Ex Situ Limb Perfusion System to Extend Vascularized...

Ex Situ Limb Perfusion System to ExtendVascularized Composite Tissue Allograft Survivalin SwineKagan Ozer, MD,1 Alvaro Rojas-Pena, MD,2 Christopher L. Mendias, PhD,1 Benjamin Bryner, MD,2

Cory Toomasian, BS,3 and Robert H. Bartlett, MD2

Background.Organ perfusion systems have successfully been applied in solid organ transplantations. Their use in limb trans-plantation and replantation has not been widely investigated. In this study, we tested the potential for ex situ perfusion system toprolong limb allograft viability in a swine forelimb amputation/replantation model.Methods. Fourteen swine were used. In group1 (n = 4), we perfused 4 amputated limbs for 12 hours usingwarm (27°C–32°C) autologous blood. Group 2 (n = 3) served as a coldpreservation control group, preserving limbs for 6 hours at 4°C. All limbs were transplanted into healthy swine (n = 7) and observedfor another 12 hours. Hemodynamic variables of circulation, as well as perfusate gases and electrolytes (pH, pCO2, pO2, O2 sat-uration, Na+, K+, Cl−, Ca2+, HCO3

−, glucose, lactate) were measured. Muscle samples were used to measure single-muscle fibercontractility.Results. In the control group, no microcirculation was observed after 6 hours of cold storage. In the pump perfusiongroup, all limbs displayed a gradual increase in lactate levels (P < 0.05) during ex situ perfusion that returned to normal after trans-plantation and reperfusion (P = 0.05). The pH and potassium remained stable throughout the experiment. Single-muscle fiber con-tractility testing showed near normal contractility at the end of the reperfusion period (P > 0.05). Limb weight did not increasesignificantly between the end of pump perfusion and reperfusion (P > 0.05). Conclusions. We demonstrated the potential topreserve limb allograft using ex vivo circulation. This approach promises to extend the narrow time frame for revascularization ofprocured extremities in limb transplantation.

(Transplantation 2015;99: 2095–2101)

Hand transplantation became a reality for failed majorupper extremity replantations.1However, the procedure

is limited by the number of available heart-beating donorsand is subjected to time restraints.2 Once a potential donoris identified, in a relatively short time, standardized screeningtests must be obtained and decisions must be made to pro-ceed.3 The harvested extremity must be revascularized pref-erably within 4 hours to prevent the muscle damage due toreperfusion injury.4 Similarly, in upper extremity trauma, re-plantation is possible if done soon after the injury, but successis limited by many factors, including the availability of the

microsurgery team and the warm ischemia time.5 Currently,these issues are addressed by immediate transfer of the patientto specialized centers and perfusing the amputated limb withclear cold solution and preserving it at cold temperature untilthe replantation can be done. In amputations proximal to thewrist joint, replantation is considered futile if the warm ische-mia time exceeds 6 hours.

Tissue preservation in extremities before revascularizationis a limiting factor in the area of limb transplantation and re-plantation. Given the time restrains imposed by tissue viabil-ity, immediate transfer of an amputated or harvested limb isessential for success. With recent advances in organ preserva-tion, a prolonged period of tissue survival is possible in solidorgan transplantations with excellent function.6-14 Up to5 days of survival is achieved in experimental kidney perfu-sion systems.6,7,13 Using the experience accumulated in previ-ous studies, we investigated the feasibility of keeping swinelimbs viable on an ex situ limb perfusion system in this study.

MATERIALS AND METHODSThe University Committee on Use and Care of Animals

approved this study. The University Committee on Use andCare of Animals, United States Department of Agricultureand Association for Assessment and Accreditation of Lab-oratory Animal Care guidelines and policies for humane useof animals in research were followed in this study.

Experimental groups: In group 1, four swine forelimbswere amputated, connected to ex situ perfusion for 11 hoursand transplanted to another healthy swine. In group 2, 3 swine

Received 6 January 2015. Revision requested 20 February 2015.

Accepted 28 February 2015.1 Department of Orthopaedic Surgery, University of Michigan, Ann Arbor, MI.2 Department of Surgery, University of Michigan, Ann Arbor, MI.3 Wayne State University Medical School, Detroit, MI.

This study is in part made possible by a grant received from the American Societyfor Surgery of the Hand.

The authors declare no conflicts of interest.

