Incidence of Central Venous Catheter-Related Bloodstream...

Research ArticleIncidence of Central Venous Catheter-Related BloodstreamInfections Evaluation of Bundle Prevention in Two IntensiveCare Units in Central Brazil

Thais Yoshida1 Ana Elisa Bauer de Camargo Silva2

Luciana Leite Pineli Simotildees1 and Rafael Alves Guimaratildees 3

1Hospital de Doencas Tropicais Dr Anuar Auad Secretaria da Saude do Estado de Goias Goiania Goias Brazil2Faculty of Nursing Federal University of Goias Goiania Goias Brazil3Institute of Tropical Pathology and Public Health Federal University of Goias Goiania Goias Brazil

Correspondence should be addressed to Rafael Alves Guimaraes rafaelalvesg5gmailcom

Received 3 September 2018 Revised 28 December 2018 Accepted 16 January 2019 Published 7 October 2019

Academic Editor Xavier Wittebole

Copyright copy 2019 Thais Yoshida et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Background Central venous catheter-associated bloodstream infections (CVC-BSIs) have been associated with increased lengthof hospital stay mortality and healthcare costs especially in intensive care units (ICUs) The aim of this study was to evaluate theincidence density of CVC-BSIs before and after implementation of the bundle in a hospital of infectious and dermatological diseasesin Central BrazilMethods A retrospective cohort study was conducted in two ICUs (adult and pediatric) between 2012 and 2015Two periods were compared to assess the effect of the intervention in incidence density of CVC-BSIs before and after interventionrelated to the stages before and after the implementation of the bundle respectively Results No significant reduction was observedin the incidence density of CVC-BSIs in adult ICU (incidence rate ratio [IRR] 0754 950 CI 0349 to 1621 p-value = 0469)despite the high bundle application rate in the postintervention period Similarly significant reduction in the incidence densityin pediatric ICU has not been verified after implementation of the bundle (IRR 1148 950 CI 0314 to 4193 p-value = 0834)Conclusion Not significant reduction in the incidence density of CVC-BSIs was observed after bundle implementation in ICUssuggesting the need to review the use of process as well as continuing education for staffs in compliance and correct application ofthe bundle Further studies are needed to evaluate the effect of bundle in the reduction of incidence density of CVC-BSIs in Brazil

1 Introduction

Healthcare-associated infections (HAIs) are a serious publichealth problem and represents significant adverse events inhospitalized patients especially in intensive care units (ICUs)[1ndash3] Central venous catheter-associated bloodstream infec-tions (CVC-BSIs) are among the most serious HAIs andhas been associated with increased length of hospital staymortality and healthcare costs [4]

It is estimated to occur per year in the United States ofAmerica (USA) 80000 cases of CVC-BSIs in ICUs [5] Inaddition in 2013 the National Healthcare Safety Network(NHSN) estimated an incidence of 12 BSIs1000 CVC-daysin medical ICUs in the USA [3] In European countriesthe incidence1000 CVC-days range from 12 in France

to 42 in England [2] In developing countries especiallythose in Latin America the dimension of CVC-BSIs is littleknown However studies conducted in Brazil indicate thatthe incidence density of CVC-BSIs in ICUs patients hasdecreased over the years [6] In 2014 it was recorded anincidence of 51 CVC-BSIs1000 CVC-days in adult ICUs inBrazil rate lower than in 2011 (59 CVC-BSIs1000 CVC-days) [6] In pediatric ICUs the incidence density of CVC-BSIs1000 CVC-days decreased from 73 in 2011 to 58 in 2014[6]

The CVC-BSIs are serious infections and can be pre-vented through appropriate techniques for insertion andmanagement of CVC [4] The application of preventivemeasures in an integratedmanner structured and systemizedhas shown positive results in reducing these infections and

Hindawie Scientific World JournalVolume 2019 Article ID 1025032 8 pageshttpsdoiorg10115520191025032

2 The Scientific World Journal

help to increase patient safety in healthcare services [4 78] In this context are bundles of prevention defined as aset of preventive practices based on evidence that must beperformed collectively The use of these measures allows theevaluation of programs of care and handling of the CVCto identify potential failures andor successes that affect thefinal results Also enables the calculation of indicators thatshow the care practice called process indicators Care impliedin care processes and evaluated through the use of bundlesare essential to improving quality and safety in patient care[9 10]

Several studies have shown decrease in the incidenceof CVC-BSIs after bundle implementation [10ndash15] A meta-analysis that examined the impact of bundles showed asignificant reduction in themedian incidence CVC-BSIs afterapplication of these strategies (641000 CVC-days versus251000 CVC-days p-value lt 0001) [10] The impact ofbundles in reducing the incidence of CVC-BSIs dependson multidisciplinary teamwork effective communicationsetting daily goals easily measurable care continuous profes-sional training and auditing processes [8] Thus despite thepositive results in decreasing CVC-BSIs after implementingreported bundles in several investigations some studies haveshown no reduction in CVC-BSIs rates in places like USATaiwan Spain and Brazil even after systematic applicationof these strategies [10]

In Brazil few studies have investigated the effect ofbundle in reducing CVC-BSIs in ICU and most of thesewere conducted in the most developed region of the country(Southeast) [13 14] Still studies in pediatric ICU are scarcein Brazil Thus this research aimed to evaluate the incidencedensity of CVC-BSIs before and after implementation of thebundle in a hospital of infectious and dermatological diseasesin Central Brazil

