologic and biochemical analytes in healthy adult Bernese Mountain dogs Vet Clin Pathol

39 20ndash28

29 Nogami K 2016 The utility of thromboelastography in inherited and acquired bleeding

disorders Br J Haematol 174 503ndash514

30 Oswald E Stalzer B Heitz E Weiss M Schmugge M Strasak A Innerhofer P Haas T

2010 Thromboelastometry (ROTEMreg) in children Age-related reference ranges and correlations

with standard coagulation tests Br J Anaesth 105 827ndash835

31 Sankarankutty A Nascimento B Teodoro da Luz L Rizoli S 2012 TEGreg and ROTEMreg

in trauma similar test but different results World J Emerg Surg 7 (Suppl 1) S3

32 Sigrist NE Schefer RJJ Kutter APN 2018 Characteristics of hyperfibrinolysis in dogs and

cats demonstrated by rotational Thromboelastometry (ROTEM) Vet J 242 67-73

33 Steiner J Teague S Williams D 2003 Development and analytic validation of an enzime-

linked immunosorbent assay for the measurement of canine pancreatic lipase immunoreactivity in

serum Can J Vet Res 68 161ndash168

34 Sucker C Tharra K Lithmate J Scharf RE Zotz RB 2011 Rotation thromboelastography

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

35

(ROTEM) parameters are influenced by age gender and oral contraception Perfusion 26 334ndash

340

35 Swanepoel A Emmerson O Pretorius E 2017 The effect of endogenous and synthetic

estrogens on whole blood clot formation and erythrocyte structure Microsc Microanal 23 599ndash

606

36 Theusinger OM Nuumlrnberg J Asmis LM Seifert B Spahn DR 2010 Rotation

thromboelastometry (ROTEM) stability and reproducibility over time Eur J Cardiothorac Surg

37 677ndash683

37 Toulon P Metge S Hangard M Zwahlen S Piaulenne S Besson V 2017 Impact of

different storage times at room temperature of unspun citrated blood samples on routine

coagulation tests results Results of a bicenter study and review of the literature Int J Lab

Hematol 3 458ndash468

38 Turner JS Kutter APN Sigrist NE 2019 Correlation of rotational thrromboelastometry

(ROTEM) parameters with platelet count and their ability to predict thrombocytopenia in dogs

Res Vet Sci 126 45ndash50

39 Versteeg HH Heemskerk JWM Levi M Reitsma PH 2013 New Fundamentals in

Hemostasis Physiol Rev 93 327ndash358

40 Wiinberg B Jensen AL Kjelgaard-Hansen M Rojkjaer R Johansson PI Gade LP

Gram DX Kristensen AT 2007 Study on biological variation of haemostatic parameters in

clinically healthy dogs Vet J 174 62ndash68

41 Wiinberg B Jensen AL Johansson PL Rozanski E Tranholm M Kristensen AT 2008

Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular

coagulation J Vet Intern Med 22 357ndash365

42 Wikkelsoslash A Wetterslev J Moslashller AM Afshari A 2016 Thromboelastography (TEG) or

thromboelastometry (ROTEMreg) to monitor haemostatic treatment versus usual care in adults or

children with bleeding Cochrane Database Syst Rev Art

httpsdoiorg10100214651858CD007871pub3 No CD007871

43 Willems A Paepe D Marynissen S Smets P Van de Maele I Picavet P Duchateau L

Daminet S 2017 Results of screening of apparently healthy senior and geriatric dogs J Vet

Intern Med 31 81ndash92

Chapter 2 Reference intervals for rotational thromboelastometry measurements using

the ROTEMreg delta device in dogs

36

44 Woodman J Wagg CR Boysen SR Leguillete R Mizen K Roy MF 2015 Evaluation

of coagulation via thromboelastography in healthy horses administered dexamethasone Can Vet

J 56 1271ndash1274

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of

standard heparin prophylaxis and correlation to haemostatic changes

Pereira JM Hewicker-Trautwein M Rohn K von Depka Prondzinski M

Mischke R

Contributions of the authors

Pereira JM performed (1) the ultrasound evaluations and blood sampling of the

patients under study (2) haematology analysis (3) injection of the anticoagulation

prophylaxis (partially) to the patients under study (4) analysis of samples (collaboration)

at Werlhof Institute Medical Care Center (5) analysis of some haemostasis test

(collaboration) (6) statistical analysis (partially) (7) data documentation (8) and wrote

the manuscript

Hewicker-Trautwein M performed the electron microscopy analysis

Rohn K performed the statistical analysis

von Depka Prondzinski M performed some haemostasis test (thrombin generation

assay and endogenous thrombin potential)

Mischke R performed the coagulation and haemostasis test was involved in designing

and supervision of the study analysis and discussion of the data and its statistical analysis

and revision edition and correction of the manuscript

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

38

Central venous catheter induced thrombosis in dogs ndash efficacy of standard heparin

prophylaxis and correlation to haemostatic changes

Joseacute Mauricio Pereira1 Marion Hewicker-Trautwein2 Karl Rohn3 Mario von Depka

Prondzinski4 Reinhard Mischke1

1Small Animal Clinic University of Veterinary Medicine Hannover Foundation

Buumlnteweg 9 D-30559 Hannover Germany

2Institute for Pathology University of Veterinary Medicine Hannover Foundation

Buumlnteweg 17 D-30559 Hannover Germany

3Institute of Biometry Epidemiology and Information Processing University of

Veterinary Medicine Hannover Foundation Buumlnteweg 2 30559 Hannover Germany

4Werlhof Institute Schillerstrasse 23 30159 Hannover Germany

Abbreviations anti-Xa anti-factor Xa BW body weight CVC central venous catheter

EJV external jugular vein ELMI electron microscopy IV intravenous JV jugular vein

LMWH low molecular weight heparin ML maximum lysis (ROTEM) PIVC

peripheral intravenous catheter ROTEM rotational elastometry SC subcutaneous

TGA thrombin generation assay TID three times a day TT thrombin time UFH

unfractionated heparin

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

39

Corresponding author

E-mail addresses matiscrgmailcom (JM Pereira) MarionHewicker-

Trautweintiho-hannoverde (M Hewicker-Trautwein) KarlRohntiho-hannoverde

(K Rohn) mailwerlhof-institutde (M von Depka Prondzinski)

ReinhardMischketiho-hannoverde (R Mischke)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

40

ABSTRACT

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens The objectives of the present study were (1) to determine the incidence of

catheter-induced thrombosis in canine intensive care patients and thereby (2) to evaluate

the efficacy of a routine prophylactic heparin treatment and finally (3) to assess whether

initial changes of selected haemostasis parameters are predictive for an increased risk for

thrombosis formation 29 dogs hospitalised in the Small Animal Clinic University of

Veterinary Medicine Hannover Foundation and receiving a central venous catheter

(CVC) in the external jugular vein (EJV) for medical reasons were included in the final

assessment 17 dogs received the standard antithrombotic dosage regimen of

unfractionated heparin (UFH 150 IUKG BW TID sc) and 12 dogs (usually surgical

patients) the reduced dosage (75 IUKG BW TID sc) A colour Doppler ultrasound of

the respective EJV in order to detect thrombus formation and blood collection for

haemostasis tests were performed immediately before the installation procedure (day 0)

and on days 1 3 5 7 and so on after the CVC insertion Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time antithrombin

activity D-dimer concentration thrombin generation and rotational elastometry as well

as the heparin activity using a chromogenic anti-factor Xa test Finally electron

microscopy (ELMI) of sections of the removed CVC was performed In 8 dogs

sonographic examination revealed thrombus formation in the lumen of EJV and in 1927

CVCs assessed using ELMI thrombi on the external CVC surface were detected without

significant correlation between both methods Plasmatic heparin activities showed great

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

41

variability but no significant differences were found between dogs with or without CVC-

associated thrombus formation From the initially performed haemostasis tests high

fibrinogen concentrations and low maximum lysis values as measured by the ROTEM

delta device (ex-tem reagent) were associated with sonographically detectable thrombi

The results of the present study indicate that CVC-associated thromboses occur (eg 28

[829] of the patients as detected by ultrasound) although severe thromboses (high

grades) are relatively rare in intensive care canine patients receiving a routine

anticoagulant treatment A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment

Keywords CVC complication venal occlusion antithrombotic treatment ultrasound

electron microscopy rotational elastometry

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

42

1 Introduction

Central venous catheters (CVCs) are very important tools for care of patients in critical

care units allowing parenteral nutrition support therapy administration of high volumes

of intravenous fluids safe drugs administration that cannot be given through peripheral

veins plasmapheresis haemodialysis blood transfusions as well as periodical blood

sampling (Geerts 2014 Vose et al 2019) CVC related thrombosis is one of the most

severe complications (apart from infections and pneumothorax) of CVCs in humans

(Geerts 2014 Comerlato et al 2017 Evans and Ratchford 2018) leading to patientrsquos

distress CVC dysfunction venous stenosis and in some cases to the development of

thromboembolism (Geerts 2014)

In the human literature there are various studies dealing with epidemiology with reference

to different anticoagulatory regimens including placebo-controlled studies (Lee and

Kamphuisen 2012) In humans CVC related thromboses represent 70ndash80 of all upper

extremity thrombosis cases and approximately 10 of thromboembolism cases in

humans (Kucher 2011) The incidence of CVC related thrombosis reported in humans

based on 25 studies ranged widely from 2 to 67 (median = 30 ) (van Rooden et al

2005 review article) This wide range of calculated incidences of CVC related

thromboses in humans reflects the use of different CVC types different locations of entry

site of the CVC (jugular femoral subclavian axillary cephalic vein) different thrombus

detection technique (venography or ultrasonography) unequal durations of follow up

varying diseases of patients (eg haematological oncological) and presence of

predisposing factors (eg previous venous thrombosis history presence of cancer) (van

Rooden et al 2005 Geerts 2014 Liu et al 2015)

Part of the authors empirically observed CVC-associated thrombosis of the external

jugular vein (EJV) as a relevant complication in their canine intensive care patients which

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

43

awakened their interest in that topic To the best of the authoracutes knowledge studies about

extra luminal CVC related thromboses in dogs are rare in the available veterinary

literature One study evaluating the complications associated with 100 CVCs placed in

the jugular vein of dogs and cats reported only 2 cases of thrombosis (confirmed by

palpation or ultrasound) but obviously did not assess the appearance of subclinical CVC

related thromboses (Adamantos et al 2010) Thereby the incidence of CVC-induced

thromboses in dogs is unknown and systematic studies regarding an effective

anticoagulatory management are lacking

In this context it is of interest whether a possible development of thrombosis or an

increased risk (despite the prevention by a standard application of unfractionated heparin

[UFH]) can be predicted by means of changes of haemostasis related parameters during

the initial examination In one study on 168 human patients without indication of

anticoagulation prophylaxis for example elevated leukocyte counts high plasminogen

activator inhibitor and high coagulation factor VIII levels (all measured on blood samples

taken after CVC placement) were associated with an increased incidence of symptomatic

CVC-related thrombosis diagnosed by ultrasonography (Boersma et al 2016)

Therefore aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (as detected by ultrasonography and

electron microscopy ELMI) and thereby (2) to evaluate the efficacy of a standard

antithrombotic regimen with unfractionated heparin (UFH) routinely used in the Small