K.O., A.R.P., R.H.B. participated in research design. K.O., A.R.P., C.L.M., R.H.B.participated in the writing of the article. K.O., A.R.P., C.L.M., B.B., C.T. participatedin the performance of the research. C.L.M. contributed new reagents or analytictools. K.O., A.R.P., B.B. participated in data analysis.

Correspondence: Kagan Ozer, MD, Department of Orthopaedic Surgery, Universityof Michigan, 2098 South Main St., Ann Arbor, MI 48105. ([email protected]).

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

ISSN: 0041-1337/15/9910-2095

DOI: 10.1097/TP.0000000000000756

OriginalBasicScienceçGeneral

Transplantation ■ October 2015 ■ Volume 99 ■ Number 10 www.transplantjournal.com 2095


forelimbs were amputated, preserved at 4°C for 6 hours,and transplanted to another healthy swine. Recipients inboth groups were observed 12 additional hours duringthe reperfusion.

Animal Model

Swine, 40 ± 5 kg, were sedated with an intramuscular mixof 5 mg/kg tiletamine HCl and zolepam HCl (Telazol WyethHoldings Corporation, Carolina, Puerto Rico) and 3 mg/kgxylazine (TranquiVed Vedco Inc, St. Joseph, MO). Animalswere intubated and ventilated with 100% O2 and 1% to3% isofluorane (Hospira, Inc., Lake Forest, IL). Initial me-chanical ventilator settings were adjusted to keep pCO2 at35 to 45 mm Hg and peak inspiratory pressures less than20 cm H2O. The right carotid artery and right internal jugu-lar vein were catheterized to monitor mean arterial bloodpressure and heart rate, and to collect blood samples. A cen-tral line was placed in the internal jugular vein to monitorcentral venous pressure and maintenance fluid administration(80–120 mL/hr). Prophylactic antibiotic (cefazolin, 1 g) wasadministered throughout the procedure every 6 hours.

Surgical Technique

Preparation of the donor forelimb: The animal was placedsupine on the operating table. Both forelimbs were shavedfirst and prepared using betadine solution for skin cleansing.Two longitudinal incisions were placed on the medial andlateral aspects of the limb, approximately 12 cm each. Bothof these incisions were connected to a circumferential inci-sion around the antecubital fossa. First, the incision on thelateral aspect of the limb was explored. Two skin flaps (ante-rior and posterior) were elevated to expose the lateral aspectof the flexor and extensor compartments. The triceps tendonwas detached from the olecranon and marked using 0 silk.On the medial side, 2 skin flaps were elevated anteriorly andposteriorly. Care was taken to protect the cephalic vein duringthe dissection. The biceps tendon was identified at the ante-cubital fossa and detached along with the brachialis muscle.

This was followed by elevation of flexor/pronator mass andextensor compartment muscles off the medial and lateralepicondyles, respectively. The median, ulnar and radial nerveswere then identified and transected. An electrode was placedon the median nerve for later use in neuromuscular electricalstimulation. Finally, the brachial artery and accompanyingveins were transected 5 cm proximal to the antecubital fossa,and the elbow was disarticulated. The brachial artery wascannulated, and limbs were flushed with 10000 units of hep-arin in 100 mL of saline without any other preservative solu-tion. In the control group, procured forelimb was taken to acold storage room. In the experimental group, the limb wasattached to the ex situ limb perfusion system.

Preparation of the recipient and transplantation: Recipientwas anesthesized 2 hours before the end of preservation. Ampu-tation was performed on the ipsilateral forelimb at the samelevel. At the end of its preservation, the limb was broughtto the operating field and fixed to the recipient’s ulnohumeraljoint using 2 cross Steinman pins. This was followed by re-approximation of biceps, triceps, forearm flexor, and exten-sor muscles. The ends of the neurovascular bundle at theantecubital fossa were approximated, and the artery was anas-tomosed using 8.0 ethilon suture under magnification alongwith its 2 accompanying veins. In addition, the cephalic veinand/or basilic veins were also anastomosed when available.Before opening the clamps for revascularization of the limb,recipient was injected with 200 μm/kg heparin subcutane-ously. Skin was approximated using 2.0 silk.