2 Materials and Methods

This is a retrospective cohort study that examined the inci-dence density of CVC-BSIs before and after implementationof bundle of prevention The research was conducted inadult and pediatric ICUs of a hospital of infectious anddermatological diseases in Central Brazil from January 2012to December 2015

The hospital provides care elective and emergencymedium and high complexity exclusively to patients of theHealth Unic System (Sistema Unico de Saude in Portuguese)in Brazil This institution has 130 beds distributed in fivesectors two of them in intensive care The adult ICU hasnine beds four of them for individual isolation of patientswhile the pediatric ICU has four hospital beds two intendedfor isolation of patients in special care Overall they have100 occupancy rate in all periods The service profile inboth ICUs is for patients with infectious diseases includingAcquired Immune Deficiency Syndrome (AIDS) tuberculo-sis meningitis dengue among others Patients are mostlyimmunosuppressed with use of antimicrobials for commu-nity infections opportunistic or related to health care (HAIs)

The data of this research were obtained by searchingthe electronic files of Hospital Infection Control Service of

the institution sector responsible for monitoring CVC-BSIsin the ICU The information files were drawn about thebundle of prevention (components of the package and thetotal number of applications-days) number of patients-daysnumber of patients with CVC-days number of episodes ofHAIs number of cases of CVC-BSIs number of deaths fromCVC-BSIs and characteristics of patients with CVC-BSIs(age sex length of stay time of use of CVC diagnosis andisolated microorganisms)

The study included all cases of CVC-BSIs diagnosed inadult and pediatric ICUs during the analysis periodThe casedefinition was based on criteria established by the NationalHealth Surveillance Agency of Brazil which are based onthe NHSN [16] The IPCS were defined based on laboratorycriteria that is diagnosed using blood cultures Thus CVC-BSIs were considered if one of the three criteria was met(i) Criterion 1 patient with one or more positive bloodcultures collected preferentially from peripheral blood withthe pathogen being not related to infection elsewhere (ii)Criterion 2 at least one of the following signs or symptomsfever (gt 38∘C) tremor oliguria (urinary volume lt 20 ml perhour) and hypotension (systolic pressure le 90mmHg) thesesymptoms being unrelated to infection (eg diphtheroidsBacillus sppPropionibacterium spp staphylococci coagulasenegative and micrococci) or (iii) Criterion 3 for children gt28 days and lt 1 yearmdashat least one of the following signs andsymptoms fever (gt 38∘C) hypothermia (lt 36∘C) bradycar-dia or tachycardia (not related to infection elsewhere) andtwo or more blood cultures (in different punctures with amaximum interval of 48 hours) with common skin contam-inants (eg diphtheroids Bacillus spp Propionibacteriumspp staphylococci coagulase negative and micrococci) [16]

The bundle of prevention of CVC-BSIs was systemati-cally implemented in the institution from September 2014(adult ICU) to November 2014 (pediatric ICU) It consistedof actions to be performed in all patients using CVCdefined from the Institutersquos recommendations for Health careImprovement (IHI) [17] This corresponds to an audit tool inthe use of the CVC process that consists of four check itemsin the form of checklists which are actions to be performeddaily in all patients using CVC The bundle includes thefollowing elements

(i) Care in catheter insertion aseptic technique forcatheter insertion (barrier maximum precautions)hand hygiene with chlorhexidine degermante 2patient skin antisepsis with degermante 2 chlorhex-idine followed by alcoholic 05 and record inthe catheter insertion of records with justificationstatement

(ii) Care in the administration of drugs aseptic guns andconnectionswith 70 alcohol beforemedications andexclusive route for infusion of blood derivatives orparenteral nutrition

(iii) Care in maintaining the catheter daily medicalrecords to assess the insertion site clean dry dressingand adhered and dated exchange of catheters insertedin emergency situations and those from other insti-tutions for a maximum of 48 hours exchange for

The Scientific World Journal 3

the infusion system every 96 hours andor in caseof suspicion of pyrogenic shock and blood visiblestuck inside the system record (date and signature)installation in equipos infusion

(iv) Daily assessment for early catheter removal removalof the catheter so that there is no more indicationof use or in the presence of signs and symptomsof catheter-related infection and evidence of medicalrecords of catheter removal with justification state-ment

In this study two periods were compared to assessthe effect of the bundle before intervention (referenceperiod) and after intervention [10]Thepreintervention phaseencompassed the period from January 2012 to August 2014in the pediatric ICU and from January 2012 to October2014 in the adult ICU and represented the period before theapplication of the bundleThus the phase of postinterventioncontemplated November 2014 to December 2015 in the adultICU and from September 2014 to December 2015 in thepediatric ICU and reflected the period after implementationof the bundle The primary outcome was the incidencedensity of CVC-BSIs preintervention phase compared withthe phase of intervention

Data analysis of adult and pediatric ICU was performedseparately Initially the calculations of process indicatorsand their respective confidence intervals of 95 (950CI) for each study period were performed For analysisof the indicator for the bundle the bundle was calculatedapplication rate the following formula

(i) total number of applications-days bundle in theperiod by the total number of patients with CVC-days

For analysis of outcome indicators the following formulaswere used

(i) Incidence density of HAIs number of episodes ofHAIs in the period by the number of patients-days x1000

(ii) CVC utilization () number of patients with CVC-days by the number of patients-days

(iii) Incidence density of CVC-BSIs number of new casesof CVC-BSIs in the period by the number of patientswith CVC-days x 1000

(iv) Lethality with CVC-BSIs number of deaths fromCVC-BSIs in the period by the number of patientswho developed CVC-BSIs