Animal Clinic of the University of Veterinary Medicine Hannover and (3) to assess

whether selected haemostasis test results are predictive for an increased risk for

thrombosis formation

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

44

2 Material and methods

21 Study design

Canine in-patients of the Small Animal Clinic University of Veterinary Medicine

Hannover receiving a CVC in the EJV for medical reasons from March 2017 to

December 2019 were included Anticoagulatory treatment was performed according to

the standard protocol of the clinic Colour Doppler ultrasonographic controls of the EJV

and blood samplings (for heparin activity basic haematology and haemostasis tests) were

done immediately before the catheter was placed (day 0) and on days 1 3 5 etc (2 hours

after the first daily UFH injection) until the CVC was removed Only patients were

included which could be examined at least until day 3 Haemostasis tests included

prothrombin time activated partial thromboplastin time thrombin time fibrinogen

concentration antithrombin activity fibrin monomers D-dimers thrombin generation

assay (TGA) and rotational elastometry as well as the heparin activity using a

chromogenic anti-factor Xa test Finally ELMI of sections of the CVC after its removal

was performed

22 Animals

Originally 32 patients hospitalised in the Small Animal Clinic and receiving a CVC in the

EJV for medical reasons entered the study Three of them were subsequently excluded

because they did not reach the defined minimum study period of 3 days Two were

euthanised and one referred back to the referring vet before day 3 22 dogs weighing more

than 15 kg were included so that the withdrawal of about 10 ml of blood on several days

was not a significant burden for the patient (lt 1 of the total blood volume) Five dogs

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

45

with a body weight of less than 15 kg but at least 8 kg were included with special

consideration of the amount of blood taken (no more than 6 ml on individual days)

All animals received standard anticoagulatory treatment with UFH 17 dogs with internal

diseases received 150 IUkg BW TID subcutaneously and 12 dogs (post surgery) received

a reduced dosage of 75 IUkg BW TID subcutaneously

The 29 dogs enclosed in the study included 4 mixed breed dogs and the following

pedigree dogs 5 Labrador Retrievers 3 German Shepherds 2 Australian Shepherds and

one from each of the following breeds Akita Inu Asian Shepherd Bavarian Mountain

Hound Beagle Belgian Shepherd Briard Canaan dog Dalmatian Golden Retriever

Havanese dog Hovawart Rhodesian Ridgeback Schnauzer Shar Pei and Wolfspitz

These dogs were aged between 2 and 164 months (median 77 months) and had body

weights ranging from 81 and 63 kg (median 27 kg) 9 were castrated males 8 intact

males 7 spayed females and 5 intact females

The diseases of the 29 patients requiring a CVC included 3 from each of the following

diseases hepatopathy and esophagus dilatation 2 from each of the following diseases

bacterial cystitis diabetes mellitus and 1 from each of the following diseases abdominal

wall perforation aspiration pneumonia cystitis intestinal round cells sarcoma carcinoma

of the nasal cavity chronic renal insufficiency cystic duct rupture heartworm disease

ivermectine intoxication lymphoma multifocal alveolar histiocytosis pancreatitis post-

surgical bleeding stomach dilatation stomach foreign body tetanus thoracic

pyogranulomatous inflammation thrombocytopenia and trigeminal nerve paresis

Animals were treated in accordance with the German Animal Welfare Law The

experimental design was approved by the Animal Welfare officer of the University of

Veterinary Medicine and reported to the responsible authority (Lower Saxony State

Office for Consumer Protection and Food Safety reference number 17A101)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

46

23 Positioning and handling of the central venous catheter

Two types of single lumen radiopaque polyurethane CVC (different on length and

diameter) were used during the study The CVC type I (n = 2529) was Cavafixreg Certoreg

Splittocanreg CVC 32 cm length 11 x 17 mm diameter16 G (BBraun Melsungen AG)

the CVC type II used (n = 429) was Certofixreg Mono S 215 15 cm length 08 x 14 mm

diameter18G (BBraun Melsungen AG) Each CVC was inserted into the EJV using a

modified Seldinger technique as follows the skin of the neck was shaved and disinfected

with antiseptic soap alcohol and iodine Local anesthesia (Lidocaine HCl 2) was

administrated subcutaneously on the catheter insertion site (a middle point in the lateral

neck over the jugular sulcus) A small skin incision was performed in the insertion site

and one peripheral plastic cannula with a metallic needle inside (14 G for CVC type I

and 20 G for CVC type II) was inserted in the EJV in cranio-caudal orientation (Hundley

et al 2018) Once the EJV was reached the needle was removed and a flexible wire

guide was inserted through the peripheral cannula following an external landmark

(4thintercostal space) (Reminga et al 2018) The peripheral cannula was removed and

the respective CVC was inserted slowly and with gentle pressure into the EJV guided

with the flexible wire and following the same external landmark to reach the right

position After the insertion of the CVC the flexible wire guide was removed The correct

positioning of the CVC was checked with a latero-lateral thorax x-ray ensuring that the

CVC tip was placed at the junction of the cranial cava vein with the right atrium Each

CVC was fixed to the neck skin (adjacent to the insertion point) with a monofilament

non-absorbable suture (gauge 3-0) Finally a bandage elaborated with Povidone-Iodine

gauze (Betaisodonareg) padded synthetic band (Artiflexreg Natur) and elastic band

(Elastomullreg) was used to protect the CVC output (Claude et al 2010) and changed

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

47

daily Several times day the catheter was flushed with isotonic sodium chloride

containing 1 IUml UFH

24 Heparin preparation and administration

Heparin-sodium-25000-ratiopharm (injection solution containing 5000 IUml sodium

heparin [mucosa]) was used SC injection was performed over the lateral thorax wall

using 1 mL syringes with fine graduation marks

25 Ultrasonography

Colour Doppler ultrasonographic examinations were performed in lateral position using

a LO-GIQ E9 ultrasound device (GE Healthcare Chicago Illinois USA) with a 15

Megahertz linear ultrasound probe The development of a thrombus or fibrin debris in the

EJV lumen around the CVC or inside it was assessed at each control Once the

ultrasound was finished a new protective bandage was placed around the neck of the dog

According to the percentage of lumen occlusion of the EJV (evaluated in cross sectional

views during the ultrasound scanning) the thrombi were classified in four different grades

(Fig 1)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

48

of the residual vein lumen C ndash catheter T ndash thrombus

Fig 1 Definition of different grades of central venous catheter induced thrombosis of the

jugular vein as detected using colour Doppler ultrasound

26 Blood collection

At each time point approx 6ndash10 ml of blood (according to the body weight) was collected

This included approx 1 ml of EDTA blood for hematological analysis and approx 5ndash9

ml of citrated blood (tubes containing one part of 011 molL [38 ] sodium citrate to

nine parts of blood) for the measurement of hemostasis tests in citrated blood (ROTEM)

or citrated plasma Blood samples were collected by puncture of peripheral veins with

disposable cannulas (09 x 40 mm [20G] or 10 x 40mm [21G]) directly into sample tubes

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

49

or from the CVC using the three-syringe-technique (Villalta-Garciacutea et al 2015) in

situations when the peripheral venipuncture was not indicated due to phlebitis or irritation

andor inflammation of the surrounding skin Briefly 2ndash3 ml of blood were taken in a

sterile disposable syringe and immediately afterwards the blood sample was taken in a

new disposable syringe and transferred into the test tubes containing the respective

anticoagulants finally the first 2-3 ml taken were re-infused to the dog through the CVC

followed by flushing with a solution of sodium chlorideUFH (02) Immediately after

blood collection each sample tube was swayed gently until blood and anticoagulant had

been mixed thoroughly To prepare citrated platelet poor plasma citrated blood was

centrifuged at 16000 g for 10 min at room temperature using a microcentrifuge The

plasma was transferred into plastic tubes and the centrifugation procedure was repeated

This platelet poor plasma was then frozen in aliquots at -70 degC until used for analysis

Directly before analysis the plastic tubes were thawed at 37 degC in a water bath

27 Laboratory tests

Haematological analyses were made automatically with Adviareg 2120i (Siemens

Healthcare GmbH Germany)

Prothrombin time activated partial thromboplastin time and thrombin time were

measured with the autoanalyser Amax Destiny Plus (Tcoag Deutschland GmbH

Germany) and commercial reagents (Thromborel S Siemens Healthcare Diagnostics

Products GmbH Marburg Germany CK Prestreg and Test Thrombin Reagent

Diagnostica Stago SAS Asniegraveres sur Seine Cedex France) Whereas the last two tests

were measured according to the manufacturer`s instructions a modified test optimised

for canine plasma was used to measure prothrombin time (25 microl citrated plasma 120 with

imidazole buffer + 25 microl of a human fibrinogen solution 2 gl prewarmed for 120 s at 37

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

50

oC addition of 25 microl of activating reagent Thromborel S) and the results converted into a

percentage activity based on a calibration curve prepared with canine pooled plasma The

fibrinogen concentration was measured with the method according to Clauss using a test

kit from Stago Diagnostica SAS D-dimer concentration was measured with a

quantitative commercial human latex turbidimetric method in the autoanalyser Cobas

c311 using a completely assembled kit and human standards (Roche Diagnostics

Mannheim Germany) Antithrombin activity was also measured with a completely

assembled chromogenic test kit in the autoanalyser Cobas c311 and calibrated with a

canine pool plasma (n = 100)

The thrombin generation assay (Technothrombinreg TGA (Technoclone GbmH Vienna

Austria) determines thrombin generation based on monitoring of the fluorescence

generated by cleavage of a fluorogenic substrate by thrombin over time after activation

of the coagulation cascade by different concentrations of tissue factor and negatively

charged phospholipid in platelet poor plasma From the fluorescence development over

time the concentration of thrombin in the sample can be calculated using a thrombin

calibration curve The increase in thrombin concentration with time then allows to

calculate generation of thrombin in the sample and to plot thrombin values over time for

the whole coagulation process

The rotational thromboelastometry analysis with ROTEM delta device (Werfen Munich

Germany) was started approx 15 minutes after the blood collection To perform the

analysis and according to the automated ROTEM software 20 microl of recalcification reagent

(200 mmoll calcium chloride solution star-tem) and 20 microl of the respective activating

reagent (ex-tem in-tem) were pipetted into and carefully mixed in a pre-warmed cuvette

using the automated pipette provided in the device immediately afterwards 300 microl of

citrated whole blood (previously gently revolved) were added and mixed with the

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

51

reagents previously deposited Finally the cuvette was carefully placed and fixed to an

oscillating pin (oscillation movements on left and right through an angle of 475deg) in the

ROTEM delta device The interaction between blood cells and subsequently the bond

between fibrin and platelets promoted by the biomechanical changes in pH electrolytes

and temperature start the clot formation with its subsequent retraction and lysis The

magnitude of the resistance produced by the clot over the oscillating pin is quantitatively

represented by different parameters measured during the analysis and also qualitatively

in a graphic illustration (Bagge et al 2016 Crochemore et al 2017)

A chromogenic substrate test (Coatest Heparin Chromogenix-Instrumentation

Laboratory Haemochrom Diagnostica Essen Germany) was used for measurements of

UFH in the auto-analyser Cobas c311 The test application was created based on

manufacturers` instructions Different dilutions of the same batch of the commercial

sodium UFH preparation used in the experiment with normal canine pool plasma

(prepared from identical aliquots of 10 healthy adult dogs) served as standards for the

calibration of the measurement

28 Electron microscopy

Selected areas on the last third of the CVC (5ndash10 cm according with the CVC) with

macroscopically detectable thrombi were analysed and some of the CVCs were cut in

two different places The cut samples were preserved in a fixative medium composed of