Ex-situ Perfusion Protocol

Throughout this study, we used pressure-limited pulsatileperfusion (with organ flow controlled by resistance), room tem-perature (27°C–32°C), open venous drainage, andmembraneoxygenation to maintain stable oxygenation and acid-basebalance (Figure 1). As a perfusate, we used fresh autologousplasma (hematocrit, 10%), with periodic addition of glucose.The perfusate did not contain white cells or platelets.

Preparation of the perfusate: first, the perfusion system(including the pump and all collateral circuit) was primed

FIGURE 1. Amputated limb is stored in the top container in a temperature-regulated environment. Deoxygenated blood is collected in a per-fusate reservoir, which is connected to the pump and oxygenator to complete the circle. Amputated limb is cannulated on the operating tableand connected to the flow probe. Venous return depends on gravity.

2096 Transplantation ■ October 2015 ■ Volume 99 ■ Number 10 www.transplantjournal.com


using 200 cc of Dulbecco’s Modified Eagle’s Medium (Lonza,Walkersville, MD). In the meantime, the donor blood obtainedafter procurement of the limb was centrifuged at 3500 RPMfor 15 minutes to remove white blood cell and platelet frac-tions. Once those fractions were removed, packed red bloodcells and plasma were mixed in 1 to 2 ratio, along with dex-tran to reach the colloid osmotic pressure of 20 to 30mmHg.Colloid osmotic pressure was measured using osmometer(Wescor Inc., Logan, UT) before this mixture was added tothe pump. Throughout the experiment, we gradually removed160 mL of perfusate every 2 hours, and substituted with re-placement perfusatewith 10%hemoglobinwith similar param-eters of plasma oncotic pressure. This exchange removed toxicmetabolites from the system in the absence of liver and kidney.We also measured perfusate glucose levels every 2 hours andsubstituted the perfusate with either glucose (1 cc D50%) ifthe level was less than 4.5 mmol/L, or with insulin (2 units) ifglucose was above 14mmol/L.

The perfusion apparatus consisted of an RM3 pulsatileperfusion pump (Waters Medical Systems, Minneapolis, MN)and a commercially available Terumo oxygenator (Baby CapioxRX; Terumo, Tokyo, Japan) 0.5 m2 connected to a 70%O2,5% CO2 gas tank. The gas mixture to the membrane lungwas 5% CO2, 95% O2. The perfusate was fresh pig plasmaand red blood cells, supplemented with glucose to maintainperfusate glucose in the range of 100 to 200 mg/dL. Pulsatileperfusion at room temperature (27–32 °C) was maintainedwith a pressure limitation of 60 to 80mmHg.

During perfusion, the limb was prepared in sterile fashionand maintained in an antiseptic Dakins solution dressing,and the perfusate was continuously sterilized with ultravioletlight. Flow was measured continuously, and vascular resis-tance was calculated frequently as an indicator of successfulperfusion. The following parameters of perfusion were mea-sured throughout organ perfusion: pressure, flow, resistance,temperature, O2 delivery and consumption, limb weight andcircumference, acid-base status, and blood gases.

Monitoring

The following variables were measured throughout theexperiment:

• Perfusion parameters (from the limb on pump perfusion)• From the perfusate and the recipient (every 2 hours): hemoglo-

bin lactate, K+, Na+, Cl−, HCO3−, pH, pCO2, pO2, glucose.

• Neuromuscular electrical stimulation every 4 hours using anerve stimulator.

Single-Muscle Contractility Testing

The contractility of permeabilized muscle fibers was testedas previously described.15,16 Muscle biopsies were taken par-allel to the longitudinal axis of muscle fibers sized 10 mm inlength and 5 mm in width (Figure 2). Biopsies were immedi-ately placed in a relaxing solution containing Ca2+ chelatingagents in a pH-buffered solution that causes muscle fibersto relax. Fiber bundles of approximately 4 to 6 mm in lengthand 0.5 to 0.75 mm in diameter were dissected from musclebiopsies. Bundles were placed in skinning solution to furtherpermeabilize the cell membranes for 30 min, and then instorage solution for 16 hours at 4°C, followed by storage at−80°C.On the day of single-fiber testing, bundleswere thawedslowly on wet ice, and individual fibers were pulled frombundles using fine forceps. Fibers were then placed in a cham-ber containing relaxing solution and secured at one end to aservomotor (Aurora Scientific, Aurora, ON) and at the otherend to a force transducer (Aurora Scientific) using 2 ties of10-0monofilament nylon suture at each end. Using a laser dif-fraction measurement system, fiber length was adjusted toobtain a sarcomere length of 2.7 μm. The average fiber cross-sectional areawas calculated assuming an elliptical cross-section,with diameters obtained at 5 positions along the fiber fromhigh-magnification images at 2 different views (top and side).Maximumfiber isometric force (Fo)was elicitedby immersingthe fiber in a solution containing a high concentration of Ca2+

