Journal of Forensic Radiology and Imaging 5 (2016) 8–14

Contents lists available at ScienceDirect

Journal of Forensic Radiology and Imaging

http://d2212-47

n CorrE-m

patriciapatrick.

1 TewebsiteToulous(Switze

journal homepage: www.elsevier.com/locate/jofri

Very economical immersion pump feasibility for postmortem CTangiography

Wolf Schweitzer n, Patricia Mildred Flach, Michael Thali, Patrick Laberke, Dominic GaschoDepartment of Forensic Medicine and Imaging, Zurich Institute of Forensic Medicine, University of Zurich, Switzerland

a r t i c l e i n f o

Article history:Received 15 September 2015Received in revised form30 October 2015Accepted 17 November 2015Available online 30 November 2015

Keywords:Postmortem CT angiographyPMCTALow costImmersion pumpVirtopsyForensic pathology

x.doi.org/10.1016/j.jofri.2015.11.00980/& 2015 Elsevier Ltd. All rights reserved.

esponding author.ail addresses: wolf.schweitzer@irm.uzh.ch (W.flach@irm.uzh.ch (P.M. Flach), michael.thali@laberke@irm.uzh.ch (P. Laberke), dominic.gascchnical Working Group Postmortem Angiogralists Lausanne (Switzerland), Hamburg (Ge

e (France), Foggia (Italy), Krakow (Poland), Lerland) and Leipzig (Germany) (in the order cit

a b s t r a c t

Background: Current postmortem CT angiography (PMCTA) methods lack machines that are low pricedand openly accessible. The aim of this technical study was to evaluate the usage of a very economicalmobile immersion pump in comparison to a standard roller pump of a heart-lung machine for PMCTA.

Methods: A Barwig Model 0444 immersion pump and a roller pump mounted inside a conventionalclinically certified Stöckert Shiley heart-lung machine were used. Contrast agent (water soluble contrastagent dissolved in PEG 200 at 1:20) and PMCTA parameters were same across both methods. Two PMCTAfull body scans of a comparable postmortem interval are compared.

Results: Calibrating voltage against flow rate yields a linear relationship for both water and PEG basedcontrast solution. Imaging yields similar results for both pump methods.

Conclusions: A more widespread and systematic implementation of PMCTA needs the premise ofaffordable equipment for facilities with tight budgets. Evaluation of an economical pump system is a stepinto that direction. This study shows that for specific postmortem application, a very low priced im-mersion pump delivers equal results to a clinically certified costly roller pump.

& 2015 Elsevier Ltd. All rights reserved.

1. Background

Ten years ago, postmortem computed tomography (PMCT) inforensic medicine gained the addition of postmortem computedtomography angiography (PMCTA), initiated by a ground breakingpaper by Jackowski et al. [1].

Postmortem angiography is not obvious due to arrested bloodcirculation and subsequent postmortem changes. First works onPMCTA focused on contrast agents [1,2] but not so much on themethod used for contrast medium circulation.

Since then, postmortem angiography has become increasinglypopular both for research and case investigation. Grabherr et al. in-troduced diesel oil based contrast agent solutions into PMCTA in 2006[2]. Subsequently, Grabherr co-authored a number of postmortemangiography patents (2009 [3], 2010 [4], 2012 [5] and 2013 [6]). Theoil-based contrast agent branded Angiofil® and a dedicated pumptermed Virtangio® [7] were then made commercially availablethrough collaborative efforts with Fumedica®, while aworking group1

. Schweitzer),irm.uzh.ch (M. Thali),ho@irm.uzh.ch (D. Gascho).phy Methods (TWGPAM): itsrmany), Munich (Germany),icester (Great Britain), Baseled [8]).

was spearheaded, employing that particular choice of devicesand materials as a de facto-standard also through a series ofpublications [9–20]. This is an impressive coordinative aspectwhile one has to acknowledge that academic authors that holdpatent applications or commercial licenses are more productivegenerally [21].

With that, the current leading commercial solution for post-mortem angiography has to be regarded to be the Virtangio®

machine that is sold in conjunction with the oil-based patentedcontrast agent Angiofil® [3–6,22]. The Virtangio® machine costs upto around 80 000 CHF, and a single postmortem scan costs anadded amount of around 500 CHF (consumables including contrastagent and tubes). Such costs are prohibitively high for many for-ensic pathologists. The Virtangio® machine offers relatively con-venient handling and controls, and at its core, it contains a rollerpump.