Analyses were performed using the Stata software ver-sion 140 [18] The indicators found in before and afterintervention were compared using the Wald statistic Forthe primary outcome (Incidence density of CVC-BSIs) thebundle effect was analysed using Poisson regression modelswith robust variance [19] The models were adjustment bybaseline severity In addition we included a dummy variablein the model representing intervention ldquo0rdquo in the preinter-vention period and ldquo1rdquo in the postintervention period [20]

The following assumptions of the Poisson regressionweremetfor model validation (ii) Independence of observations itwas verified by comparing errors based on standard modelswith robust errors to determine large differencesmdashfor adultICU the difference of errors between the standard and robustmodels for the two models was 45 and for the pediatricICU it was 34 suggesting independence (ii) distributionfollowing a classical Poisson distribution verifying that theobserved and expected data were similarmdashthis assumptionwas verified by predicting the mean values of the dependentvariable and compared by the t test with the observed values[21]mdashfor the adult ICU the observed and expected valueswere similar (t = 0000 df 47 p value = 1000) as observedin the pediatric ICU model (t = 0000 df 47 p value =1000) and (iii) the mean and variance of the model arethe same or similar as assessed by Pearsonrsquos chi-square ofPearsonrsquos chi-square for the pediatric ICUmodel with a valueof 1067 and for the pediatric ICUmodel a chi-Pearson squarewith a value of 1078 indicating small overdispersion of thedata (values less than 1 indicate subdispersion equal to 1equidispersion and greater than 1 overdispersion) thereforenot causing serious problems in the model In addition toreinforce data suitability to Poisson models the goodness-of-fit Deviance was performed - the result for the adult ICUwas a chi-square of 5370 (df 45 p value = 0175) and forpediatric ICU a chi-square of 28621 (df 45 p value = 0973)indicating that both data fit the Poisson model well Thusthe adjusted incidence rate ratio (IRR) was calculated andits 950 CI for the difference in incidence density of CVC-BSIs among the investigated periods P values lt 005 wereconsidered statistically significant In addition a descriptiveanalysis of the variables related to the patient with CVC-BSIswas carried out (total and ICU)

The studywas approved by the Ethics Committee in Trop-ical Diseases Hospital Research Dr Anuar Auad protocol n0112012 and all ethical and legal principles were consideredunder Resolution n 4662012 of the National Health Council[22]

3 Results

The results of this study are presented by ICU Table 1 showsthe variables and indicators related to the bundle in the adultand paediatric ICU In adult ICU a total of 2282 applications-days of bundle was observed resulting in an application rateof 898 in the postintervention period Furthermore therewas an overall compliance of 856The item with the lowestapplication in the adult ICU was Item 4

In pediatric ICU there was a total of 438 applications-days resulting in bundle application rate of 541 and fullcompliance of 43 Items 1 and 4 presented a low applicationrate in this ICU (Table 1)

In the adult ICU during the study period we observed atotal of 11446 patients-days 9387 CVC-days and an overallutilization rate of 820 CVC A higher CVC usage fee inthe preintervention period compared to the postinterventionperiod (856 vs 736 p value lt 0001) We believe that thechange in the patient profile in adult ICU (with decreasedseverity in the postintervention period) is responsible for


Table 1 Variables and bundle indicators in adult and pediatric ICUsCentral Brazil 2014-2015

Variables and indicatorsbundle Adult ICU Pediatric ICU

Applications-day 2282 438Bundle application rate() 898 541

Item 1 602 486Item 2 998 998Item 3 998 959Item 4 881 57Total Compliance 856 43ICU Intensive Care Unit

the decrease of CVC usage fee Still there was a decreasein the number of cases of HAIs ranging from 270 in thepreintervention phase to 77 in the postintervention periodThe overall incidence density of HAIs was 303 per 1000patient-days (950 CI 273 to 336) A reduction in HAIsdensity per 1000 patients-days between the periods wasfound (p value lt 0001) (Table 2)

It was also observed in the adult ICU that the occurrenceof 32 cases of CVC-BSIS throughout the investigation period(25 in the preintervention period and 7 in the postinter-vention period) In the preintervention period there was anincidence density of CVC-BSIs per 1000 CVC-days of 365(950 CI 247 to 538) Despite the reduction in the absolutenumber of cases of CVC-BSIs after the implementationof the bundle (after intervention) there was no significantreduction in the incidence density (IRR 0754 950 CI0349 to 1621 p value = 0469) (Table 2)

In the pediatric ICU a total of 3791 patients-days and2078 CVC- days were observed resulting in an overall rateof use of CVC of 548 It was found that the CVC usage feesignificantly increased in the analyzed periods (p lt 0001)In this unit there were a total of 51 cases of HAIs (densityof 135 per 1000 patients-days 950 CI 103 to 176) Inaddition the incidence density of HAIs was reduced in thepostintervention period (p value lt 0001) (Table 3)

During the study period there was an overall per 1000CVC-days of 336 (950 CI 163-693) in the pediatric ICUIn the preintervention period there was an incidence densityof CVC-BSIs of 315 by 1000 CVC-days (950 CI 122 to807) There was no significant reduction in the incidencedensity of CVC-BSIs between the periods (IRR 1148 950CI 0314 to 4193 p value = 0834) obtained in fitted Poissonmodel (Table 3)

The characterization of these patients is presented inTable 4 It is noteworthy that most patients who developedCVC-BSIs in adult ICU were male (781) while in the pedi-atric ICU they were female In general the main diagnosisin patients with CVC-BSIs was AIDS (487) followed bytuberculosis (128) The median length of hospital stay andCVC use was 395 days and 105 days respectively