andadenosinetriphosphate (ATP).Specific forceof fibers (sFo)was calculated by dividing Fo by fiber cross-sectional area. Fi-bers were categorized as fast or slow by examining their forceresponse to rapid, constant-velocity shortening, and 10 fastfibers were tested from each biopsy sample.

Statistical Analysis

Data are presented as mean ± SEM. Differences betweenthe control and perfusion groups were tested using paired

FIGURE 2. Muscle fiber contractility testing diagram. A, Overview of the general orientation of the fiber in the testing system, and a rep-resentative image demonstrating the mid-portion of a porcine muscle fiber in the system. (B) Calculation of muscle fiber CSA. Five widthmeasurements from the top and side views of the mid-portion fiber are fitted to the formula of an ellipse to calculate fiber CSA. CSA, cross-sectional area.

© 2015 Wolters Kluwer Ozer et al 2097


t tests (α=0.05) inExcel statistical software, v12.2.3. (Microsoft,Redmond, WA).

RESULTSAll animals in group 1 displayed satisfactory flow to the

extremity after transplantation and remained hemodynami-cally stable. In group 2, after 6 hours of cold storage andtransplantation, blood flowwas established only in large sizevessels (brachial and radial arteries), there was no flow inmicrocirculation or in muscle planes.

All recipients remained hemodynamically stable. Perfusedlimbs initially showed a higher resistance within the first3 hours (P < 0.05), which later returned to normal (Figure 2).Overall weight gain at the end of perfusion was statisticallysignificant compared to limb weight before pump perfusion(P<0.05).No significant gainwas observed between thebe-ginning and end of reperfusion (P = 0.05; Figure 3).

Among all measured parameters in Figure 3, K+, pH,pCO2, and pO2 did not show any statistically significantchange before (on pump perfusion) or after transplantation(Figure 4, P > 0.05). Lactate levels however showed a gradual

increase on pump perfusion (P < 0.05) and returned to nor-mal levels after the transplantation (during reperfusion).

Single-muscle fiber contractility testing showed a gradualdecline in the cold preservation group, whereas in the pumpperfusion group, contractility remained close to normal con-trols during the perfusion and reperfusion afterward. Inthe control group (Figure 5, right side), muscle contractil-ity shows a statistically significant decline at 12, 18, and24 hours after the amputation compared to the contralat-eral side, as seen on Fo and sFo values (P < 0.05). In thepump perfusion group, muscle contractility returned closeto contralateral limb values at 24 hours (P > 0.05; Figure 5,left column).

DISCUSSIONIn this study, we demonstrated the feasibility of using an

ex situ warm perfusion system to keep limb allografts via-ble up to 12 hours. This is demonstrated by ongoing observ-able muscle contraction in response to electrical stimulationthroughout the reperfusion phase, as well as marked improve-ments in sFo values at the end of 12 hours of reperfusion.

FIGURE 3. Data in left column show hemodynamic parameters of perfusion while the limbwas attached to ex situ pump. Data in right columnshow hemodynamic changes in the recipient after the transplantation. Data do not include cold perfusion control group as there was no flow inthe microcirculation after 4 hours of ischemia. Error bars represent standard error of means (SEM). MAP, mean arterial pressure; HR, heart rate;bpm, beats per minute.



FIGURE 4. Data in left column show metabolic parameters of limb perfusion when the limb was attached to ex situ perfusion. Data in rightcolumn show metabolic parameters of recipient after transplantation of the limb. Error bars represent standard error of means (SEM).



Using alternating doses of insulin/glucose and changing theperfusate every 2 hours, we were able to maintain potassium,pH, and lactate at constant levels. Oxygen delivery and con-sumption were also both well maintained under at room tem-perature (27°C–32°C).