The choice of a particular pump system is a technical issue andtherefore warrants technical considerations. Why PMCTA startedout specifically using roller pumps is not clear [1,2,23]. Heart-lungmachines [24] contain relatively costly peristaltic or roller pumps[25] also to preserve the integrity of corpuscular blood compo-nents [26]. Roller pumps excel in transporting fluids with non-Newtonian characteristics (i.e., non-constant viscosity such asblood, more generally containing emulsions, suspensions, slurries;illustrative examples are ketchup or paint) [27,28]. Postmortemangiography requires homogenous watery or oily solutions to be

W. Schweitzer et al. / Journal of Forensic Radiology and Imaging 5 (2016) 8–14 9

quite simply pumped from a container into the body of a deceasedperson. More cost-effective and therefore ideal PMCTA pumps maybetter be inspired by typical oil or water pumps, so we chose asimple centrifugal pump with a small paddle wheel.

Administration of contrast agent solution in PMCTA seems torequire a specific flowrate. Ross et al. published a comparison ofcontrast agents [23], where he described perfusion volumes ofabout 2000 mL for head, neck, thorax, abdomen and the upperextremities and about 400 mL for the lower extremities; he used adecommissioned pressure controlled heart-lung machine (HL20,Maquet) with a perfusion pressure of around 80 mmHg. Morganet al. ([29], following [11]) reported an Angiofil® flow rate of800 mL/Min and a volume of 1200 mL (arterial) and 1800 mL(venous) before adding 500 mL @ 200 mL/min for the “dynamic”phase. Contrast agent solutions such as polyethylene glycol (PEG)substrates exhibit concentration dependent viscosity [30]. Robertset al. [31] report a flow rate of 10 mL/s (600 mL/min). So fromliterature, one may assume that approximate flow rates some-where between 200 mL/min and 800 mL/min should result inacceptable organ perfusion.

Based on the expectation that any centrifugal pump musteringa flow rate of watery or oily fluids as low as 1 L/min would sufficefor sufficient vascular filling, we invested into a relatively powerfulpump ceiling at a 10 L/min flow rate. As this first feasibility testresult exceeded our expectations, this technical note thusdescribes the first use of this very affordable immersion pump(Fig. 1), effectively achieving immersion pump PMCTA (IP PMCTA,in full nomenclature accordance [32]) that results in data that isequivalent to PMCTA performed with a standard heart-lung-machine (HLM PMCTA) (Figs. 2–4). The aim of this technicalstudy was to evaluate PMCTA using a very economical mobileimmersion pump in comparison to a standard roller pump of aheart-lung machine. We show a side by side comparison of firstresults.

pump

*acces

sf

m

o1cmo c

p

powersupply

Fig. 1. Immersion pump set up for usage in postmortem CT angiography. The pump (A, B:outflow and conical tube connector (o), cable (c), motor (m), shaft (s) and paddle wheel ((shown in D and E). The pump draws electricity from a voltage controllable power supplyconnects the pump that is immersed inside the contrast agent container to the femora

2. Methods

2.1. PMCTA: vascular access, contrast medium, PEG

Water soluble contrast agent and polyethylene glycol (PEG)were used as this has been found to provide a superior quality inpostmortem angiographic visualization compared to an oil basedapproach [23]. Vessels were catheterized at the level of sub-inguinal femoral vessels by an autopsy technician (Cannula PU 14F100 mm, Sorin Group International SA, Lausanne, Switzerland; 1/4–1/4 straight connector). A water-soluble contrast medium (Op-tiray 300, Guerbet, Paris, France) was diluted in PEG (PEG 200,Schaerer and Schlaepfer, Rothrist, Switzerland) with a ratio ofabout 1:20 to achieve a mean density of around 350 Hounsfieldunits (HU) (calibrated via CT scan of the contrast media mixture)and injected in the arterial vascular system only, using two dif-ferent pumps: an immersion pump and a heart-lung machine.

2.2. Specification of immersion pump

We used an immersion pump (12–24 V, Model 0444 max.600 L/h ∼10 L/min, 145 g, 30 �80 mm, 0.9–1.7 A, 40 W, BarwigWasserversorgung, D-34385 Bad Karlshafen; cost around 16–20CHF) that we controlled by a variable DC power supply (0–30 V, 0–5 A, power consumption 230 V/AC710% @ 50 Hz, weight 4.95 kg,þ5 to þ40 °C operating temperature, Basetech BT-305, Basetech,Conrad Electronic SE, D-92240 Hirschau; cost 130 CHF).

The pump is constructed to be safely operated in a submergedposition. It is approved for pumping drinking water. Its motor issealed, it features service free lifetime lubrication also for main-tenance free dry runs. Due to the build of the pump, maximal cycletimes of 30 min are permissible but for continuous operation inexcess of 30 min, voltages not higher than 9 V DC should be ap-plied. Approved operating temperatures are between 0 and 50 °C.