Table 5 shows the characterization of themicroorganismsidentified in culture for diagnosis of CVC-BSIs Most (618)of the causative agents of CVC-BSIs in the institution were

Gram-negative with a predominance of Pseudomonas aerug-inosa (282 ) Gram- positive ones accounted for 308of the isolated microorganisms highlighting Staphylococcusaureus Fungi accounted for 103 of microorganisms themost prevalent non-albicans Candida

4 Discussion

Currently the CVC-BSIs control has been the subject ofnational and international targets [17 23] The reduction ofthese infections is feasible and possible since its occurrence isdirectly related to adoption of safe practice and protocol com-pliance including the systematic use of bundles of prevention[24] However even well established in the practice great arethe challenges and constant is the quest for membership ofprofessional best practice There are few published studieson the evaluation of bundles in reducing CVC-BSIs in LatinAmerica This research adds to the literature about the effectof these strategies in rates CVC-BSIs rates in Brazil Theresults showed that even after implementation of bundlesof prevention significant reduction in incidence density ofCVC-BSIs did not occur in both units assessed

This investigation found an incidence density of CVC-BSIs per 1000 CVC-days of 34 in the adult ICU index belowpercentile 90 of Brazilian ICUs (118 CVC-BSIs per 1000CVC-days) [6] andhigher than theAmerican ICUs (28CVC-BSIs per 1000 CVC-days) [3] Similarly in the pediatric ICUthe incidence of overall incidence density was 336 CVC-BSIsper 1000CVC-days rate belowpercentile 90 in pediatric ICUin Brazil in 2014 (142 CVC-BSIs per 1000 CVC-days) [6] andhigher than that found in pediatric ICU USA (20 CVC-BSIsper 1000 CVC-days) [3]

Regarding the CVC utilization rate in the adult ICUmost patients used the device for most of the length of stayalthough this rate shows lower postintervention periodHow-ever in the pediatric ICU it was found that this ratio signif-icantly increased in the postintervention period Regardlessof these differences there was high use rate of this deviceduring the period analyzed in both unitsThe utilization ratesof CVC in adult and pediatric ICU are above the percentileof 750 of American hospitals evaluated by NHSN [3]probably reflecting the greater severity of patients admittedto the institution under study in relation to those hospitalsthat integrate the NHSN system The CVC utilization ratereveals the degree of exposure to BSIsMesiano andMerchan-Hamann [25] point out that the maintenance of vascularaccess for a long time andwith greater frequency of use resultsin increased infections related to that device

Poisson regression models showed no significant reduc-tion in the incidence density of CVC-BSIs in adult andpediatric ICUs after implementation of the bundle (afterintervention) (p value gt 005) This corroborates with otherstudies conducted in different geographical locations thathave shown no significant reduction of infections afterimplementation of ICU prevention packages [26ndash29] InUSA a randomized clinical trial in ICU of 60 hospitals alsofound no significant reduction after application of preventivebundles (242 to 273 CVC-BSIs per 1000 CVC-days p value= 059) [29] In Taiwan a study conducted in two ICUs found


Table 2 Evaluated variables and indicators in the adult ICU Central Brazil 2012-2015

Variables and indicators All Periods p valuePreinterventiona Postinterventionb

Number of patients-days 11446 7995 3451 -Number of episodes of HAIs 347 270 77 -Incidence density of HAIs (950 CI) 303 (273-336) 338 (300-380) 223 (179-278) lt 0001c

CL utilization () (950 CI)b 820 (813-827) 856 (849-864) 736 (721-751) lt 0001c

Number of new cases of CLABSIs 32 25 7 -Number of deaths from CLABSIs 1 7 3 -Lethality with CVC-BSIs () (950 CI)b 313 (180-496) 280 (143-476) 429 (158-750) 0459c

CVC-days 9387 6847 2540 -Incidence density of CVC-BSIs(950CI)b 340 (241-480) 365 (247-538) 275 (133-547) 0469d

950 CI 950 Confidence Interval a Preintervention period January 2012 to October 2014 b Postintervention period November 2014 to December 2015c Wald Statistics d Wald Statistics obtained in fitted Poisson model CL Central Line CVC Central venous catheter CVC-BSIs Central venous catheter-associated bloodstream infections HAIs Healthcare-associated infections ICU Intensive Care Unit

Table 3 Evaluated variables and indicators in the pediatric ICU Central Brazil 2012-2015


Number of patients-days 3791 2575 1216 -Number of episodes of HAIs 51 35 16 -Incidence density of HAIs (950 CI) 135(102-176) 136 (98-188) 122 (81-213) 0783c


Number of new cases of CLABSIs 7 4 3 -Number of deaths from CLABSIs - - - -Lethality with CVC-BSIs () (950 CI)b - - - -CVC-days 2078 1269 809 -Incidence density of CVC-BSIs(950CI)b 336(163-693) 315(122-807) 370(126-1084) 0834d

950 CI 950 Confidence Interval a Preintervention period January 2012 to August 2014 b Postintervention period September 2014 to December 2015c Wald statistics d Wald Statistics obtained in fitted Poisson model CL Central Line CVC Central venous catheter CVC-BSIs Central venous catheter-associated bloodstream infections HAIs Healthcare-associated infections ICU Intensive Care Unit

a similar rate of BSIs between periods of preinterventionand after systematic implementation of bundles (158 to 106CVC-BSIs per 1000 CVC-days p value = 031) [28] InSpain a study conducted in a university hospital found noreduction in pre- and post-application bundles (55 to 38CVC-BSIs per 1000 CVC-days p value = 049) [27] In Brazila study conducted by Wolf et al [25] in the ICU of SaoPaulo showed that even after bundle implementation nosignificant reduction in incidence density of CVC-BSIs (20to 11 CVC-BSIs per 1000 CVC- days p value = 007) [26]