Current standard of care for the preservation ofvascularized composite tissue allografts is static cold storageat 4°C using UW solution. This method has been widely usedin solid organ transplantations since the 1960s.17 The firststatic storage solutions were introduced by Collins et al.17

and were subsequently modified throughout the 1980s asEuroCollin’s solution.18 Cooling slows metabolic processesand prolongs the time during which the organ can be deprivedof oxygenwithout loss of viability.However, energy consump-tion due to metabolic activity is not halted but reduced and,ultimately, the limitation of cold storage is determined bymitochondrial damage and subsequent ATP depletion.19

In cold storage, metabolism is slowed 1.5- to 2-fold for every10°C drop in temperature,20 and anaerobicmetabolism leadsto depletion of ATP stores and build up of acidotic environ-ment.21 To prevent and restore energy charges, normother-mic preservation methods gained popularity in recent years.Studies have shown maintenance of normal metabolic func-tion as in liver, kidney, and lung transplant models.13,22,23

In this study, we used blood at room temperature (27°C–32°C)andwere able tomaintain stable oxygen delivery and consump-tion after transplantation and 12 hours of reperfusion. This isfurther supported in the current study by measures of muscleforce production at the cellular level. Limbs on pump perfu-sion had smaller decreases in sFo, and when the limb wasreattached, the force values improved markedly. This is in

stark contrast to the limbs on cold perfusion, which rapidlylost sFO and did not regain force production once the limbwas reattached. The recovery of force production is likelydue to improved muscle cell survival and a reduction in apo-ptosis. As muscle tissue death can limit the viability of a limbafter reattachment, this suggests that not only does pumpper-fusion improve muscle force production acutely, it also likelycontributes to long-term survival of the transplanted limb.

Initial perfusion experiments using blood were mostlyfocused on hemodynamic parameters of limb perfusion.24,25

Blood, or hemoglobin as an oxygen carrier, has not beenused for decades because of its potential to propagate meta-bolic overload and stimulate oxidative metabolism. It hasbeen shown on many animal models as well as in the clinicalsetting that the use of blood as an artificial oxygen carrier issafe and reliable. Few studies however have shown the effec-tive use of normothermic blood on composite tissue allo-grafts. Constantinescu et al26 showed successful perfusionof amputated swine limbs up to 12 hours. In their feasibilitystudy, they used a modern extracorporeal heart-lung circuitand autologous blood at 32°C for perfusion. Their resultsindicated effective perfusion of the limbs as evidenced bysteady mean arterial perfusion, high pH, and low lactatelevels.26 The study however did not include the replanta-tion of the limb to another recipient, and therefore did notaccount for the ischemia-reperfusion injury. Later, the samegroup perfused swine forelimbs for 12 hours and replantedto another animal to observe changes associated with ischemia/reperfusion injury.27 Using histopathology and markers ofinflammation, they demonstrated that extracorporeal circu-lation had a negligible effect on reperfusion-induced injuries.

FIGURE 5. Arrow on x axis indicates the time of transplantation. After transplantation, biopsies were obtained at first, sixth, and 12th hours.Graphs in the left column show data regarding muscle contractility within the first 12 hours on pump perfusion and 12 hours after thetransplantation. Data on the right show muscle contractility of control group, which was subjected to 6 hours of cold preservation andtransplantation afterward.



Our findings are similar and further show not only survivalbut also promising functional recovery as evidenced bysingle-muscle fiber contractility testing.

Limitations of the current study include limited reperfu-sion period (12 hours). As a result, we were able to obtainthe information regarding the initial phase of the reperfusioninjury. To observe its long-term effects on nerve regenera-tion and limb function, we need to conduct a separate setof experiments to keep recipient animals alive under lowdose immunosuppressants until after the nerve regenerationis completed.

The animal model presented in the current study is closerto human upper extremities in terms of its metabolism andcirculatory hemodynamics. The method has a potential toapplication in limb replantation and transplantation.

REFERENCES1. Petruzzo P, Lanzetta M, Dubernard JM, et al. The international registry

on hand and composite tissue transplantation. Transplantation. 2008;86:487.

2. Hettiaratchy S, Randolph MA, Andrew Lee WP. Long-term considerationof hand transplantation. Transplantation. 2003;75:1605.

3. McDiarmid SV, Azari KK. Donor-related issues in hand transplantation.Hand Clin. 2011;27:545.

4. Hartzell TL, Benhaim P, Imbriglia JE, et al. Surgical and technical aspectsof hand transplantation: is it just another replant?Hand Clin. 2011;27:521.