The pump is sold and delivered with an electric cable sealedinside and exiting the pump. The outlet contains a conical

sc

i

ump

powersupply

CT of the pump, view on front facing longitudinal virtual cut surface with inflow (i),f)) is immersed in the contrast agent solution that we placed into a plastic container(control panel in C; whole device in D, E). Display of the entire setup (E), a PVC tubel region (*access) of the deceased positioned on the table of the postmortem CT.

IP PMCTA HLM PMCTAa’ b’

c’

d’ e’

f’

f’

HLM PMCTAg’

h’

i’

IP PMCTAg

h

i

a b

c

d e

f

f

1

2

3

4

5

6

j j’

Fig. 2. Comparing the new very affordable immersion pump PMCTA (IP PMCTA) and conventional heart-lung machine PMCTA (HLM PMCTA), we find comparable quality inthe cheaply afforded scans in an axial side by side image comparison. Both carotid arteries (a/b and ′ ′a /b ) are equally well shown as the basilar artery ( ′c/c ). The Circle ofWillis exhibits equally distinct filling of the arteries ( ′d/d , ′e/e ) despite the presence of a vessel leak in the IP PMCTA case (ruptured aneurysm, subarachnoid hemorrhagealready present at native PMCT prior to PMCTA). Subcutaneous arteries at the temple similar ( ′f/f ). The left circumflex coronary artery ( ′g/g ) and the left anterior descendingcoronary artery ( ′h/h ) are depicted clearly in both techniques. Side by side of pelvic arteries (external ( ′i/i ) and internal ( ′j/j ) iliac arteries) with similar delineation.

W. Schweitzer et al. / Journal of Forensic Radiology and Imaging 5 (2016) 8–1410

extension that we used to heat shrink a polyvinyl-chloride (PVC)tube onto it.

2.2.1. Flow rate and voltage calibrationPositive displacement pumps such as roller pumps are char-

acterized by a flow rate that is fairly unresponsive to variation offluid viscosity; conversely, centrifugal pumps such as the very low

IP PMCTA

Fig. 3. Immersion pump PMCTA (IP PMCTA) contains similarly clearly reconstructed temperformed with a conventional heart lung machine (HLM PMCTA).

cost model we employed exhibit lower flow rates with higherviscosity with the same voltage. For that we employed a scaledcontainer and a manually operated stop watch. A total of 35measurements were obtained for both (less viscous) water and(more viscous) contrast agent PEG solution. Linear regression wasperformed using statistics software (JMP 11, SAS, Cary NC, USA)(Fig. 5).

HLM PMCTA

poral and occipital arteries (arrows) in this VRT side-by-side display as the PMCTA

HLM PMCTAIP PMCTA

a’

b’

c’ d’

f’

g’

e’a

b

c d

f

g

e

Fig. 4. Vascular imaging using volume rendering of whole body reconstructions (magnified and omitting most of the legs) yield comparable results for both the immersionpump (IP PMCTA) and the conventional heart lung machine (HLM PMCTA) systems. The arm arteries ( ′i/a ) are displayed into the periphery ( ′b/b ); carotid arteries ( ′c/c , ′d/d )as well as abdominal arteries ( ′e/e ), kidneys ( ′f/f ) and lower extremities ( ′g/g ) are featured equally well.

W. Schweitzer et al. / Journal of Forensic Radiology and Imaging 5 (2016) 8–14 11

2.2.2. HandlingCleaning is performed by immersion into detergent and sub-

sequent pumping (Fig. 6). The pump is very small and light, whichreflects on the overall handling (details in Table 1). The immersionpump including extras can easily be fit into a suitcase or bag, andthus is mobile. One-handed manipulation risks to spill watery PEGsolution onto the floor, which may be hazardous due to

0

2

4

6

8

10

flow

rate

L/m

in

5 10 15 20 25

volt

PEG

water

Fig. 5. Voltage-dependent flow-rates for both water (diagram: þ) and more viscous Pmethod section) (linear regression; water =R 0.962 , PEG =R 0.992 ). Linear regressionlogarithmic flowrate scale) shows a linear relation between pump voltage and both wabetween 0.2 and 0.8 L/min, PEG is best pumped with pump voltages between 5 and 10 VThis coincides with this particular pump model's permanent usage's comfort zone of 5

slipperiness, as a first practical test of this tiny pump revealed; bi-manual manipulation of tubing and pump is thus strongly advised.