The studies did not identify reduction in the incidencedensity of CVC-BSIs after bundles application emphasizingthat their use in isolation does not bring decrease in infec-tions requiring a multidisciplinary approach and to considerthe epidemiological profile of the institution and focusedactive leaders in continuous improvement processes [26ndash29]In addition factors such as high rate of use of CVC lowfull compliance bundles application [26] low adherence tobundle and constant vigilance are factors that can decreasethe effectiveness of intervention strategies In fact in this

study total compliance in the pediatric ICU and especially forItem 4 ldquoassessment for early catheter removalrdquo was very lowwhich contributed to the absence of significant reduction

In the present study there was a greater proportion ofGram-negative than Gram-positive microorganisms unlikemost studies conducted in North America that show a higherfrequency ofGram-positivity inCVC-BSIs [30ndash32]Howeverit corroborates with other studies previously published inseveral countries and regions [32ndash35] In fact studies in LatinAmerica such as Brazil have shown a higher prevalence ofGram-negativity in CLABSI compared to American studiesInvestigations such as SCOPE (Surveillance and Control ofPathogens of Epidemiological Importance) [36] and EPICII (Extended Prevalence of Infection in Intensive Care) [32]show this difference These studies discuss the possibility ofa climate influence [32 36] As Brazil is a tropical countryit has a warmer climate than in USA and some studiesshow a higher prevalence of Gram-negative summerspringinfections than in the autumnwinter where there wouldbe more Gram-positive infections [37] Another possibility


Table 4 Characterization of patients with CVC-BSIs Central Brazil 2012-2015

Variables All (n = 39) Adult ICU (n = 32) Pediatric ICU (n = 7)Age (yeas) (Median IIQ) 445 (230) 465 (180) 50 (80)Length of stay (days) (Median IIQ) 395 (380) 420 (440) 330 (300)CL usage time (days) (Median IIQ)c 105 (80) 100 (80) 120 (80)SexMale 27 (692) 25 (781) 2 (286)Female 12 (308) 7 (219) 5 (714)DiagnosesAIDS 19 (487) 190 (594) -Tuberculosis 5 (128) 5 (156) -Viral hepatitis 1 (26) 1 (31) -Leishmaniasis 2 (51) 2 (62) -Dengue 1 (26) - 1 (143)Leprosy 1 (26) 1 (31) -Meningitis 2 (51) 1 (31) 1 (143)Other lung infections 2 (51) - 2 (286)Tetanus 1 (26) 1 (31) -Others 6 (154) 3 (94) 3 (429)AIDS Acquired immunodeficiency syndrome ICU Intensive Care Unit

Table 5 Characterization of the microorganisms identified in culture for diagnosis of CVC-BSIs

Microorganisms All (n = 39) Adult ICU (n = 32) Pediatric ICU (n = 7)Gram-positive 12 (308) 10 (312) 2 (286)Staphylococcus aureus 6 (154) 2 (125) 2 (286)Staphylococcus epidermidis 3 (77) 3 (94) -Staphylococcus coagulase negativos 1 (26) 1 (31) -Enterococcus faecalis 2 (51) 2 (62) -Streptococcus salivarius 1 (26) 1 (31) -Gram-negative 24 (618) 20 (625) 4 (571)Pseudomonas auruginosa 11 (282) 8 (250) 3 (429)Acinetobacter spp 5 (128) 5 (156) -Enterobacter spp 5 (128) 3 (94) 2 (286)Klebsiella pneumoniae 5 (128) 5 (156) -ESBL Klebsiella 1 (26) 1 (31) -Achromobacter xylosoxidans 1 (26) 1 (31) -Fungi 4 (103) 3 (94) 1 (143)Candida albicans 2 (51) 1 (31) 1 (143)Candida nao albicans 3 (77) 3 (94) -ICU Intensive Care Unit

would be a higher proportion of infections secondary to lungand urinary tract infections than in American studies [36]

This study has some limitations First in retrospectiveanalyses there is no possibility of reporting bias with theinability to control confounding variables (lack of infor-mation) Second data as catheter insertion site and otherrisk factors of patients with CVC-BSIs were not subject tocollection by the lack of information in the source data datathat could explain the lack of reduction in incidence densityof CVC-BSIs Thirdly the number of new cases of CLABSIsin the postintervention period was very small in both ICUs(adult and pediatric)This may have diminished the power of

the study to verify statistical differences Other studies withlarger samples and in several hospitals are needed Fourththe analysis period after intervention period was relativelyshort to evaluate the effect of long-term bundle Finally theresults cannot be generalized to all ICUs because they are onlyconsidered units of an institution

5 Conclusion

In conclusion there was no significant reduction in theincidence density CVC-BSIs in adult ICU (p value = 0469)and pediatrics (p = 0834) after implantation of the bundle


of prevention There was an increase of CVC utilizationrate in both ICUs and low total bundle compliance in thepediatric ICU in the postintervention period which indicatebias application of care for CVC-BSIs prevention