1 glutaraldehyde and 4 formalin Once the sample was fixed it was prepared

according to the osmium thiocarbohydrazide procedure (OTOTO-procedure) (Malick et

al 1976) The fixed CVCs were dehydrated in seven series as follow first three series of

120 minutes each one on 30 50 and 70 ethanol respectively followed by two series

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

52

of 120 min each one on 90 ethanol and then two series of 24 hours each one on 100

ethanol After the dehydration process the samples were placed two times during 24

hours each one on isoamyl acetate and then placed on in a critical point drier Finally the

samples were coated with gold in a splutter-coater with 25 kV acceleration voltage in an

argon atmosphere with a current of 20 mA for one minute The samples were examined

in an EVO 15 scanning electron microscope (ZEISS International Carl Zeiss Jena

Germany) and photographed with magnifications of 600x and 6000x avoiding mechanical

artifacts (Toskala et al 1995) Thrombi on the external wall of the CVC were analysed

and measured using the software ImageJ (National Institutes of Health and the Laboratory

for Optical and Computational Instrumentation University of Wisconsin USA)

According to their extension (surface size) and thickness (related to the CVC wall

thickness in a cross sectional view) thrombi on the external wall and additionally within

the lumen of the CVC were each classified in four grades (Fig 2 Fig 1S [suppl

material]) In case of two cuts the highest degree of thrombosis detected at one of the two

sites was used for the final assessment

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

53

surface sizepercent of CVC wall thickness

Fig 2 Definition and illustration of different grades of central venous catheter induced

(extraluminal) thrombosis as detected using electron microscopy

Fig 1S (suppl material) Definition and illustration of different grades of intraluminal

thrombosis in central venous catheters as detected using electron microscopy

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

54

29 Statistical methods

Comparison between thrombus graduation either based on sonographic and ELMI

findings respectively was performed using the McNemar test with calculation of the

concordance index (κ) Standard normal distribution was assessed using the

Kolomogorov-Smirnov test

Due to non-normal distribution of part of data sets heparin activities and results of

haemostasis assays are reported using non-parametric variables (median minimumndash

maximum box and whisker plots) Heparin activities and results of haemostasis tests

between dogs with or without thrombosis were compared using Mann-Whitney-U test

3 Results

31 Sonographic detection of thrombus formation

Thrombus formation was sonographically detected in 829 patients during the study

period Based on examination on the final day in half of these eight patients (n = 4) the

thrombus formation was scored grade 1 in 3 dogs grade 2 and only in one grade 3 In all

of these 8 dogs this was the maximum grade during the study period One dog with fibrin

precipitates was excluded from further analysis because it could not be clearly assigned

to one of the groups In 78 cases thrombi were attached mainly to the venous wall

whereas only in one case was attached to the CVC external wall Seven of these 8 dogs

which developed thrombi received the standard heparin dosage (150 IUkg TID 41

[717] of dogs receiving this dosage) and only one the reduced heparin dosage (75 IUkg

TID 8 [112] of dogs receiving this dosage)

The time point of first detection of thrombus formation varied between days 3 and 11

(median day 8) In addition detritus (presumably composed mainly of fibrin) was

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

55

detectable in the lumen of the EJV in one further dog and within the lumen of the CVC

in further 5 dogs

32 Thrombus formation by electron microscopy

27 from 29 CVC catheter tips were analysed by electron microscopy 827 samples were

without detectable thrombus (grade 0) and 1927 with detectable thrombi 72 showed

thrombus formation of grade 1 427 of grade 2 and 827 of grade 3 There was no

concordance between thrombus graduation based on sonographic and the ELMI

graduation respectively The concordance index (McNemar test) was negative (κ = -

00718) indicating that the degree of agreement is less than by chance

Of the in total 19 dogs with detectable thrombus formation by electron microscopy 11

dogs received 150 IUkg TID (in total 16 dogs of these 27 dogs received that heparin

dosage) and 8 received 75 IUkg TID (in total 11 of these 27 dogs received that heparin

dosage 73 811) and Of the 12 patients with grade 2 and 3 thrombus formation 5 dogs

received 75 IUkg TID and 7 dogs 150 IUkg

The supplementary analysis of intraluminal thrombus formation revealed no detectable

thrombi in 7 cases and thrombus formation of grade 1 (lt 25 occlusion) in 11 cases of

grade 2 (25ndash50 ) in 3 cases and of grade 3 (gt 50 ) in 6 cases

33 Heparin activities and relationship to thrombus formation

Heparin activities varied remarkably in both dosage groups (Fig 3 ab) There was no

significant difference of heparin activities between blood samples collected via

venipuncture or from the CVC (Table 1S supplementary material) In addition there was

no significant difference between heparin activities in dogs with or without

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

56

sonographically or electron microscopically detectable thrombosis (Tables 1 and 2)

Table 1 refers to sonographically detectable thromboses and is based on median heparin

activities calculated from the measurements on day 1 and following days per patient

firstly for the total number of patients and secondly for the patients receiving 150 IUkg

TID In addition in the eight dogs heparin activities on the day of first thrombus detection

(023 IUml 019ndash036 IUml) were not significantly different from the median values

(Table 1 P = 05214 Wilcoxon test) Table 2 compares heparin activities (median values

of all days and results on the final day) between dogs with and without electron

microscopically detectable CVC-associated thrombi

Table 1S (Supplementary material) Heparin activities (IUml median values

minimumndashmaximum) measured in 27 dogs receiving 75 or 150 IUkg unfractionated

(standard) heparin TID subcutaneously 2 hours after the first injection on the day

Comparison of samples collected via puncture of the peripheral veins and those collected

from the central venous catheter (CVC) using the three-syringe-technique

Peripheral vein

puncture CVC

Mann-Whitney-U

test

Total number

(75 and 150

IUkg TID)

019

(00ndash108)

(n = 65)

023

(000ndash105)

(n = 65)

P = 03033

75 IUkg TID

010

(000ndash108)

(n = 25)

012

(000ndash023)

(n = 21)

P = 06907

150 IUkg

TID

030

(000ndash096)

(n = 40)

036

(000ndash105)

(n = 44)

P = 01223

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

57

Fig 3ab Heparin activities (box- and whisker plot arithmetic means) in dogs receiving

unfractionated heparin at a dosage of 75 IUkg BW tid (a) or 150 IUkg BW tid (b) two

hours after the first daily heparin administration

day 1 represents the first day after initiation of heparin treatment

only days were considered with at least results from 5 patients

Table 1 Heparin activities (IUml median values minimumndashmaximum) in 28 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(grades 1ndash3) or without (grade 0) sonographically detectable thrombus Blood collections

were performed 2 hours after the first daily injection

Grade 0

(sonography)

Grades 1ndash3

(sonography)

Mann-Whitney U

test

Total number

(75 and 150 IUkg

TID)

020 (000ndash072)

(n = 20)

027 (018ndash059)

(n = 8)

P = 03212

150 IUkg TID 042 (000ndash072)

(n = 10)

031 (018ndash059)

(n = 7)

P = 04593

one dog with fibrin precipitates in the lumen of the jugular vein but without defined thrombus formation

was not considered because it could not be clearly assigned to one of the two groups

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

58

Table 2 Heparin activities (IUml median values minimumndashmaximum) in 27 dogs

receiving 75 or 150 IUkg unfractionated (standard) heparin TID subcutaneously with

(score 1ndash3) or without (score 0) detectable thrombus via electron microscopy

Grade 0

(electron

microscop)

(n = 8)

Grades 1ndash3

(electron

microscop)

(n = 19)

Mann-Whitney U

test

Median value of

different days 028 (005ndash060)

022 (000ndash072)

P = 05770

Final day 020 (001ndash045)

015 (000ndash105)

P = 06706

34 Relationship between haemostasis tests and thrombus formation

From the initially performed haemostasis tests high fibrinogen concentrations and low

maximum lysis values as measured by the ROTEM delta device (using the ex-tem

reagent) were associated with sonographically detectable thrombi (Tab 3) Additional

analyses based on the median values of different days for individual patients revealed that

low maximum lysis values as measured by the ROTEM delta device (using the in-tem

reagent) were associated with sonographically detectable thrombi (maximum lysis score

0 850 [100ndash175 ] score 1ndash3 375 [100ndash900 ] median [minimumndash

maximum] P = 00166) but no other test Comparison of the results of haemostasis tests

on the final day between dogs with and without sonographically detectable thromboses

did not show any significant difference (P gt 005)

Comparison of results of haemostasis tests on the final day in dogs with detectable

thrombosis via ELMI and dogs without significant thrombus formation showed several

significant differences including lower fibrinogen concentrations and longer clotting

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

59

times (ex-tem reagent) and clot formation times (ex-tem and in-tem reagents) using the

ROTEM delta analyser in general expressing a less coagulable status (Tab 4) If values

of dogs graded 0 and 1 (n =15) are compared to those of grades 2 and 3 (n = 12) the last-

mentioned had longer clotting times (ROTEM delta in-tem reagent) as well as longer clot

formation times and lower alpha angles (ROTEM delta in-tem reagent) (P lt 005 results

not shown)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

60

Table 3 Comparison of initial results of haemostasis parameters (median values

minimumndashmaximum in dogs with (grades 1ndash3) and without (grade 0) sonographically

detectable central venous catheter-induced thromboses

Parameter Unit

Sonographic

thrombus grade

P-value

(Mann-Whitney

U test)

0

(n = 20)

123

(n = 8)

Haematocrit 355 (228ndash641)

382 (245ndash499)

06111

Platelet count x103microl 292 (420ndash483)

218 (860ndash458)

02320

Prothrombin time 944 (359ndash169)

896 (471ndash136)

05758

Activated partial thromboplastin

time s 128

(955ndash249) 140

(117ndash164) 02322

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 448 (153ndash770)

747 (153ndash116)

00235

Antithrombin activity 639 (289ndash107)

670 (464ndash746)

05249

D-dimers microgml 006 (000ndash148)

015 (003ndash372)

03997

Thrombin generation assay nmol 130 (000ndash365)

912 (190ndash277)

08787

Clotting time (ROTEM ex-tem) s 420 (290ndash930)

430 (170ndash159)

07989

Clot formation time (ROTEM ex-

tem) s 665

(310ndash143) 470

(240ndash800) 00633

Alpha angle (ROTEM ex-tem) degrees 795 (620ndash840)

815 (750ndash860)

01667

Max clot firmness (ROTEM ex-

tem) mm 750

(570ndash860) 775

(700ndash830) 01753

Max clot elasticity (ROTEM ex-

tem) - 299

(131ndash619) 347

(235ndash484) 01544

Maximum lysis (ROTEM ex-tem) 135 (300ndash290)

400 (000ndash800)

00037

Clotting time (ROTEM in-tem) s 208 (840ndash810)

249 (720ndash371)

05085

Clot formation time (ROTEM in-

tem) s 835

(370ndash634) 745

(390ndash235) 08988

Alpha angle (ROTEM in-tem) degrees 745 (250ndash820)

750 (500ndash820)