2.3. Specification of heart-lung machine

As comparison, we used a standard Stöckert Shiley heart-lungmachine with four multiflow roller pumps, one of which was used

0.1

0.2

0.3

0.4

0.50.60.70.81

2

3

4

5678

10

flow

rate

L/m

in

5 10 15 20 25

volt

PEG

water

EG (diagram: � ) using the very low cost immersion pump (model 0444, Barwig,(same data in both diagrams – left diagram: linear flowrate scale; right diagram:tery (higher) and viscous (lower) fluid's flow rates. For PMCTA targeted flow rates, which is easily identified on the logarithmic voltage-flow rate plot (right diagram).–9 V to avoid overheating under prolonged usage.

Fig. 6. Cleaning the very low cost immersion pump is achieved by immersing it indetergent containing water and pumping that fluid, thus both cleaning the pumpfrom its outside and inside while transporting the water from the fluid container(orange portal) to the sink (blue portal). (For interpretation of the references tocolor in this figure caption, the reader is referred to the web version of this paper.)

Table 1Device feature comparison.

Features Barwig immersion pump Stöckert Shiley roller pump

Cost of hardware ∼200 CHF, brand new, incl.power supply, plugs, tubing,etc.

∼1500 CHF, used/refurbished

Weight 0.145 kg þ5 kg power supply 25 kgSize Can be transported with a

bag, storage can be shelfspace

big, transport requires van,storage requires room space

Max. flow rate 10 L/min 10 L/min

W. Schweitzer et al. / Journal of Forensic Radiology and Imaging 5 (2016) 8–1412

for postmortem angiography (max. 10 L/min, 4 �80 W, 25 kg, 29�19�47 cm). Such a second hand device currently has an esti-mated street price of around 1500 CHF [33].

2.3.1. HandlingWhile we used a typical clinically certified decommissioned

heart-lung machine that contains a sturdy build, its 25 kg requirethat the pump is mounted on a wheely frame. Considerable sto-rage space is required.

2.4. PMCT and PMCTA scanning details

PMCT and PMCTA were both obtained on a dual-energy CTscanner (Somatom Flash Definition, Siemens, Forchheim, Ger-many). PMCT scan parameters as published [34]: tube voltage100 kVp, automatic dose modulation software (CARE dose 4D,Siemens, Forchheim, Germany), slice thickness 1.0 mm, increment0.6 mm, image reconstruction with both soft tissue and bonekernels. Image read-out and volume rendering technique (VRT)imagery (Figs. 3 and 4) were obtained with standard workstationsoftware using factory preset settings (Syngo®.via, Version VA30A,Siemens, Medical Solutions, Erlangen, Germany) on standardcommercial hardware.

2.5. Cases

Examined with IP PMCTA: 47 year old male, death scene ex-amination around 2–5 h after death, cooled storage of the body forabout 2 1

2days before PMCT, PMCTA and autopsy. Cause of death

was a ruptured aneurysm as cause of a subarachnoid hemorrhage.

Examined with HLM PMCTA: 52 year old male, death sceneexamination around 1 day after death, cooled storage of the bodyfor about 2 1

2days before PMCT, PMCTA and autopsy. Cause of

death was attributed to an aortic dissection (not ruptured).Both cases were scanned as part of our standard procedure

when examining cause and manner of death. Examinations wereperformed under an agreement with the prosecutor's office, andPMCTA indications were part of the routine evaluation for bothcases. Research publication was authorized by the ethics com-mittee of the Canton of Zürich, Switzerland (KEK number: 30-2015).

3. Results

3.1. IP PMCTA

This immersion pump's flow rate is determined by both fluidviscosity and voltage. A mostly linear flow rate to voltage relationresulted from calibration (Fig. 5). Flow rate was effectively con-trolled with the voltage knob of the power supply.

There was no pressure measurement in place, but visual checksof the femoral access revealed no leakage, and continuous reduc-tion of the container content was observed during contrast agentadministration. Firstly the PEG contrast agent solution into thebody for half a minute at a pump voltage of 3 V, then turned up to5 V and left it there for about 5 min (yielding an approximate flowrate of 250 mL/min). When the container showed that around2500 mL had been injected into the arterial system, the pump wasstopped.

3.2. HLM PMCTA

Standard protocol using a flow rate of around 0.8 L/min using atotal of about 2500 mL of PEG contrast agent solution was applied.

3.3. Imaging

The methods IP PMCTA and HLM PMCTA yielded similar resultswhen comparing images (Figs. 2–4).

IP PMCTA revealed a leak in the vicinity of a subarachnoidhemorrhage that was already there at PMCT prior to PMCTA. Au-topsy showed a ruptured aneurysm as cause of the subarachnoidhemorrhage, thus explaining the contrast agent leak. Thus, nounexpected extravasation occurred.