The results of this study show a need to reassess the strat-egy as well as continuous training for the application bundleand measurement of compliance with discussion of processindicators with the care team It is the multidisciplinary teamtreating the patient that takes responsibility in this chainof transmission adhering to the protocols of preventionManagers remain with the implicit responsibility to managethe processes train professionals and provide favorableconditions for the implementation of preventive measures inhealth care practice The implications for the managementdeserve attention since joining the bundle of practice isbased on actions that do not require additional costs butthe adoption of preventive measures by professionals sincehealth institutions are already well structured with respect tohuman resources and materials The findings of this studysuggest managers periodically investigate the indicators ofthe CVC application process (bundles) and the occurrenceof CVC-BSIs to identify the root causes and implement newpreventivemeasures and evaluation of bundles of preventionFurther studies are needed to evaluate the effect of bundleprevention of CVC-BSIs long term in Brazil

Data Availability

The data will be made available by the corresponding authorif requested

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

References

[1] M Blake ldquoUpdate Catheter-related bloodstream infection ratesin relation to clinical practice and needleless device typerdquoCanadian Journal of Infection Control vol 23 pp 150ndash160 2008

[2] E Tacconelli G Smith K Hieke A Lafuma and P BastideldquoEpidemiology medical outcomes and costs of catheter-relatedbloodstream infections in intensive care units of four Europeancountries literature- and registry-based estimatesrdquo Journal ofHospital Infection vol 72 no 2 pp 97ndash103 2009

[3] M A Dudeck J R Edwards K Allen-Bridson et al ldquoNationalhealthcare safety network report data summary for 2013device-associated modulerdquo American Journal of Infection Con-trol vol 43 no 3 pp 1ndash19 2015

[4] J Marschall L A Mermel M Fakih L Hadaway A Kallenand N P OrsquoGrady ldquoStrategies to prevent central linendashassociatedbloodstream infections in acute care hospitals 2014 updaterdquoInfection Control and Hospital Epidemiology vol 35 pp 753ndash771 2014

[5] N P OrsquoGrady M Alexander E P Dellinger et al ldquoGuidelinesfor the prevention of intravascular catheterndashrelated infectionsrdquoAmerican Journal of Infection Control vol 39 no 11 pp S1ndashS342011

[6] Ministerio da Saude ldquoAgencia Nacional de Vigilancia Sanitaria(ANVISA)rdquo Boletim Informativo Seguranca do Paciente eQualidade em Servicos de Saude n11990011 Ano VI Avaliacao dosindicadores nacionais de infeccao relacionada a assistencia anode 2014 e relatorio de progresso 2015 37

[7] P Pronovost D Needham S Berenholtz D Sinopoli HChu S Cosgrove et al ldquoAn intervention to decrease catheter-related bloodstream infections in the ICUrdquo England Journal ofMedicine vol 355 pp 2725ndash2732 2006

[8] E Pina E Ferreira A Marques and B Matos ldquoInfeccoesassociadas aos cuidados de saude e seguranca do doenterdquo RevPort Saude Publica pp 27ndash39 2010

[9] J Schulman R Stricof T P Stevens et al ldquoStatewide NICUcentral-line-associated bloodstream infection rates decline afterbundles and checklistsrdquo Pediatrics vol 127 no 3 pp 436ndash4442011

[10] E Ista B van der Hoven R F Kornelisse et al ldquoEffectivenessof insertion and maintenance bundles to prevent central-line-associated bloodstream infections in critically ill patients ofall ages a systematic review and meta-analysisrdquo The LancetInfectious Diseases vol 16 no 6 pp 724ndash734 2016

[11] E Y Furuya A Dick E N Perencevich et al ldquoCentral linebundle implementation in us intensive care units and impacton bloodstream infectionsrdquo PLoS ONE pp 1ndash6 2011

[12] V D Rosenthal D G Maki C Rodrigues et al ldquoImpact ofinternational nosocomial infection control consortium (inicc)strategy on central linendashassociated bloodstream infection ratesin the intensive care units of 15 developing countriesrdquo InfectionControl andHospital Epidemiology vol 31 no 12 pp 1264ndash12722010

[13] R D Lobo A S Levin M S Oliveira et al ldquoEvaluation ofinterventions to reduce catheter-associated bloodstream infec-tion Continuous tailored education versus one basic lecturerdquoAmerican Journal of Infection Control vol 38 no 6 pp 440ndash448 2010

[14] M G Menegueti K M Ardison F Bellissimo-Rodrigues et alldquoThe impact of implementation of bundle to reduce catheter-related bloodstream infection ratesrdquo Journal of ClinicalMedicineResearch vol 7 no 11 pp 857ndash861 2015

[15] C F Padilla Fortunatti ldquoImpact of two bundles on centralcatheter-related bloodstream infection in critically ill patientsrdquoRevista Latino-Americana de Enfermagem vol 25 2017

[16] Agencia Nacional de Vigilancia Sanitaria ldquoCriteriosDiagnosticos de Infeccoes Relacionadas a Assistencia a SauderdquoIn Agencia Nacional de Vigilancia Sanitaria Brasılia Brasil2017 httpportalanvisagovbrdocuments338523507912Caderno+2+-+CritC3A9rios+DiagnC3B3sticos+de+InfecC3A7C3A3o+Relacionada+C3A0+AssistC3AAncia+C3A0+SaC3BAde7485b45a-074f-4b34-8868-61f1e5724501

[17] R Resar F A Griffin C Haraden and T W Nolan ldquoUsingCare Bundles to Improve Health Care Qualityrdquo in IHI Innova-tion Series white paper Institute for Healthcare ImprovementCambridge MA USA 2012 httpwwwihiorg

[18] StataCorp ldquoStata Statistical Software Release 14rdquo College Sta-tion TX StataCorp LP 2015

[19] K F Sellers S Borle and G Shmueli ldquoThe COM-Poissonmodel for count data a survey of methods and applicationsrdquoApplied Stochastic Models in Business and Industry vol 28 no2 pp 104ndash116 2012