09797

Max clot firmness (ROTEM in-

tem) mm 700

(540ndash810) 740

(610ndash770) 03587

Max clot elasticity (ROTEM in-

tem) - 235

(118ndash424) 287

(156ndash332) 04764

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

61

Maximum lysis (ROTEM in-tem) 850 (000ndash210)

350 (000ndash220)

01327

Table 4 Comparison of results of haemostasis parameters (median values minimumndash

maximum) on the last day in dogs with (grades 1ndash3) and without (grade 0) central venous

catheter-associated thrombus formation as detected by electron microscopy

Parameter Unit

Thrombus grade

(ELMI)

P-value

(Mann Whitney

U test)

0

(n = 8)

123

(n = 19)

Haematocrit 322 (241ndash454)

330 (216ndash510)

05953

Platelet count 103microl 446 (321ndash592)

268 (101ndash675)

00558

Prothrombin time 145 (104ndash264)

120 (561ndash148)

00462

Activated partial thromboplastin

time s 123

(109ndash140) 136

(109ndash188) 00665

Thrombin time s 129 (121ndash178)

141 (650ndash160)

02028

Fibrinogen gl 465 (275ndash866)

347 (189ndash992)

00491

Antithrombin activity 793 (448ndash981)

692 (398ndash889)

04572

D-dimers microgml 018 (000ndash173)

007 (000ndash118)

01920

Thrombin generation assay nmol 501 (000ndash160)

93 (000ndash287)

09562

Clotting time (ROTEM ex-tem) s 360

(270ndash410) 400

(310ndash620) 00238

Clot formation time (ROTEM ex-

tem) s 395

(120ndash740) 650

(240ndash174) 00275

Alpha angle (ROTEM ex-tem) degrees 835 (760ndash880)

790 (580ndash850)

00484

Max clot firmness (ROTEM ex-

tem) mm 780

(700ndash850) 720

(550ndash860) 00586

Max clot elasticity (ROTEM ex-

tem) - 346

(239ndash555) 254

(121ndash637) 00495

Maximum lysis (ROTEM ex-tem) 105 (200ndash190)

110 (000ndash380)

03796

Clotting time (ROTEM in-tem) s 178 (100ndash222)

201 (139ndash554)

00525

Clot formation time (ROTEM in-

tem) s 435

(330ndash111) 790

(350ndash882) 00209

Alpha angle (ROTEM in-tem) degrees 815 (710ndash830)

740 (240ndash830)

00191

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

62

Max clot firmness (ROTEM in-

tem) mm 785

(660ndash810) 690

(270ndash840) 00586

Max clot elasticity (ROTEM in-

tem) - 362

(192ndash434) 218

(370ndash507) 00559

Maximum lysis (ROTEM in-tem) 950 (200ndash190)

800 (100ndash180)

06890

4 Discussion

The main result of our study is that under the study conditions part of the dogs developed

CVC-associated thromboses However only a very low part of these thromboses were

obviously of a clinically significant degree regarding the amount of occlusion of the EJV

and none of the dogs had a symptomatic thrombosis whose definition includes presence

of clinical signs such as localised swelling pain and redness (Boersma et al 2008 Lee

and Kamphuisen 2012) In the present study we particularly refer to the extraluminal

thromboses with regard to the assessment of the antithrombotic management

One major problem of the statistical analysis and result presentation was that the two

methods used to define the occurrence and degree of thrombus formation (ultrasound and

ELMI) revealed discrepant results Therefore we performed a statistical analysis

independently regarding thrombus formation with either of the two methods It is well

known that ELMI which has already been used to detect CVC induced thromboses in

humans (Lucas et al 2017) and experimental animals (Xiang et al 1998 Xiang et al

2001) is a very sensitive diagnostic technique which allows the ex vivo diagnosis of very

small or fibrin rich thrombi that perhaps cannot be detected by ultrasonography or

venography during the in vivo evaluation of patients (Xiang et al 2001) A major reason

for the discrepant results between the two methods used to detect CVC associated

thromboses is that different CVC segments are evaluated Using ultrasonography it was

only possible to evaluate the EJV in the neck area of the patient and thereby only its

proximal third In contrast ELMI assessed mainly the CVC tip and a human study

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

63

revealed that even different positions of the catheter tip are associated with various risks

for development of CVC associated thrombosis (Ballard et al 2016) One additional point

to consider which also may contribute to the discrepancy is that it is possible that thrombi

developed on the CVC surface can be detached during the CVC removal from the vein

and thereby before its analysis by ELMI (Xiang et al 2001)

The contrast venography technique is used as gold standard to confirm thrombosis in

human medicine However this diagnostic technique is normally used only in the case of

a strong clinical suspicion of venous thrombosis that cannot be detected by

ultrasonography (Laurenson et al 2010 Langston and Eatroff 2018) Our study protocol

with privately owned clinic patients did not allow to use this method In a study performed

in 44 human patients requiring a CVC (Koumlksoy et al 1995) a sensitivity of 94 and a

specificity of 96 were determined for the colour Doppler ultrasound diagnosis when it

was compared with the diagnosis performed with the contrast venography technique and

that makes the colour Doppler ultrasound an accurate diagnosis method to detect CVC

related thrombosis even in humans (Koumlksoy et al 1995 van Rooden et al 2003)

The incidences of CVC related thrombosis detected by colour Doppler ultrasound of the

internal jugular vein reported in 3 human studies were 2 30 and 56 (Harter et al

2002 Lordick et al 2003 Wu et al 1999) and although with a wide range the median

value of these rates (30 ) is very well suited to the incidence of 28 (829) which we

calculated for our canine study based on sonographic findings The wide variation of

incidences of CVC related thrombosis in different human studies can in general be

explained by differences regarding the type and entry site of the CVC (jugular femoral

subclavian axillary cephalic vein) the time period with the CVC in place the thrombus

detection technique (venography or ultrasonography comment in the cited 3 studies

consistently the last-mentioned method was used) and criteria for diagnosis of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

64

thrombosis the duration of follow up diseases and presence of predisposing factors in

these patients (eg previous venous thrombosis history presence of cancer) and finally

prophylactic treatment with anticoagulants (van Rooden et al 2005 Geerts 2014 Liu et

al 2015)

Different pathomechanisms contribute to the development of CVC induced thromboses

During the CVC insertion the puncture of the skin on the CVC insertion site and the

rupture of the blood vessels integrity activate the coagulation cascade and the CVC acts

as a matrix for adherence and accumulation of fibrin and cellular components (Smith et

al 2012) The attachment of proteins and blood cells to the internal andor external wall

of the CVC can lead to a partial or complete occlusion of the catheter lumen and

sometimes to a partial or complete obstruction of the blood flow andor

thromboembolism (Langston and Eatroff 2018) There is a strong interrelationship

between thrombus and biofilm formation (Smith et al 2012) Prevention of thrombus

formation has thereby also relevance regarding the increased risk for microbial

colonisation of the catheter (Boersma et al 2008) leading at least in humans to an

increased number of CVC related systemic infections (Lordick et al 2003)

In this context it has to be considered that CVC produced from polyurethane which are

routinely used in our clinic and thereby in the study are usually associated with a

relatively high thrombosis potential and thereby definitely provide an effective test for

an antithrombotic regimen Silicone catheters have a significantly reduced risk for

thrombotic complications but this advantage has to be purchased by a higher rate of

mechanical failure (eg rupture of the catheter) (Wildgruber et al 2016) Promising

experimental approaches work on surface modifications of the standard catheters For

example one group was able to demonstrate that a coating of the surface of the catheter

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

65

with a zwitterionic polymeric sulfobetaine showed a significant reduction of thrombus

formation in vitro with human blood and in an in vivo canine model (Smith et al 2012)

Due to ethical reasons we did not performed a placebo-controlled study However

interestingly not all human studies report systemic anticoagulation and part of the human

studies was even placebo-controlled (review Lee and Kamphuisen 2012) Obviously

there is no consensus in the available human literature regarding the beneficial use of

antithrombotic prophylaxis to prevent CVC related thrombosis Several studies did not

find a statistically significant difference in the incidence of CVC related thrombosis

between patients receiving the antithrombotic prophylaxis compared with the control

groups without systemic anticoagulation (Verso et al 2005 P = 035 Karthaus et al

2006 P gt 005 Niers et al 2007 P = 049) whereas in other studies a lower incidence

of CVC related thrombosis was found in the group receiving anticoagulants as

prophylaxis compared with the control groups (Abdelfeki et al 2004 150 vs 126

P = 003 De Cicco et al 2009 400 vs 526 P = 005)

Although it is well known that subcutaneously administered UFH has a limited and

variable bioavailability and thereby leads to unpredictable blood activities (Hirsh and

Raschke 2004) the extremely wide range of blood activities after a defined dosage in our

study was relatively unexpected Blood was collected 2 hours after the first UFH

injection and thereby at the time of suspected maximum heparin activity considering the

relatively low UGH dosage (Mischke and Jacobs 2001) The cited study was performed

on healthy experimental dogs Possible additional influencing factors in critical ill

patients include circulatory insufficiency which may reduce and delay absorption from

subcutanenous tissue decreased clearance function of the liver and kidneys as well as

altered blood plasma concentrations of heparin binding plasma proteins such as

fibrinogen and platelet factor 4 and alterations of the heparin binding capacity of

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

66

endothelial cells which altogether are reasons for a great inter- but also intrapatient

variability (Hirsh and Raschke 2004) In individual patients this can lead to heparin

resistance It has also to be considered that antithrombin activities which are necessary

for an adequate heparin effect (Bjoumlrk and Lindahl 1982) were reduced in our patients

Availability of approved blood tests which can predict the risk of development of CVC

associated thrombosis already at the moment of the CVC placement are important

(Boersma et al 2016) One of the most interesting results was that dogs which developed

sonographically detectable thrombi had nearly twofold higher fibrinogen levels Well in

agreement with our results one study performed in anticoagulated human patients

demonstrated that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 76 (1621) of the venous

thromboses were suggested to be associated with CVCs Fibrinogen measurement a test

which is widely available is therefore a useful and practicable indicator of an increased

risk for CVC induced thromboses and may advice the clinician to treat these patients

especially carefully (eg with low molecular weight heparin) In the study performed by

Boersma et al (2016) in 168 human patients without indication of anticoagulation

prophylaxis elevated leukocyte counts high levels of plasminogen activator inhibitor

and of coagulation factor VIII were associated with an increased incidence of

symptomatic CVC related thrombosis diagnosed by ultrasonography A limitation of that

study is that all measurements were performed on blood samples taken after CVC

placement Interestingly in our study ROTEM results were partly associated with

thrombus formation which reflects the well-known fact that results of viscoelastic

measurements are a good indicator of hypercoagulability and thromboembolic events

(Harahsheh and Ho 2018) It is of further interest and possibly unexpeted that on the final

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

67

day of examination dogs with ELMI signs of thromboses had partly lower coagulation

activities than dogs without thromboses Due to the low number of dogs without

detectable thrombi using ELMI this implausible result has to be interpreted with caution