4. Discussion

We showed that for postmortem CT angiography (PMCTA), re-sults equivalent to what is achieved with a peristaltic roller pumpare generated using a very low cost immersion pump. The reasonfor that is technical: the particular pumped liquid in this instanceis adequately transported using a properly used centrifugal pump.

Generally, PMCTA is interesting for forensic pathologists. Cur-rent PMCTA applications with particularly useful results appearedto be trauma cases, natural death [35], decomposed bodies [36] aswell as selected vascular access (SVA) in context of PMCTA whichone might consequentially term SVAPMCTA [31,37] and now alsofor PMMRI (postmortem magnetic resonance imaging) andSVAPMMRI [38] (following terminology agreements [32]).

4.1. Cost

IP PMCTA is calculated for hardware costs of less than a percentof a commercial top-of-the-line solution. Also, standard injectors

W. Schweitzer et al. / Journal of Forensic Radiology and Imaging 5 (2016) 8–14 13

have been demonstrated to achieve successful vascular filling [39].There are even more economical methods. Administering CPR typechest massage [29] to a deceased in order to circulate contrastagent is free of hardware cost. However, it may confound findingsor technically fail in decomposed or trauma cases. Another tech-nique would require manually administering contrast agent intothe body by use of regular injection or infusion techniques [40].

4.2. Handling

Handling of the small immersion pump has both advantagesand challenges. It is very small and light so it is mobile and doesnot require a lot of storage. Handling characteristics could be im-proved by an affordable wire frame that fixes the pump to thecontainer.

4.3. Pump performance

While so far, no specific advantage of a roller or peristalticpump over a centrifugal pump for PMCTA has been made public,our first results show that they perform equally well. The very lowcost pump exhibits the typical characteristics of a centrifugalpump. Its flow rate both depends on voltage and the fluid's visc-osity (Fig. 5). A quantitative flow rate and voltage correlation canbe achieved within a short amount of time using simple means. Todeliver carefully dosed PMCTA that automatically reduces pres-sure, roller pump systems would require a costly control systemlogic connected to intravascular pressure monitoring. This smallcentrifugal pump will reduce output against significantly in-creased pressure by itself [41].

4.4. Imaging

The vessel filling both regarding overall aspects (Figs. 3 and 4)and details (Fig. 2) are similar between both IP PMCTA and HLMPMCTA.

4.5. Outlook

For the most part, the end user of forensic PMCTA will be atypical government institute that covers forensic medicine.

Free accessibility, the option for critical appraisal as well asaffordability are paramount characteristics of the materials used.Whole industries were able to unfold their potential only afteraffordable and economic products were built and sold, such as theautomobile industry [42]. The International Society of ForensicRadiology and Imaging (ISFRI)2 is a platform that engages in andencourages open dialog. With that, one question is just how manytechnical features are really needed and what the cheapest optionsreally are.

Advanced postmortem angiography techniques employ botharterial and venous filling phases with an added back-and-forthtype third phase [11,43]. Manually switching between two si-multaneously attached immersion pumps – one for arterial, theother for venous access – could be a relatively simple optionwithout significantly increasing hardware expenses. Furthermore,adding intravascular pressure sensors to the setup may also helpto exceed visualization quality to beyond what current setups mayprovide without excessive extra cost.

The best choice for very affordable PMCTA contrast agent so-lution is not clear either. In some jurisdictions drainage of oilysubstances into the common sewerage and burial of oil con-taminated bodies in gravesite burials may be regarded as an

2 Website: http://www.isfri.com.

environmental hazard [44–48].

5. Conclusions

Very affordable postmortem CT angiography may not only befinancially easy to accomplish, but not difficult to perform tech-nically. In order to become a widely accepted tool, PMCTA must beperformed with materials that are as affordable as possible, andthis is a clear step into that direction. This IP PMCTA technique issimple in handling, at low cost and therefore disposable for in-stitutes that work within tight economical budgeting and offers apotential PMCTA solution for any forensic facility world-wide.

Competing interests

The authors declare that they have no competing interests. Inparticular, they have no pending or published patents regardingcontrast agents, pumping devices or other materials relevant toconduct postmortem computed tomography angiography.

Acknowledgments

The authors thank members of the Virtopsy team for theirvaluable discussion and paper editing.

References

[1] Christian Jackowski, Martin Sonnenschein, Michael J. Thali, Emin Aghayev,Gabriel von Allmen, Kathrin Yen, Richard Dirnhofer, Peter Vock, Virtopsy:postmortem minimally invasive angiography using cross section techniques—implementation and preliminary results, J. Forensic Sci. 50 (5) (2005)1175–1186.