[20] J L Bernal S Cummins and A Gasparrini ldquoInterruptedtime series regression for the evaluation of public health


interventions a tutorialrdquo International Journal of Epidemiologyvol 46 no 1 pp 348ndash355 2017

[21] T K Kim ldquoT-test as a Parametric Statisticrdquo Korean Journal ofAnesthesiology vol 68 no 6 pp 540ndash546 2015

[22] Ministerio da SaudemdashConselhoNacional de Saude ldquoResolucaon∘466 de 12 de dezembro de 2012 [Internet]rdquo Brasılia Brasil2012 httpbvsmssaudegovbrbvssaudelegiscns2013res0466 12 12 2012html

[23] BrasilmdashAgencia Nacional de Vigilancia Sanitaria ldquoProgramaNacional de Prevencao e Controle de Gerencia Geral deTecnologia em Servicos de Sauderdquo Brasılia Brasil 2016

[24] P J Pronovost C A Goeschel and E Colantuoni ldquoSustainingreductions in catheter related bloodstream infections in Michi-gan intensive care units observational studyrdquo British MedicalJournal vol 340 p 462 2010

[25] E R A B Mesiano and E Merchan-Hamann ldquoBloodstreaminfections among patients using central venous catheters inintensive care unitsrdquo Revista Latino-Americana de Enfermagemvol 15 no 3 pp 453ndash459 2007

[26] R D Lobo A S Levin L M Brasileiro Gomes et al ldquoImpactof an educational program and policy changes on decreasingcatheter-associated bloodstream infections in a medical inten-sive care unit in Brazilrdquo American Journal of Infection Controlvol 33 no 2 pp 83ndash87 2005

[27] M Espiau M Pujol M Campins-Martı et al ldquoIncidencia debacteriemia asociada a cateter venoso central en una unidad decuidados intensivosrdquo Anales de Pediatrıa vol 75 no 3 pp 188ndash193 2011

[28] P P Wu C E Liu C Y Chang et al ldquoDecreasing catheter-related bloodstream infections in the intensive care unit Inter-ventions in a medical center in central Taiwanrdquo Journal ofMicrobiology Immunology and Infection vol 45 no 5 pp 370ndash376 2012

[29] T Speroff E W Ely R Greevy et al ldquoQuality improvementprojects targeting health care-associated infections Comparingvirtual collaborative and toolkit approachesrdquo Journal of HospitalMedicine vol 6 no 5 pp 271ndash278 2011

[30] C OrsquoNeil K Ball H Wood et al ldquoA central line care main-tenance bundle for the prevention of central linendashassociatedbloodstream infection in nonndashintensive care unit settingsrdquoInfection Control amp Hospital Epidemiology vol 37 no 06 pp692ndash698 2016

[31] P V Sreeramoju J Tolentino S Garcia-Houchins and S GWeber ldquoPredictive Factors for the Development of CentralLinendashAssociated Bloodstream Infection Due to Gram-NegativeBacteria in Intensive Care Unit Patients After Surgeryrdquo InfectionControl amp Hospital Epidemiology vol 29 pp 51ndash56 2008

[32] J L Vincent J Rello J Marshall et al ldquoInternational studyof the prevalence and outcomes of infection in intensive careunitsrdquo Journal of the AmericanMedical Association vol 302 no21 pp 2323ndash2329 2009

[33] K Lin A Cheng Y Chang et al ldquoCentral line-associatedbloodstream infections among critically-ill patients in the era ofbundle carerdquo Journal ofMicrobiology Immunology and Infectionvol 50 no 3 pp 339ndash348 2017

[34] E Braun K Hussein Y Geffen G Rabino Y Bar-Lavie andMPaul ldquoPredominance of Gram-negative bacilli among patientswith catheter-related bloodstream infectionsrdquo Clinical Micro-biology and Infection European Society of Clinical InfectiousDiseases vol 20 no 10 pp O627ndashO629 2014

[35] M L Rinke A M Milstone A R Chen et al ldquoAmbulatorypediatric oncology CLABSIs Epidemiology and risk factorsrdquoPediatric Blood amp Cancer vol 60 no 11 pp 1882ndash1889 2013

[36] A R Marra L F A Camargo and A C C Pignatari ldquoNosoco-mial bloodstream infections in Brazilian hospitals analysis of2563 cases from a prospective nationwide surveillance studyrdquoJournal of Clinical Microbiology vol 49 no 5 pp 1866ndash18712011

[37] E N Perencevich J C McGregor M Shardell et al ldquoSummerpeaks in the incidences of gram-negative bacterial infectionamong hospitalized patientsrdquo Infection Control amp HospitalEpidemiology vol 29 no 12 pp 1124ndash1131 2008

Stem Cells International

Hindawiwwwhindawicom Volume 2018


MEDIATORSINFLAMMATION

of

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Immunology ResearchHindawiwwwhindawicom Volume 2018

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine


Behavioural Neurology

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary andAlternative Medicine

Volume 2018Hindawiwwwhindawicom

Submit your manuscripts atwwwhindawicom


help to increase patient safety in healthcare services [4 78] In this context are bundles of prevention defined as aset of preventive practices based on evidence that must beperformed collectively The use of these measures allows theevaluation of programs of care and handling of the CVCto identify potential failures andor successes that affect thefinal results Also enables the calculation of indicators thatshow the care practice called process indicators Care impliedin care processes and evaluated through the use of bundlesare essential to improving quality and safety in patient care[9 10]