One possible explanation could be that temporary hypercoagulability in these dogs may

have finally resulted in consumption of haemostasis substrates

Limitations of the present study were as a consequence of the study design based on

clinical patients the relatively low total number of patients inclusion of different UFH

treatment regimens and the lack of a control group

5 Conclusions

The results of the present study indicate that the used UFH regimen was not completely

effective to prevent CVC-associated thromboses which were detectable in 28 [829]

of the patients by ultrasound However severe thromboses (high grades) are relatively

rare in intensive care canine patients receiving a routine anticoagulant treatment Because

heparin plasma activities were not different in dogs with or without CVC-associated

thrombus formation other local and systemic factors must play a significant role as well

A high fibrinogen concentration and low maximum lysis (ROTEM ex-tem) may be

useful indicators for an increased thrombotic risk in individual patients which may

require special anticoagulatory treatment

6 Acknowledgments

Jose Mauricio Pereira received a scholarship from the Universidad Nacional de Costa

Rica (reference code JB-C-1107-2016)

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

68

7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb S

Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim A

2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled trial

Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications associated

with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash257

3 Bagge A Schoumltt U Kander T 2016 Effects of naturopathic medicines on Multiplate and ROTEM

a prospective experimental pilot study in healthy volunteers BMC Complement Altern Med 16 64

4 Ballard DH Samra NS Gifford KM Roller R Wolfe BM Owings JT 2016 Distance of the

internal central venous catheter tip from the right atrium is positively correlated with central venous

thrombosis Emerg Radiol 23 269ndash273

5 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological malignancies Ann

Oncol 19 433ndash442

6 Boersma R S Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten HC

2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients with

hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Bjoumlrk I Lindahl U 1982 Mechanism of the anticoagulant action of heparin Mol Cell Biochem

48 161ndash182

8 Claude A Riedesel D Riedesel E 2010 Electrocardiography-guided and retrospective analysis of

central venous catheter placement in the dog Vet Anaesth Analg 37 97ndash105

9 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti M M Schaan

DB Rados VD 2017 Complications of central venous catheter insertion in a teaching hospital

Rev Assoc Med Bras 63 613ndash620

10 Crochemore T Piza FMT Rodriacutegues RDR Guerra JCC Ferraz LJR Correcirca TD 2017

A new era of thromboelastometry Einstein (Sao Paulo) 15 380ndash385

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

69

11 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin D

Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in cancer

patients a randomized controlled study based on serial venographies Ann Oncol 20 1936ndash1942

12 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

13 Geerts W 2014 Central venous catheterndashrelated thrombosis Hematology Am Soc Hematol Educ

Program 2014 306ndash311

14 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic events A

systematic review and meta‐analysis Eur J Haematol 100 113ndash123

15 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-related

infection and thrombosis of the internal jugular vein in hematologic-oncologic patients undergoing

chemotherapy a prospective comparison of silver-coated and uncoated catheters Cancer 94 245ndash251

16 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188S-203S

17 Hundley D Brooks A Thomovsky E Jhonson P Freeman L Schafbuch R Heng HG Moore

G 2018 Comparison of ultrasound-guided and landmark-based techniques for central venous

catheterization via the external jugular vein in healthy anesthetized dogs Am J Vet Res 79 628ndash

636

18 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of and

risk factors for venous thromboembolism during surgical treatment for esophageal cancer a single-

institution study Surg Today 46 445ndash452

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber C

Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti C

Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer patients with

central venous catheters final results of a double‐blind placebo‐controlled phase III trial Ann Oncol

17 289ndash296

20 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of colour

Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687-689

21 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J Med

364 861ndash869

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

70

22 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J Vet

Emerg Crit Care 28 366ndash371

23 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and conditions

in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

24 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related thrombosis in

patients with cancer J Thromb Haemost 10 1491ndash1499

25 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash449

26 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C 2003

Ultrasound screening for internal jugular vein thrombosis aids the detection of central venous catheter-

related infections in patients with haemato-oncological diseases a prospective observational study Br

J Haematol 120 1073ndash1078

27 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters Rev

Esc Enferm USP 51 1ndash9

28 Malick L Wilson R Stetson D 1976 Modified thiocarbo- hydrazide procedure for scanning

electron microscopy routine use for normal pathological and experimental tissue Stain Technol 50

265ndash269

29 Mischke R Jacobs C 2001 The monitoring of heparin administration by screening tests in

experimental dogs Res Vet Sci 70 101ndash108

30 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007 Prevention

of catheter‐related venous thrombosis with nadroparin in patients receiving chemotherapy for

hematologic malignancies a randomized placebo‐controlled study J Thromb Haemost 5 1878ndash

1882

31 Reminga C Silverstein D Drobatz K Clarke D 2018 Evaluation of the placement and

maintenance of central venous jugular catheters in critically ill dogs and cats J Vet Emerg Crit Care

28 232ndash243

32 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver D

Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012 Vascular

Chapter 3 Central venous catheter induced thrombosis in dogs ndash efficacy of standard

heparin prophylaxis and correlation to haemostatic changes

71

catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus formation and

microbial attachment Sci Transl Med 4 153ra132 doi101126scitranslmed3004120

33 Toskala E Nuutinen J Rautiainen M 1995 Scanning electron microscopy findings on human

respiratory cilia in chronic sinusitis and in recurrent respiratory infections J Laryngol Otol 109 509ndash

514

34 Van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash2419

35 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise P

Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin for the

prevention of venous thromboembolism associated with central vein catheter a double‐blind placebo‐

controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

36 Villalta-Garciacutea P Loacutepez-Herranz M Mazo-Pascual S Honrubia-Fernaacutendez T Jaacutentildeez-Escalada

L Fernaacutendez-Peacuterez C 2015 Reliability of blood test results in samples obtained using a 2-mL discard

volume from the proximal lumen of a triple-lumen central venous catheter in the critically ill patient

Nurs Crit Care 22 298ndash304

37 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in healthy

dogs PeerJ 7 e7072 httpsdoiorg107717peerj7072

38 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler M

Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124 doi101016jejca201602011

39 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi after

short-term central venous catheterization of the internal jugular vein J Clin Anesth 11 482ndash485

40 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

41 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

Chapter 4 General discussion

72

Chapter 4 General discussion

Incidence of CVC related thrombosis

In our study the incidence of CVC related thrombosis detected by ultrasound in critical

care canine patients was 28 To the best of the authoracutes knowledge similar studies

performed in dogs are rare in the available veterinary literature In one study performed

in 81 dogs and 12 cats (Adamantos et al 2010) an incidence of 2 was found for vein

thrombosis related to CVC This low incidence can be due to a lacking of systematic

ultrasound evaluations of the jugular vein containing the CVC In the mentioned study in

dogs and cats the vein thrombosis was confirmed by palpation or ultrasound presumably

after the presentation of clinical symptoms however in many cases as in our study

thrombus into the jugular vein can be developed without manifestation of clinical

symptoms what makes a routinely ultrasound evaluation of the vein containing the CVC

very important Our results are well aligned with the findings in humans in which a

median incidence of 30 was found for thrombosis related to CVC based on 25 studies

performed in patients requiring a CVC (review article van Rooden et al 2005) In the

mentioned 25 human studies the incidence of CVC related thrombosis varied within a

wide range (2ndash67 ) according to factors related to (1) the CVC (eg material type

entry site [jugular femoral subclavian axillary cephalic vein] time period in place (2)

diagnostic thrombus detection technique (ultrasonography or venography) and criteria

(3) patient collective (diseases and presence of predisposing factors [eg previous venous

thrombosis history presence of cancer] and (4) prophylactic antithrombotic treatment

(Geerts 2014 Liu et al 2015 van Rooden et al 2005)

The incidence of CVC related thrombosis detected by colour Doppler ultrasound in our

study fits also with the median incidence (30 ) found in three studies performed in

humans in which colour Doppler ultrasound evaluations of the jugular vein containing a

Chapter 4 General discussion

73

CVC were performed (Harter et al 2002 Lordick et al 2003 Wu et al 1999) In these

three studies the individual incidence of CVC related thrombosis was 2 (Harter et al

2002) 30 (Lordick et al 2003) and 56 (Wu et al 1999) The high thrombus

incidence reported in the study performed by Wu et al (1999) could be due to the criteria

for diagnosis of thrombosis because in that study both fibrin sleeves and compact

thrombi were considered as cases of thrombosis related to CVC The incidence for

compact thrombi was 25 which is more similar to our results In the study performed

by Harter et al (2002) the CVC related thrombosis incidence (2 ) was very low

compared with the incidence found in our study In that study the jugular veins were

examined just before or less than 24 hours after CVC removal (median period of CVC

use = 1025 days) In our study the ultrasound evaluations were performed before the

CVC placement (day 0) and then on days 1 3 5 etc until the CVC removal (median

period of CVC use = 7 days median interval until thrombus detection = 7 days) In our

study we detected thrombus formation as early as on the 3rd day so it could be possible

that in the cited study possibly only temporarily existing thrombi could have been missed

by ultrasonography in the final stage and thereby for the incidence calculation However

this scenario is not very likely because results of the present study show that

sonographically detectable thrombi during the observation period had in all cases their

maximum grade on the final day

In the study performed by Lordick et al (2003) the CVC related thrombosis incidence

determined for 43 human patients was very similar to our study and in both studies only

compact thrombi detectable by ultrasound were considered for assessment of the

incidence In the mentioned study all the human patients were screened for the presence

of thrombosis every 4 days after CVC insertion with real time B-mode ultrasound

(Lordick et al 2003)

Chapter 4 General discussion

74

In our study 70 of CVCs analysed by ELMI showed thrombi of grade 1ndash3 on the

external CVC wall To the best of the authoracutes knowledge studies about incidence of CVC

related thrombosis on the outside surface of the CVC detected by ELMI cannot be found

in the human and small animal literature In one study performed in humans (Lucas et al

2017) thrombus formation was analysed by ELMI in 78 central venous catheters That

study focussed on microstructural assessment of thrombus formation inside the distal

openings of the CVC which were found in 85 of the evaluated CVCs The percentage

of occurrence of thrombosis in the mentioned study in humans was nearly similar to the

incidence of intraluminal thrombus formation as detected in the presented study by ELMI

(74 ) In general intraluminal thrombi were not specially considered in the present

study because they mainly express the efficacy of the flushing technique rather than the

effectivity of the antithrombotic treatment In one study performed in rats the

microstructure of thrombi developed in the jugular vein containing the CVC was also

evaluated by ELMI (Xiang et al 1998) In the mentioned study pericatheter thrombosis

was detected on the proximal section of the CVC in 100 of rats (n = 12) after 1 and 3

days of catheterisation The analyses of the CVCs placed in rats were performed with the

catheter inside the vein in contrast to our study where the CVC was removed from the

jugular vein for the respective analysis Xiang et al (2001) hypothesised in one study

performed in rabbits and rats in which CVCs were placed in the jugular vein that fibrin

rich thrombi associated to the CVC can be detached during the catheter removal and this

could explain the difference between the thrombosis incidence assessed in our study and

that of the study performed by Xiang et al (1998)

In our study the incidence of CVC related thrombosis detected by ultrasonography in the

lumen of the EJV did not correlate with the incidence of CVC related thrombosis detected

by ELMI on the external wall of the CVC This was mainly due to the fact that the CVC

Chapter 4 General discussion

75

segment evaluated with both methods was different Using ultrasonography it was only

possible to evaluate the EJV in the neck area of the patient and thereby only the proximal

third of the CVC In contrast with the ELMI analysis only the distal section of the CVC