[2] Silke Grabherr, Valentin Djonov, Armin Friess, Michael J. Thali, Gerhard Ranner,Peter Vock, Richard Dirnhofer, Postmortem angiography after vascular perfu-sion with diesel oil and a lipophilic contrast agent, Am. J. Roentgenol. 187 (5)(2006) W515–W523.

[3] Silke Grabherr, C.H. Murten, Erich Gygax, C.H. Bern, X-ray Contrast Agent forPost-Mortem Experimental and Diagnostic Angiography, European PatentEP2063916, June 2009.

[4] Silke Grabherr, C.H. Murten, Erich Gygax, C.H. Bern, Radiographic ContrastAgent for Postmortem, Experimental and Diagnostic Angiography, US-Patent20100021389, January 2010.

[5] Silke Grabherr, C.H. Denges, Erich Gygax, C.H. Bern, X Ray Contrasting Agent forPost-Mortem, Experimental and Diagnostic Angiography, European PatentEP2526973, November 2012.

[6] Silke Grabherr, C.H. Murten, Erich Gygax, C.H. Bern, Radiographic ContrastAgent for Postmortem, Experimental and Diagnostic Angiography, US-Patent20130045168, February 2013.

[7] Forim-X-AG, Richard Dirnhofer, Virtangio(R), Swiss Brand Name Registry51685/2010, June 2010.

[8] Silke Grabherr, TWGPAM Organization and Management ⟨http://www.postmortem-angio.ch/twgpam/twgpam-members/⟩ (accessed August 2015).

[9] Silke Grabherr, Marco Dominietto, Lisa Yu, Valentin Djonov, Bert Müller, Se-bastian Friess, Angiofil: a novel radio-contrast agent for post-mortem micro-angiography, In: Optical Engineeringþ Applications, International Society forOptics and Photonics, Bellingham, Washington, USA, 2008, p. 70781O.

[10] Silke Grabherr, Andreas Hess, Marek Karolczak, Michael J. Thali,Sebastian D. Friess, Willi A. Kalender, Richard Dirnhofer, Valentin Djonov,Angiofil®-mediated visualization of the vascular system by microcomputedtomography: a feasibility study, Microsc. Res. Tech. 71 (7) (2008) 551–556.

[11] Silke Grabherr, Francesco Doenz, Beat Steger, Richard Dirnhofer,Alejandro Dominguez, Barbara Sollberger, Erich Gygax, Elena Rizzo,Christine Chevallier, Reto Meuli, et al., Multi-phase post-mortem CT angio-graphy: development of a standardized protocol, Int. J. Leg. Med. 125 (6)(2011) 791–802.

[12] Silke Grabherr, Christèle Widmer, Katia Iglesias, Frank Sporkert,Marc Augsburger, Patrice Mangin, Cristian Palmiere, Postmortem biochem-istry performed on vitreous humor after postmortem CT-angiography, Leg.Med. 14 (6) (2012) 297–303.

[13] Chevallier Christine, Doenz Francesco, Vaucher Paul, Palmiere Cristian,Dominguez Alejandro, Binaghi Stefano, Mangin Patrice, Grabherr Silke, Post-mortem computed tomography angiography vs. conventional autopsy:

W. Schweitzer et al. / Journal of Forensic Radiology and Imaging 5 (2016) 8–1414

advantages and inconveniences of each method, Int. J. Leg. Med. 127 (5) (2013)981–989.

[14] Dominic Wichmann, Axel Heinemann, Clemens Weinberg, Hermann Vogel,Wilhelm Wolfgang Hoepker, Silke Grabherr, Klaus Pueschel, Stefan Kluge,Virtual autopsy with multiphase postmortem computed tomographic angio-graphy versus traditional medical autopsy to investigate unexpected deaths ofhospitalized patients: a cohort study, Ann. Intern. Med. 160 (8) (2014)534–541.

[15] Silke Grabherr, Holger Wittig, Fabrice Dedouit, Krzysztof Wozniak, HermannVogel, Axel Heinemann, Florian Fischer, Artur Moskala, Guiseppe Guglielmi,Patrice Mangin, Jochen Grimm, Pitfalls in post-mortem CT-angiography-in-travascular contrast induces post-mortem pericardial effusion, Leg. Med.(Tokyo Japan), 17(3), 2014, 218-219.

[16] Silke Grabherr, Jochen Grimm, Pia Baumann, Patrice Mangin, Application ofcontrast media in post-mortem imaging (CT and MRI), La Radiol. Med. (2015),1–11.