Several studies have shown decrease in the incidenceof CVC-BSIs after bundle implementation [10ndash15] A meta-analysis that examined the impact of bundles showed asignificant reduction in themedian incidence CVC-BSIs afterapplication of these strategies (641000 CVC-days versus251000 CVC-days p-value lt 0001) [10] The impact ofbundles in reducing the incidence of CVC-BSIs dependson multidisciplinary teamwork effective communicationsetting daily goals easily measurable care continuous profes-sional training and auditing processes [8] Thus despite thepositive results in decreasing CVC-BSIs after implementingreported bundles in several investigations some studies haveshown no reduction in CVC-BSIs rates in places like USATaiwan Spain and Brazil even after systematic applicationof these strategies [10]

In Brazil few studies have investigated the effect ofbundle in reducing CVC-BSIs in ICU and most of thesewere conducted in the most developed region of the country(Southeast) [13 14] Still studies in pediatric ICU are scarcein Brazil Thus this research aimed to evaluate the incidencedensity of CVC-BSIs before and after implementation of thebundle in a hospital of infectious and dermatological diseasesin Central Brazil

2 Materials and Methods

This is a retrospective cohort study that examined the inci-dence density of CVC-BSIs before and after implementationof bundle of prevention The research was conducted inadult and pediatric ICUs of a hospital of infectious anddermatological diseases in Central Brazil from January 2012to December 2015

The hospital provides care elective and emergencymedium and high complexity exclusively to patients of theHealth Unic System (Sistema Unico de Saude in Portuguese)in Brazil This institution has 130 beds distributed in fivesectors two of them in intensive care The adult ICU hasnine beds four of them for individual isolation of patientswhile the pediatric ICU has four hospital beds two intendedfor isolation of patients in special care Overall they have100 occupancy rate in all periods The service profile inboth ICUs is for patients with infectious diseases includingAcquired Immune Deficiency Syndrome (AIDS) tuberculo-sis meningitis dengue among others Patients are mostlyimmunosuppressed with use of antimicrobials for commu-nity infections opportunistic or related to health care (HAIs)

The data of this research were obtained by searchingthe electronic files of Hospital Infection Control Service of

the institution sector responsible for monitoring CVC-BSIsin the ICU The information files were drawn about thebundle of prevention (components of the package and thetotal number of applications-days) number of patients-daysnumber of patients with CVC-days number of episodes ofHAIs number of cases of CVC-BSIs number of deaths fromCVC-BSIs and characteristics of patients with CVC-BSIs(age sex length of stay time of use of CVC diagnosis andisolated microorganisms)

The study included all cases of CVC-BSIs diagnosed inadult and pediatric ICUs during the analysis periodThe casedefinition was based on criteria established by the NationalHealth Surveillance Agency of Brazil which are based onthe NHSN [16] The IPCS were defined based on laboratorycriteria that is diagnosed using blood cultures Thus CVC-BSIs were considered if one of the three criteria was met(i) Criterion 1 patient with one or more positive bloodcultures collected preferentially from peripheral blood withthe pathogen being not related to infection elsewhere (ii)Criterion 2 at least one of the following signs or symptomsfever (gt 38∘C) tremor oliguria (urinary volume lt 20 ml perhour) and hypotension (systolic pressure le 90mmHg) thesesymptoms being unrelated to infection (eg diphtheroidsBacillus sppPropionibacterium spp staphylococci coagulasenegative and micrococci) or (iii) Criterion 3 for children gt28 days and lt 1 yearmdashat least one of the following signs andsymptoms fever (gt 38∘C) hypothermia (lt 36∘C) bradycar-dia or tachycardia (not related to infection elsewhere) andtwo or more blood cultures (in different punctures with amaximum interval of 48 hours) with common skin contam-inants (eg diphtheroids Bacillus spp Propionibacteriumspp staphylococci coagulase negative and micrococci) [16]

The bundle of prevention of CVC-BSIs was systemati-cally implemented in the institution from September 2014(adult ICU) to November 2014 (pediatric ICU) It consistedof actions to be performed in all patients using CVCdefined from the Institutersquos recommendations for Health careImprovement (IHI) [17] This corresponds to an audit tool inthe use of the CVC process that consists of four check itemsin the form of checklists which are actions to be performeddaily in all patients using CVC The bundle includes thefollowing elements

(i) Care in catheter insertion aseptic technique forcatheter insertion (barrier maximum precautions)hand hygiene with chlorhexidine degermante 2patient skin antisepsis with degermante 2 chlorhex-idine followed by alcoholic 05 and record inthe catheter insertion of records with justificationstatement

(ii) Care in the administration of drugs aseptic guns andconnectionswith 70 alcohol beforemedications andexclusive route for infusion of blood derivatives orparenteral nutrition

(iii) Care in maintaining the catheter daily medicalrecords to assess the insertion site clean dry dressingand adhered and dated exchange of catheters insertedin emergency situations and those from other insti-tutions for a maximum of 48 hours exchange for















3 Results
















4 Discussion































5 Conclusion





Data Availability




References












































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of

































3 Results
















4 Discussion































5 Conclusion





Data Availability




References












































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of































4 Discussion































5 Conclusion





Data Availability




References












































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of













































5 Conclusion





Data Availability




References












































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of



























5 Conclusion





Data Availability




References












































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of






















Data Availability




References












































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of










































of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of























of




Disease Markers



OncologyJournal of





PPAR Research



Volume 2018


Journal of

ObesityJournal of



















Date post:	28-Dec-2019
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Incidence of Central Venous Catheter-Related Bloodstream...

Documents