(last third) was evaluated because macroscopically detected thrombi mainly developed

near the CVC tip which was routinely positioned at the junction of the cranial cava vein

with the right atrium In a retrospective review of medical records of 169 human patients

requiring a CVC (Ballard et al 2016) the position of the CVC tip was determined by

radiologic evaluation and it was determined that the catheter tip position can be a

predisposing factor for central venous thrombosis development Thereby it could be of

interest to evaluate in further studies in dogs if a high incidence of thrombosis detected

by ELMI as was shown in our study could be related to the CVC tip position

In our study polyurethane catheters were tested which are widely used and are the

standard CVCs used in the Small Animal Clinic University of Veterinary Medicine

Hannover In vivo and in vitro studies have assessed the thrombogenicity of different

catheter materials (Borow and Crowley 1985 Smith et al 2012) In general terms

polyurethane catheters (central or peripheral) had shown a higher thrombogenicity

compared with coated (eg with hydromer) and silicone catheters Due to this property

of polyurethane catheters they can be supposed as a good test system for antithrombotic

regimens as evaluated in our study On the other hand despite silicone catheters have been

associated with a lower thrombogenicity their mechanical flexibility could be a

disadvantage in small clinical practice due to the increased risk of rupture detected even

in humans (Wildgruber et al 2016)

Chapter 4 General discussion

76

Heparin thromboprophylaxis for CVC related thrombosis

The occurence of CVC related thrombosis detected by colour Doppler ultrasound and

ELMI in our study demonstrates that the UFH in the given dosage was not completely

effective to prevent CVC related thrombosis in canine intensive care patients In human

medicine LMWH is commonly used in prophylactic anticoagulation regimens however

studies performed in human patients requiring a CVC and receiving a prophylactic

anticoagulation (with LMWH or UFH) did not achieve consensus regarding the efficacy

of prophylactic heparin to prevent CVC related thrombosis (Abdelfeki et al 2004

Lordick et al 2003 Wu et al 1999) To the best of the authoracutes knowledge studies

performed in dogs evaluating systemic prophylactic anticoagulation to prevent CVC

related thrombosis cannot be found However according to the evidence found in the

human literature and the results of our study further investigations using comparative

groups with different drugs dosages andor administration routes are required to establish

an effective antithrombotic treatment to prevent CVC related thrombosis

The plasma heparin activities measured in canine patients in our study varied significantly

after a defined subcutaneous dosage The limited and variable bioavailability of SC

administered UFH is well documented in humans (Hirsh and Raschke 2004)

Considerable variations of the circulatory status of the liver- and kidney function and

concentration of heparin-binding proteins in intensive care patients obviously amplify

this phenomenon leading to an extremely high inter-individual and intra-individual

variation of the heparin blood levels in our patients A different application route (IV)

may have reduced the influence of part of these factors and therefore could have led to

more predictable UFH blood levels However continuous infusion requires more

technical equipment and is also associated with a wide range of heparin activities in

canine patients (Scott et al 2009)

Chapter 4 General discussion

77

Prediction of CVC related thrombosis

Of 7 haemostasis tests (including ROTEM with 12 different parameters [6 for each

reagent ex-tem and in-tem]) evaluated as possible predictors of CVC related thrombosis

in our study initial values of high fibrinogen and low maximum lysis (ROTEM ex-tem)

were associated with sonographically detectable thrombus formation Some studies in

humans have established the association of laboratory biomarkers as predictors of risk

thrombosis (Boersma et al 2016 Cheng et al 2013 Kato et al 2016 Liu et al 2015)

One study performed in anticoagulated human patients showed ndash well in agreement with

our results ndash that high fibrinogen levels were associated with an increased risk of

perioperative venous thromboembolism (P = 0004) (Kato et al 2016) The cited study

does not exclusively refer to CVC associated thrombosis but 762 (1621) of the

venous thromboses were suggested to be associated with CVCs Fibrinogen concentration

measurements are normally available in the human and veterinary practice and the

association of this parameter with a high risk of development of CVC related thrombosis

can be useful to choose special antithrombotic treatment in individual critical care patients

to prevent such severe complications

Thrombin generation assay (TGA) abnormalities such as higher velocity and endogenous

thrombin potential (ETP) values measured initially ie before any anticoagulation

therapy have been associated with a higher thromboembolism risk in general (ie without

specific association with CVCs) in one study performed on 105 human patients (Espitia

et al 2017) In contrast in our study the initial TGA assay results did not show an

association with formation of CVC related thrombosis

The ROTEM parameters evaluated in the present study were partially associated with

formation of CVC thrombosis detected by colour Doppler ultrasound This finding

confirms the results obtained in humans in which ROTEM measurements are regarded as

Chapter 4 General discussion

78

a valuable indicator of hypercoagulability (Harahsheh and Ho 2018 Tuumlrk et al 2018)

However in our study individual ROTEM parameters showed partly lower coagulation

activities in dogs with CVC related thrombosis detected by ELMI at the final day of

evaluation Because the number of patients without signs of thrombosis detected by ELMI

was low compared with those who showed signs these implausible results should be

interpreted carefully The usefulness of ROTEM analysis which has also many other

indications demonstrates that is was worth to perform a detailed methodological approval

of the device including the establishment of reference values in the preliminary study

Conclusion

In conclusion the prethrombotic or thrombotic changes detected in our study did not

correlate between ultrasonography and ELMI however both methods show accurately

signs of CVC related thrombosis The UFH in the given dosage was not completely

effective to prevent CVC induced thrombus formation in canine intensive care patients

Plasma heparin activities varied significantly after a defined subcutaneous dosage

However plasma heparin activities were not different between dogs with and without

detectable CVC associated thromboses Initial high fibrinogen values and low maximum

lysis (ROTEM ex-tem) values were associated with sonographically detectable thrombi

and may be useful as indicators for an increased thrombotic risk

Finally our study in canine intensive care patients is to the best of the authoracutes

knowledge the first study in the available veterinary literature performed to evaluate the

efficacy of an antithrombotic treatment with UFH to prevent CVC related thrombosis

Chapter 5 Summary

79

Chapter 5 Summary

Central venous catheter induced thrombosis in dogs ndash occurrence under an

antithrombotic treatment regimen and possible indicators

Joseacute Mauricio Pereira Mora

Central venous catheters (CVC) are of importance for intensive care management of

canine patients Thrombosis of the jugular vein is a severe complication related to the

CVC use However only little is known about the incidence of CVC-associated

thromboses in dogs and the necessity and efficacy of prophylactic antithrombotic

regimens

Therefore the aims of the present study were (1) to determine the incidence of catheter-

induced thrombosis in canine intensive care patients (2) to evaluate the efficacy of a

prophylactic heparin regimen routinely used in the Small Animal Clinic of the University

of Veterinary Medicine Hannover and (3) to assess whether selected haemostasis test

results are predictive for an increased risk for thrombosis formation

The thesis also includes a preliminary methodological study on the use of rotational

elastometry in dogs using the ROTEM delta device In this preliminary study reference

intervals were determined based on 125 healthy dogs for different clot activator reagents

and for measurements without activator Coefficients of variation (CVs) were calculated

based on results of fourfold measurements of 3 normal and 3 abnormal blood samples in

order to assess precision of the method Influence of age groups sex neuter status and

size were also assessed The median CVs for most of the parameters evaluated in the

precision analysis were less than 10 and this confirmed that the ROTEM delta analyser

delivers repeatable results Regarding the influence of age sex neuter status and animal

Chapter 5 Summary

80

size only neuter status in female dogs appeared to be a relevant influencing factor on

most of important ROTEM parameters

The main study of the thesis comprises 29 patients hospitalised at the Small Animal Clinic

requiring a CVC for medical reasons These patients received a prophylactic

anticoagulation treatment with unfractionated heparin (UFH 150 IU [surgical dogs

received 75 IU]kg BW TID subcutaneously) A colour Doppler ultrasound of the

respective external jugular vein (EJV) in order to detect thrombus formation in the lumen

of the EJV and blood collection for haemostasis tests were performed directly before the

installation procedure (day 0) and on days 1 3 5 7 and so on after the CVC insertion

Haemostasis tests included prothrombin time activated partial thromboplastin time

thrombin time antithrombin activity D-dimer concentration thrombin generation and

rotational elastometry as well as the heparin activity using a chromogenic anti-factor Xa

test Finally when the CVC was removed electron microscopy (ELMI) of sections of the

CVC with macroscopically visible thrombi in the last third (5ndash10 cm) of the CVC was

performed 3 different grades were defined for CVC-associated (extraluminal) thrombi

detected by either ultrasound or ELMI respectively

829 patients evaluated by colour Doppler ultrasound during the study developed thrombi

into the lumen of EJV (4 x grade 1 3 x grade 2 and 1 x grade 3) 1927 CVCs assessed

using ELMI showed detectable thrombi on the external CVC surface (7 x grade 1 4 x

grade 2 and 8 x grade 3) There was no significant correlation between thrombus

formation detected by ultrasound or ELMI Thus statistical group comparisons of

patients with or without CVC-associated thrombus formation was performed

independently for both diagnostic procedures The plasmatic heparin activities showed

great variability but no significant differences were found between dogs with or without

CVC-associated thrombus formation regardless whether this was based on sonographic

Chapter 5 Summary

81

or ELMI findings From the initially performed haemostasis tests high fibrinogen

concentrations and low maximum lysis values as measured by the ROTEM delta device

(using the ex-tem reagent) were associated with sonographically detectable thrombi

In conclusion the preliminary methodological study on the ROTEM delta in dogs shows

that the method delivers repeatable results in canine citrated whole blood and the

reference values obtained should offer a reliable basis for the interpretation of test results

of this device in dogs with suspected haemostasis disorders The results of the main study

indicate that CVC-associated thromboses occur (eg 28 [829] of the patients as

detected by ultrasound) although severe thromboses (high grades) are relatively rare in

intensive care canine patients receiving a routine anticoagulant treatment Thereby

prophylactic antithrombotic protocol (unfractionated heparin [UFH]) in the given dosage

was not completely effective to prevent CVC induced thrombus formation in canine

intensive care patients A high fibrinogen concentration and low maximum lysis

(ROTEM ex-tem) may be useful indicators for an increased thrombotic risk in individual

patients which may require special anticoagulatory treatment Further studies are

required to evaluate whether different thromboprophylactic treatment regimens eg with

low molecular weight heparin are more effective in preventing CVC-associated

thromboses in dogs

Chapter 6 Zusammenfassung

82

Chapter 6 Zusammenfassung

Zentralvenenkatheter-induzierte Thrombose bei Hunden - Auftreten unter einem

antithrombotischen Behandlungsschema und moumlgliche Indikatoren

Joseacute Mauricio Pereira Mora

Zentralvenenkatheter (ZVK) sind fuumlr das Intensivmanagement von Hunden von

Bedeutung Die Entstehung einer Thrombose der Vena jugularis ist eine schwerwiegende

Komplikation im Zusammenhang mit der Verwendung eines ZVKs Uumlber die Inzidenz

von ZVK-assoziierten Thrombosen bei Hunden und die Notwendigkeit und Wirksamkeit

prophylaktischer antithrombotischer Protokolle liegen jedoch nur wenige fundierte

Angaben in der Literatur vor

Daher waren die Ziele der vorliegenden Studie (1) die Inzidenz der Katheter-induzierten