5. Ozer K, Kramer W, Gillani S, et al. Replantation versus revision of amputatedfingers in patients air-transported to a level 1 trauma center. J Hand SurgAm. 2010;35:936.

6. Butler AJ, Rees MA, Wight DG, et al. Successful extracorporeal porcineliver perfusion for 72 hr. Transplantation. 2002;73:1212.

7. Hoffman FM. Outcomes and complications after heart transplantation: areview. J Cardiovasc Nurs. 2005;20:S31.

8. Karck M, Vivi A, Tassini M, et al. The effectiveness of University ofWisconsin solution on prolonged myocardial protection as assessed byphosphorus 31-nuclear magnetic resonance spectroscopy and functionalrecovery. J Thorac Cardiovasc Surg. 1992;104:1356.

9. Mankad PS, Severs NJ, Lachno DR, et al. Superior qualities of Universityof Wisconsin solution for ex vivo preservation of the pig heart. J ThoracCardiovasc Surg. 1992;104:229.

10. McLarenAJ, Friend PJ. Trends in organ preservation. Transpl Int. 2003;16:701.

11. Muhlbacher F, Langer F, Mittermayer C. Preservation solutions fortransplantation. Transplant Proc. 1999;31:2069.

12. Southard JH, Belzer FO. Organ preservation. Annu Rev Med. 1995;46:235.

13. St Peter SD, Imber CJ, Friend PJ. Liver and kidney preservation byperfusion. Lancet. 2002;359:604.

14. Stringham JC, Paulsen KL, Southard JH, et al. Prolonging myocardialpreservation with a modified University of Wisconsin solution containing2,3-butanedione monoxime and calcium. J Thorac Cardiovasc Surg. 1994;107:764.

15. Gumucio JP, DavisME, Bradley JR, et al. Rotator cuff tear reducesmusclefiber specific force production and induces macrophage accumulationand autophagy. J Orthop Res. 2012;30:1963.

16. Mendias CL, Kayupov E, Bradley JR, et al. Decreased specific force andpower production of muscle fibers from myostatin-deficient mice areassociatedwith a suppression of protein degradation. JAppl Physiol (1985).2011;111:185.

17. Collins GM, Bravo-Shugarman M, Terasaki PI. Kidney preservation fortransportation. Initial perfusion and 30 hours’ ice storage. Lancet. 1969;2:1219.

18. Aydin G, Okiye SE, Zincke H. Successful 24-hour preservation of theischemic canine kidney with Euro-Collins solution. J Urol. 1982;128:1401.

19. Net M, Valero R, Almenara R, et al. The effect of normothermic recirculationis mediated by ischemic preconditioning in NHBD liver transplantation.Am J Transplant. 2005;5:2385.

20. Clavien PA, Harvey PR, Strasberg SM. Preservation and reperfusioninjuries in liver allografts. An overview and synthesis of current studies.Transplantation. 1992;53:957.

21. Carini R, Autelli R, Bellomo G, et al. Alterations of cell volumeregulation in the development of hepatocyte necrosis. Exp Cell Res.1999;248:280.

22. Brockmann J, Reddy S, Coussios C, et al. Normothermic perfusion: anew paradigm for organ preservation. Ann Surg. 2009;250:1.

23. Cypel M, Rubacha M, Yeung J, et al. Normothermic ex vivo perfusion preventslung injury compared to extended cold preservation for transplantation.Am J Transplant. 2009;9:2262.

24. Delorme TL, Shaw RS, Austen WG. A method of studying “normal”function in the amputated human limb using perfusion. J Bone JointSurg Am. 1964;46:161.

25. O’Donovan MJ, Rowlerson A, Taylor A. Proceedings: the perfusedisolated human limb: an assessment of its viability. J Physiol. 1976;256:27P.

26. Constantinescu MA, Knall E, Xu X, et al. Preservation of amputatedextremities by extracorporeal blood perfusion; a feasibility study in aporcine model. J Surg Res. 2011;171:291.

27. Muller S, Constantinescu MA, Kiermeir DM, et al. Ischemia/reperfusioninjury of porcine limbs after extracorporeal perfusion. J Surg Res. 2013;181:170.



Ex Situ Limb Perfusion System to Extend Vascularized...

Documents

Transcript of Ex Situ Limb Perfusion System to Extend Vascularized...