[17] Guy N. Rutty, Jade Barber, Jasmin Amoroso, Bruno Morgan, Eleanor A.M. Graham, The effect on cadaver blood DNA identification by the use oftargeted and whole body post-mortem computed tomography angiography,Forensic Sci. Med. Pathol. 9 (4) (2013) 489–495.

[18] Beatrice Vogel, Axel Heinemann, Antonios Tzikas, Chanasom Poodendaen,Helmut Gulbins, Hermann Reichenspurner, Klaus Püschel, Hermann Vogel,Post-mortem computed tomography (PMCT) and PMCT-angiography aftercardiac surgery. Possibilities and limits, Arch. Med. Sadowej Kryminol. 63 (3)(2013) 155–171.

[19] Artur Moskała, Krzysztof Woźniak, Piotr Kluza, Filip Bolechała, Ewa Rzepecka-Woźniak, Jan Kołodziej, Katarzyna Latacz, Validity of post-mortem computedtomography angiography (PMCTA) in medico-legal diagnostic management ofstab and incised wounds, Arch. Med. Sadowej. Kryminol. 62 (4) (2012)315–326.

[20] Beatrice Vogel, Axel Heinemann, Axel Gehl, Iwao Hasegawa, Wilhelm-Wolfgan Höpker, Chanasom Poodendaen, Antonios Tzikas, Helmut Gulbins,Hermann Reichenspurner, Klaus Püschel, et al., Post-mortem computed to-mography (PMCT) and PMCT-angiography after transvascular cardiac inter-ventions, Arch. Med. Sadowej Kryminol. 63 (4) (2012) 255–266.

[21] Bernard Lo, Marilyn J. Field, Conflict of Interest in Medical Research, Education,and Practice, The National Academies Press, Washington, DC, USA, 2009.

[22] Forim-X-AG, Rolf Steinegger, Angiofil(R), Swiss Brand Name Registry 54916/2006, June 2006.

[23] Steffen Ross, Danny Spendlove, Stephan Bolliger, Andreas Christe,Lars Oesterhelweg, Silke Grabherr, Michael J. Thali, Erich Gygax, Postmortemwhole-body ct angiography: evaluation of two contrast media solutions, Am. J.Roentgenol. 190 (5) (2008) 1380–1389.

[24] Denton A. Cooley, Development of the roller pump for use in the cardio-pulmonary bypass circuit, Texas Heart Inst. J. 14 (2) (1987) 112.

[25] Rufus Porter, Jonathan D. Bradley, Elastic-Tube Pump, US Patent 12,753, April17, 1855.

[26] Eugene E. Allen, Instrument for Transfusion of Blood, US Patent 249,285, 1881.[27] Craig I. Walker, Apostolos Goulas, Performance characteristics of centrifugal

pumps when handling non-newtonian homogeneous slurries, Proc. Inst.Mech. Eng. Part A: J. Power Energy 198 (1) (1984) 41–49.

[28] Alfa Laval, All You Need To Know…Alfa Laval Pump Handbook, Alfa Laval,2002.

[29] Bruno Morgan, Namiko Sakamoto, Seiji Shiotani, Silke Grabherr, Postmortemcomputed tomography (PMCT) scanning with angiography (PMCTA): a de-scription of three distinct methods, In: Guy N. Rutty (Ed.), Essentials of Au-topsy Practice, Springer, London, 2014, pp. 1–21.

[30] Stanislava Kirinčič, Cveto Klofutar, Viscosity of aqueous solutions of polyethylene glycols at 298.15 K, Fluid Phase Equilib. 155 (2) (1999) 311–325.

[31] Ian S.D. Roberts, Rachel.E. Benamore, Charles Peebles, Carl Roobottom, ZoeC. Traill, Diagnosis of coronary artery disease using minimally invasive au-topsy: evaluation of a novel method of post-mortem coronary CT angiography,

Clin. Radiol. 66 (7) (2011) 645–650.[32] Guy N. Rutty, Gil Brogdon, Fabrice Dedouit, Silke Grabherr, Gary M. Hatch,

Christian Jackowski, Peter Leth, Anders Persson, Thomas D. Ruder,Seiji Shiotani, et al., Terminology used in publications for post-mortem cross-sectional imaging, Int. J. Legal Med. 127 (2) (2013) 465–466.

[33] Used Heart-Lung Machine—Stockert Shiley—10-00-00 Roller Pump, Item535330733. ⟨http://www.medwow.com/used-heart-lung-machine/stockert-shiley/10-00-00-roller-pump/535330733.item⟩, 2015.