Thrombose bei caninen Intensivpatienten zu bestimmen (2) die Wirksamkeit eines

prophylaktischen Heparin-Regimes zu bewerten das routinemaumlszligig in der Klinik fuumlr

Kleintiere der Stiftung Tieraumlrztliche Hochschule Hannover angewendet wird und (3) um

zu beurteilen ob ausgewaumlhlte Haumlmostasetestergebnisse ein erhoumlhtes Risiko fuumlr die

Thrombosebildung vorhersagen koumlnnen

Die PhD-Arbeit enthaumllt zudem eine vorbereitende methodische Studie zur Verwendung

der Rotationselastometrie bei Hunden mit Hilfe des ROTEM delta-Geraumltes In dieser

Vorstudie wurden Referenzintervalle basierend auf 125 gesunden Hunden fuumlr

verschiedene Gerinnungsaktivatorreagenzien und fuumlr Messungen ohne Aktivator

bestimmt Um die Genauigkeit der Methode einzustufen wurden Variationskoeffizienten

(VKs) basierend auf den Ergebnissen von Vierfachmessungen von drei normalen und drei

abnormalen Blutproben berechnet Der Einfluss von Alter Geschlecht Kastration und

Groumlszlige wurde ebenfalls getestet Die Medianwerte der VKs fuumlr die meisten in der

Chapter 6 Zusammenfassung

83

Praumlzisionsanalyse bewerteten Parameter betrugen weniger als 10 und dies bestaumltigte

dass das ROTEM delta-Analysegeraumlt wiederholbare Ergebnisse liefert In Bezug auf den

Einfluss von Alter Geschlecht Kastration und Tiergroumlszlige erwies sich nur eine Kastration

bei Huumlndinnen als relevanter Einflussfaktor fuumlr die meisten wichtigen ROTEM-

Parameter

Die Hauptstudie der Arbeit umfasst 29 Patienten welche zum Zeitpunkt der

Untersuchung stationaumlr in der Kleintierklinik aufgenommen worden waren und aus

medizinischer Indikation einen ZVK benoumltigten Diese Patienten erhielten eine

prophylaktische Antikoagulation mit unfraktioniertem Heparin (UFH 150 IE

[chirurgische Hunde erhielten 75 IE]kg KG 3 x taumlglich subkutan) Bildgebende

Untersuchungen mittels Farbdoppler-Ultraschall der jeweiligen Vena jugularis externa

(VJE) zum Nachweis der Thrombusbildung im Lumen der VJE und Blutentnahmen fuumlr

Haumlmostasetests wurden jeweils direkt vor dem Installationsvorgang (Tag 0) sowie an den

Tagen 1 3 5 und 7 nach Legen des ZVKs durchgefuumlhrt Haumlmostasetests umfassten die

Prothrombinzeit die aktivierte partielle Thromboplastinzeit die Thrombinzeit die

Antithrombinaktivitaumlt die D-Dimer-Konzentration die Thrombingenerierung und die

Rotationselastometrie sowie die Heparinaktivitaumlt unter Verwendung eines chromogenen

Anti-Faktor-Xa-Tests Schlieszliglich wurde nach Entfernung des ZVKs eine

Elektronenmikroskopie (ELMI) von Abschnitten aus dem letzten Drittel (5ndash10 cm) des

ZVKs mit makroskopisch sichtbaren Thromben durchgefuumlhrt Fuumlr ZVK-assoziierte

(extraluminale) Thromben die entweder durch Ultraschall oder ELMI nachgewiesen

wurden wurden drei verschiedene Grade definiert

829 Patienten die waumlhrend der Studie mittels Farbdoppler-Ultraschall untersucht

wurden entwickelten Thromben im Lumen der VJE (4 x Grad 1 3 x Grad 2 und 1 x Grad

3) 1927 ZVKs die unter Verwendung von ELMI bewertet wurden zeigten

Chapter 6 Zusammenfassung

84

nachweisbare Thromben auf der aumluszligeren ZVK-Oberflaumlche (7 x Grad 1 4 x Grad 2 und 8

x Grad 3) Es gab keine signifikante Korrelation zwischen der durch Ultraschall oder

ELMI festgestellten Thrombusbildung Daher wurden statistische Gruppenvergleiche

von Patienten mit oder ohne ZVK-assoziierter Thrombusbildung fuumlr beide diagnostische

Verfahren unabhaumlngig voneinander durchgefuumlhrt Die plasmatischen Heparinaktivitaumlten

zeigten eine groszlige Variabilitaumlt es wurden jedoch keine signifikanten Unterschiede

zwischen Hunden mit oder ohne ZVK-assoziierter Thrombusbildung gefunden

unabhaumlngig davon ob diese auf sonographischen oder ELMI-Befunden beruhte Von den

initial durchgefuumlhrten Haumlmostasetests waren hohe Fibrinogenkonzentrationen und

niedrige maximale Lyse-Werte der Messungen mit ROTEM delta (ex-tem Reagenz) mit

sonographisch nachweisbaren Thromben assoziiert

Zusammenfassend zeigt die vorbereitende methodische Studie zum ROTEM delta bei

Hunden dass die Methode wiederholbare Ergebnisse in mit Zitrat

antikoaguliertemVollblut liefert Die erarbeiteten Referenzwerte sollten eine zuverlaumlssige

Grundlage fuumlr die Interpretation der Testergebnisse dieses Geraumlts bei Hunden mit

Verdacht auf Haumlmostasestoumlrungen bieten Die Ergebnisse der Hauptstudie zeigen dass

ZVK-assoziierte Thrombosen auftreten (z B 28 [829] der Patienten bezogen auf den

Ultraschall-gestuumltzten Nachweis) obwohl schwere Thrombosen (hohe Grade) bei

intensivmedizinisch versorgten Hundepatienten die eine Routineantikoagulation

erhalten relativ selten sind Somit war das prophylaktische antithrombotische Protokoll

(unfraktioniertes Heparin [UFH]) in der angewendeten Dosierung nicht vollstaumlndig

wirksam um die ZVK-induzierte Thrombusbildung bei caninen Intensivpatienten zu

verhindern Eine hohe Fibrinogenkonzentration und eine niedrige maximale Lyse

(ROTEM ex-tem) koumlnnen als Indikatoren eines erhoumlhten Thromboserisikos bei einzelnen

Patienten nuumltzlich sein fuumlr die moumlglicherweise eine spezielle gerinnungshemmende

Chapter 6 Zusammenfassung

85

Behandlung notwendig ist Weitere Studien sind erforderlich um zu untersuchen ob

verschiedene thromboprophylaktische Behandlungsschemata z B mit

niedermolekularem Heparin bezuumlglich der Verhinderung von ZVK-assoziierten

Thrombosen bei Hunden wirksamer sind

Chapter 7 References

86

Chapter 7 References

1 Abdelkefi A Ben Othman T Kammoun L Chelli M Ben Romdhane N Kriaa A Ladeb

S Torjman L Lakhal A Achour W Ben Hassen A Hsaiumlri M Ladeb F Ben Abdeladhim

A 2004 Prevention of central venous line-related thrombosis by continuous infusion of low-dose

unfractionated heparin in patients with haemato-oncological disease A randomized controlled

trial Thromb Haemost 92 654ndash661

2 Adamantos S Brodbelt D Moores A L 2010 Prospective evaluation of complications

associated with jugular venous catheter use in a veterinary hospital J Small Anim Pract 51 254ndash

257

3 Ballard D H Samra N S Gifford K M Roller R Wolfe B M Owings J T 2016

Distance of the internal central venous catheter tip from the right atrium is positively correlated

with central venous thrombosis Emerg Radiol 23 269ndash273

4 Baumann-Kreuziger L Onwuemene O Kolesar E Crowther M Lim W 2015 Systematic

review of anticoagulant treatment of catheter-related thrombosis Thromb Res 136 1103ndash1109

5 Bliss SP Bliss SK Harvey HJ 2002 Use of recombinant tissue‐plasminogen activator in a

dog with chylothorax secondary to catheter‐associated thrombosis of the cranial vena cava J Am

Anim Hosp Assoc 38 431ndash435

6 Boersma RS Hamulyak K van Oerle R Tuinenburg A Ten Cate-Hoek AJ Schouten

HC 2016 Biomarkers for prediction of central venous catheter related-thrombosis in patients

with hematological malignancies Clin Appl Thromb Hemost 22 779ndash784

7 Boersma RS Jie KS Verbon A van Pampus EC Schouten HC 2008 Thrombotic and

infectious complications of central venous catheters in patients with hematological

malignancies Ann Oncol 19 433ndash442

8 Bonnet F Loriferne JF Texier JP Texier M Salvat A Vasile N 1989 Evaluation of

Doppler examination for diagnosis of catheter-related deep vein thrombosis Intensive Care Med

15 238ndash240

9 Borow M Crowley J G 1985 Evaluation of central venous catheter thrombogenicity Acta

Anaesthesiol Scand Suppl 81 59ndash64

10 Cheng Y Cui T Fu P Liu F Zhou L 2013 Dyslipidemia is associated with tunneled-cuffed

catheter-related central venous thrombosis in hemodialysis patients a retrospective multicenter

Chapter 7 References

87

study Artif Organs 37 155ndash161

11 Comerlato PH Rebelatto TF Santiago de Almeida FA Klein LB Boniatti MM

Schaan BD Rados DV 2017 Complications of central venous catheter insertion in a teaching

hospital Rev Assoc Med Bras 63 613ndash620

12 De Cicco M Matovic M Balestreri L Steffan A Pacenzia R Malafronte M Fantin

D Bertuzzi CA Fabiani F Morassut S Bidoli E Veronesi A 2009 Early and short‐term

acenocumarine or dalteparin for the prevention of central vein catheter‐related thrombosis in

cancer patients a randomized controlled study based on serial venographies Ann

Oncol 20 1936ndash1942

13 Espitia O Fouassier M Hardouin JB Pistorius MA Agard C Planchon B Trossaert M

Pottier P 2017 Thrombin generation assay in hospitalized nonsurgical patients A new tool to

assess venous thromboembolism risk Clin Appl Thromb Hemost 23 45ndash51

14 Evans NS Ratchford EV 2018 Catheter related-venous thrombosis Vasc Med 23 411ndash413

15 Geerts W 2014 Central Venous Catheter-related thrombosis Hematology Am Soc Hematol

Educ Program 2014 306ndash311

16 Harahsheh Y Ho KM 2018 Use of viscoelastic tests to predict clinical thromboembolic

events A systematic review and meta‐analysis Eur J Haematol 100 113ndash123

17 Harter C Salwender HJ Bach A Egerer G Goldschmidt H Ho AD 2002 Catheter-

related infection and thrombosis of the internal jugular vein in hematologic-oncologic patients

undergoing chemotherapy a prospective comparison of silver-coated and uncoated

catheters Cancer 94 245ndash251

18 Hirsh J Raschke R 2004 Heparin and low-molecular-weight heparin The Seventh ACCP

Conference on Antithrombotic and Thrombolytic Therapy Chest 126 (Suppl) 188Sndash203S

19 Karthaus M Kretzschmar A Kroning H Biakhov M Irwin D Marschner N Slabber

C Fountzilas G Garin A Abecasis NG Baronius W Steger GG Sudhoff T Giorgetti

C Reichardt P 2006 Dalteparin for prevention of catheter‐related complications in cancer

patients with central venous catheters final results of a double‐blind placebo‐controlled phase III

trial Ann Oncol 17 289ndash296

20 Kato F Takeuchi H Matsuda S Kawakubo H Omori T Kitagawa Y 2016 Incidence of

and risk factors for venous thromboembolism during surgical treatment for esophageal cancer a

Chapter 7 References

88

single-institution study Surg Today 46 445ndash452

21 Kohler TR Kirkman TR 1998 Central venous catheter failure is induced by injury and can

be prevented by stabilizing the catheter tip J Vasc Surg 28 59ndash66

22 Koumlksoy C Kuzu A Kutlay J Erden I Ozcan H Ergicircn K 1995 The diagnostic value of

colour Doppler ultrasound in central venous catheter related thrombosis Clin Radiol 50 687ndash

689

23 Kucher N 2011 Clinical practice Deep-vein thrombosis of the upper extremities N Engl J

Med 364 861ndash869

24 Langston CE Eatroff AE 2018 Hemodialysis catheter-associated fibrin sheath in a dog J

Vet Emerg Crit Care 28 366ndash371

25 Langston CE Eatroff AE Poeppel K 2014 Use of tissue plasminogen activator in catheters

used for extracorporeal renal replacement therapy J Vet Intern Med 28 270ndash276

26 Laurenson MP Hopper K Herrera MA Johnson EG 2010 Concurrent diseases and

conditions in dogs with splenic vein thrombosis J Vet Intern Med 24 1298ndash1304

27 Lee AYY Kamphuisen PW 2012 Epidemiology and prevention of catheter‐related

thrombosis in patients with cancer J Thromb Haemost 10 1491ndash1499

28 Liu G Fu ZQ Zhu P Li SJ 2015 Central venous catheter-related thrombosis in senile male

patients New risk factors and predictors J Huazhong Univ Sci Technolog Med Sci 35 445ndash

449

29 Lordick F Hentrich M Decker T Hennig M Pohlmann H Hartenstein R Peschel C

2003 Ultrasound screening for internal jugular vein thrombosis aids the detection of central

venous catheter-related infections in patients with haemato-oncological diseases a prospective

observational study Br J Haematol 120 1073ndash1078

30 Lucas TC Silva EED Souza DOD Santos ARD Lara MO 2017 Microstructural

evaluation by confocal and electron microscopy in thrombi developed in central venous catheters

Rev Esc Enferm USP 51 1ndash9

31 Mismetti P Mille D Laporte S Charlet V Buchmuller‐Cordier A Jacquin JP Fournel

P Dutrey‐Dupagne C Decousus H 2003 Low‐molecular‐weight heparin (nadroparin) and

very low doses of warfarin in the prevention of upper extremity thrombosis in cancer patients with

indwelling long‐term central venous catheters a pilot randomized trial Haematologica 88 67ndash

Chapter 7 References

89

73

32 Niers TM Di Nisio M Klerk CP Baarslag HJ Buller HR Biemond BJ 2007

Prevention of catheter‐related venous thrombosis with nadroparin in patients receiving

chemotherapy for hematologic malignancies a randomized placebo‐controlled study J Thromb

Haemost 5 1878ndash1882

33 Scott K C Hansen B D DeFrancesco T C (2009) Coagulation effects of low molecular

weight heparin compared with heparin in dogs considered to be at risk for clinically significant

venous thrombosis J Vet Emerg Crit Care 19 74ndash80

34 Smith RS Zhang Z Bouchard M Li J Lapp HS Brotske GR Lucchino DL Weaver

D Roth LA Coury A Biggerstaff J Sukavaneshvar S Langer R Loose C 2012

Vascular catheters with a nonleaching poly-sulfobetaine surface modification reduce thrombus

formation and microbial attachment Sci Transl Med 4 153

35 Tuumlrk S M Cansu D Uuml Teke H Uuml Kaşifoğlu T Meltem Akay O Bilgin M amp Korkmaz

C 2018 Can we predict thrombotic tendency in rheumatoid arthritis A thromboelastographic

analysis (with ROTEM) Clin Rheumatol 37 2341ndash2349

36 van Rooden CJ Rosendaal FR Barge RM van Oostayen JA van der Meer FJM

Meinders AE Huisman MV 2003 Central venous catheter related thrombosis in haematology

patients and prediction of risk by screening with Doppler-ultrasound Br J Haematol 123 507ndash

512

37 van Rooden CJ Tesselaar ME Osanto S Rosendaal FR Huisman MV 2005 Deep vein

thrombosis associated with central venous catheters - a review J Thromb Haemost 3 2409ndash

2419

38 Verso M Agnelli G Bertoglio S Di Somma FC Paoletti F Ageno W Bazzan M Parise

P Quintavalle R Naglieri E Santoro A Imberti D Soraru M Mosca S 2005 Enoxaparin

for the prevention of venous thromboembolism associated with central vein catheter a double‐

blind placebo‐controlled randomized study in cancer patients J Clin Oncol 23 4057ndash4062

39 Vose J Odunayo A Price JM Daves M Schildt JC Tolbert MK 2019 Comparison of

heparinized saline and 09 sodium chloride for maintaining central venous catheter patency in

healthy dogs PeerJ 7e7072 httpdoi107717peerj7072

40 Wang LH Wei F Jia L Zhi Lu Wang B Dong HY Yu HY Sun GJ YangJ Li B

Chapter 7 References

90

Meng J Zhang RN Bi XQ Chen HY Jiang AL 2015 Fibrin sheath formation and

intimal thickening after catheter placement in dog model role of hemodynamic wall shear stress J

Vasc Access 16 275ndash284

41 Wildgruber M Lueg C Borgmeyer S Karimov I Braun U Kiechle M Meier R Koehler

M Ettl J Berger H 2016 Polyurethane versus silicone catheters for central venous port devices

implanted at the forearm Eur J Cancer 59 113ndash124

42 Whitesell RT Steenburg SD 2014 Imaging findings of acute intravascular thrombus on non-

enhanced computed tomography Emerg Radiol 21 271ndash277

43 Wu X Studer W Skarvan K Seeberger MD 1999 High incidence of intravenous thrombi

after short-term central venous catheterization of the internal jugular vein J Clin Anesth 11

482ndash485

44 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 1998 Composition and

formation of the sleeve enveloping a central venous catheter J Vasc Surg 28 260ndash271

45 Xiang DZ Verbeken EK Van Lommel AT Stas M De Wever I 2001 Sleeve-related

thrombosis a new form of catheter-related thrombosis Thromb Res 104 7ndash14

91

Acknowledgements

First I thank God for every day in my life for His guide and because He was is

and will be always with me and that certainty gives me strength to take a step

forward

My sincere thanks to Prof Dr Reinhard Mischke for the opportunity to perform

my PhD under his supervision for his support and guidance for all that I learned

with him and for his friendship

I also thank to Prof Dr Manfred Kietzmann and PD Dr Mario von Depka

Prondzinski for being part of my supervision group for their time and beneficial

advices and discussion during my PhD

My gratitude to Prof Hewicker-Trautwein and Mrs Rohn for their valuable work

and availability to help me

I thank the Werlhof Institute Medical Care Center Hannover for the valuable

cooperation especially to Mrs Ober-Cassebaum Anne and Franziska for all their

help

I thank all the staff of the Small Animal Clinic of the University of Veterinary

Medicine Hannover (TiHo) My special gratitude to the internal medicine and

critical care surgery neurology ultrasound and anesthesia departments and to

the TiHo students all their cooperation made possible my research

I thank the haematology clinical chemistry and haemostasis laboratory

department of the Small Animal Clinic TiHo especially to Dirk because he was

always available for me thanks for his guide and friendship all his explanations

in the laboratory were of great value but also the conversations about different

topics of the life were of big value for me

92

I thank the Hannover Graduate School for Veterinary Pathobiology

Neuroinfectology and translational Medicine (HGNI) for the well-organized

PhD program and especially thanks to Prof Dr Beatrice Grummer Dr Tina

Selle and Mrs Tania Czeslik for his guide and help during my studies

My sincere thanks to Mrs Maritta Ledwoch for her guidance at the International

office of the TiHo ldquoI remember you since my externship in Germanyrdquo

I also thank the Student Union Hannover for all the help with the accommodation

for me and my family

I want to thank the Universidad Nacional de Costa Rica and therefore to Costa

Rica (my loved country) for the financial support to perform my PhD studies

but also for my professional formation as a veterinarian Special thanks to the

Veterinary Medicine School from Universidad Nacional de Costa Rica To all the

academic and administrative staff for their support and for believing in me

Sincerely thanks to Dr Mauricio Jimenez Soto my mentor my adviser but most

important my friend Thanks for believing in me and for always being by my side

ldquoAgradezco profundamente a mis papaacutes ldquodon Tito (Papillo) y dontildea Eida (Mami)rdquo

Todo lo bueno que hay en mi proviene de Dios y de ellos Gracias a mis hermanos

(as) cuntildeados (as) sobrinos y a toda la familia relacionada a ellos Siempre han

sido un gran apoyo y los amordquo

ldquoGracias a dontildea Alba Fran y David Gracias por toda la ayuda y por estar siempre

pendientesrdquo

I would like to thank all my friends that during a short part or during my whole

life have been with me For all the good time shared really thanks

To Willy thanks ldquogordordquo

93

I sincerely thank to Adriano and Johanna for their help and hospitality Thanks

Adriano for your guidance and support during my beginning at the TiHo for make

me feel in Costa Rica with your friendship for the tourist walks and for the bike

training I missed the coffee times

I want to thank Ina the first person to give me a smile at the start of my PhD

Thanks for your hugs and friendship

I also thank Franz for his help and for being a very nice partner Thanks for help

me always when I requested you

ldquoFinalmente quiero expresar mi gratitud y amor a Karo Faacutetima y Abigaiacutel A Karo

por tu sacrificio y soporte por las palabras de aliento por estar a mi lado en las

buenas pero en especial en las malas A Faacute por tu sacrificio por tus sonrisas por

tus abrazos y besos por bailar con migo siempre por ser mi modelo e inspiracioacuten

A Abi por ser una nueva razoacuten in mi vida para seguir adelante por todo lo que

deseo compartir con tigo por mostrarme junto con Faacutetima la felicidad en una

sonrisardquo

Page 12: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 13: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 14: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 15: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 16: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 17: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 18: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 19: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 20: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 21: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 22: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 23: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 24: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 25: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 26: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 27: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 28: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 29: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 30: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 31: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 32: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 33: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 34: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 35: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 36: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 37: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 38: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 39: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 40: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 41: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 42: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 43: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 44: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 45: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 46: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 47: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 48: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 49: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 50: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 51: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 52: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 53: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 54: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 55: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 56: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 57: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 58: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 59: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 60: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 61: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 62: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 63: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 64: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 65: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 66: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 67: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 68: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 69: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 70: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 71: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 72: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 73: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 74: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 75: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 76: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 77: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 78: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 79: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 80: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 81: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 82: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 83: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 84: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 85: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 86: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 87: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 88: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 89: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 90: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 91: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 92: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 93: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 94: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 95: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 96: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 97: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 98: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 99: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 100: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 101: THESIS DOCTOR OF PHILOSOPHY (PhD)
Page 102: THESIS DOCTOR OF PHILOSOPHY (PhD)

Recommended