[34] Patricia M. Flach, Dominic Gascho, Wolf Schweitzer, Thomas D. Ruder,Nicole Berger, Steffen G. Ross, Michael J. Thali, Garyfalia Ampanozi, Imaging inforensic radiology: an illustrated guide for postmortem computed tomo-graphy technique and protocols, Forensic Sci. Med. Pathol. 10 (4) (2014)583–606.

[35] Steffen G. Ross, Stephan A. Bolliger, Garyfalia Ampanozi, Lars Oesterhelweg,Michael J. Thali, Patricia M. Flach, Postmortem ct angiography: capabilities andlimitations in traumatic and natural causes of death, Radiographics 34 (3)(2014) 830–846.

[36] Thomas D. Ruder, Katja Schulze, Steffen Ross, Garyfalia Ampanozi,Dominic Gascho, Patrick Laberke, Michael J. Thali, Gary M. Hatch, Into thedecomposed body—feasibility of post-mortem CT angiography in a decom-posed cadaver, J. Forensic Radiol. Imaging 2 (3) (2014) 149–152.

[37] Sarah L. Saunders, Bruno Morgan, Vimal Raj, Claire E. Robinson, Guy N. Rutty,Targeted post-mortem computed tomography cardiac angiography: proof ofconcept, Int. J. Leg. Med. 125 (4) (2011) 609–616.

[38] Patrick J. Laberke, Steffen Ross, Patricia Mildred Flach, Garyfalia Ampanozi,Wolf Schweitzer, Michael Thali, Jürgen Fornaro, Selective vascular access forpostmortem magnetic resonance angiography of the coronary arteries, In:23rd Congress of the International Academy of Legal Medicine, 2015.

[39] Marianne Jolibert, Christophe Bartoli, Pierre-Eloi Laurent, Catherine Boval,Vincent Vidal, Marie-Dominique Piercecchi, Frederic Cohen, George Leonetti,Jean-Michel Bartoli, Guillaume Gorincour, Do we really need a dedicatedequipment to achieve a multiphase post-mortem {CT} angiography? J. For-ensic Radiol. Imaging 1 (2) (2013) 83.

[40] Sabrina Mauf, Dominic Gascho, Rilana Baumeister, Patricia Flach, Jürgen For-naro, Thorsten Jentzsch, Michael Thali, Patrick Laberke, Low-cost postmortemcomputed tomography angiography: everybody can do it!, In: AAFS 68thAnnual Scientific Meeting, Las Vegas, Nevada, USA, AAFS, February 2016.

[41] Chris O'Donnell, When it goes wrong: complications of post-mortem CT an-giography (PMCTA) as performed at VIFM, J. Forensic Radiol. Imaging 1 (2)(2013) 84–85.

[42] Lindsay Brooke, Ford Model T: The Car That Put the World on Wheels, MBIPublishing Company, Minneapolis, MN, USA, 2008.

[43] Christine Bruguier, Pascal J. Mosimann, Paul Vaucher, Antoine Uské,Francesco Doenz, Christian Jackowski, Patrice Mangin, Silke Grabherr, Multi-phase postmortem CT angiography: recognizing technique-related artefactsand pitfalls, Int. J. Legal Med. 127 (3) (2013) 639–652.

[44] Bestattungsverordnung. Kapitel 2 Paragraph 2/2. Kanton Aargau 371.112.(“Friedhöfe dürfen die öffentliche Gesundheit und die Umwelt nichtgefährden”).

[45] Bestattungsverordnung. 3. Abschnitt, Kapitel C, Paragraph 29. Kanton Zürich818.61. (sinngemäss: “nur beigesetzt oder ausgebracht werden, wenn dieBestimmungen des Forst-, Gewässerschutz-, Luftfahrt-, Bau- und Umwel-trechts eingehalten werden”).

[46] Bundesversammlung der Schweizerischen Eidgenossenschaft. 814.01 Bun-desgesetz ber den Umweltschutz. Systematische Rechtssammlung -Landesrecht.

[47] Bundesministerium der Justiz und für Verbraucherschutz. Altölverordnung inder Fassung der Bekanntmachung vom 16. April 2002 (BGBl. I S. 1368), diezuletzt durch Artikel 5 Absatz 14 des Gesetzes vom 24. Februar 2012 (BGBl. I S.212) geändert worden ist. Juris, Saarbrücken, Deutschland.

[48] Daniel Hartmann, Benjamin Meylan, Beat Jordi, Management des Grundwas-sers in der Schweiz, Bundesamt für Umwelt BFU, Berne, Switzerland, 2008.