Liver Splitting During Normothermic Organ Preservation

LETTERS FROM THE FRONTLINE Liver Splitting During Normothermic Organ Preservation TO THE EDITOR: Although widely established as a means to increase the number of patients who can benefit from transplantation, the place of liver splitting is restricted: this is one cause for disappointing graft and recipient survival in early studies. (1) Cold ischemia exacerbated by long transport times for shared grafts had a negative impact on the outcome and led to the development of the in situ split technique to reduce the cold ischemia time. (2) However, this technique inevitably prolongs the operating time in the donor hospital which, as well as the cost implications, may compromise the quality of other retrieved organs. A further complexity is the need for the relevant surgical expertise and specialist equipment at the time of donor retrieval. Nevertheless, the pressure of wait-list mortality and donor organ shortage provides a powerful incentive for expanding the scope of liver splitting. Although excellent results of pediatric split-liver and extended right graft transplants are reported, (3) ex situ split techniques are still limited by the consequences of prolonged cold ischemia. This is inevitable if the organ is retrieved, transported to a transplant center, split, and then 1 graft is transported to a second center for implantation. A similar problem ensues if the 2 implantations take place in a single center sequentially. Very few liver transplant centers have the infrastructure (both Abbreviations: ALT, alanine aminotransferase; DCD, donation after cardiac death; IVC, inferior vena cava. Address reprint requests to Jens G. Brockmann, M.D., Organ Transplant Center, King Faisal Specialist Hospital and Research Center, CSSD Building, Riyadh, Saudi Arabia 11211. Telephone: 1 966 500156931; FAX: 1 966 11 4427772; E-mail: brockmannjgb@me.com Received November 9, 2015; accepted November 29, 2016. Copyright VC 2016 by the American Association for the Study of Liver Diseases. View this article online at wileyonlinelibrary.com. DOI 10.1002/lt.24693 Potential conflict of interest: Nothing to report. facilities and surgical manpower) to carry out 2 liver transplants simultaneously. There is increasing interest in normothermic perfusion as a means to resuscitate suboptimal donor organs and minimize the effects of preservation. (4) Experimental transplant studies have shown superiority compared with conventional cold storage as well as enabling reliable prediction of viability. (5) Previously published nontransplant studies showed the potential for very prolonged storage of up to 72 hours. Normothermic perfusion provides the possibility of carrying out an ex situ liver split in an organ that is still perfused, thereby combining the advantages of the in situ and ex situ techniques. We have recently started a preclinical study in which human livers, retrieved with clinical intent but discarded, are placed on the normothermic perfusion device for periods of 24 hours. We report here the feasibility of carrying out a normothermic split in one such liver following a 24-hour period of perfusion. Materials and Methods A 1297-g liver was retrieved from a 56-year-old female donor. Donation was performed after cardiac death (DCD) with a first warm ischemia time of 20 minutes. This organ was not accepted for transplantation by the recovering center because of appearances suggesting steatosis and fibrosis and was transported to a second transplant center. Although less concerned about the morphological appearances, the second center declined the organ because of logistic reasons, ie, the anticipated blood products needed for the assigned recipient. There were additional concerns about prolonged cold ischemia time. Basically, this organ was recovered for the purpose of transplantation but was discarded during postretrieval assessments. Such organs qualified for research purposes, and the normothermic perfusion study of discarded organs was evaluated and approved by the national ethics review committee of the UK (REC reference 09/H0605/85). Accordingly, this organ was then transported to the Oxford Transplant Center and placed on an automated version of a previously described perfusion apparatus (4,5) (Fig. 1). In summary, this is a blood-based, oxygenated, LETTERS FROM THE FRONTLINE 701

BROCKMANN ET AL. LIVER TRANSPLANTATION, May 2017 Third, metabolic viability is shown by the proof and increase of protein (albumin and total protein) synthesis starting 6 hours after normothermic machine perfusion (Fig. 3) as well as by significant lactate clearance (Fig. 4). The liver showed evidence of oxygen consumption throughout the perfusion. O 2 consumption per 100 mg was between 1.46 to 1.98 ml O 2 /minute (data not shown). Lactate level before connection of the liver was 14.7 mmol/l. Under normothermic machine perfusion, lactate was cleared (end value 0.6 mmol/l). This is in clear contrast to what is described in hypothermic machine perfusion where lactate usually increases constantly over the perfusion period and is a sensitive marker of tissue ischemia. Finally, alanine FIG. 1. Levels during normothermic perfusion for perfusion pressures. nonpulsatile closed circuit providing separate hepatic and portal venous inflow and inferior vena cava (IVC) outflow, using a centrifugal pump and infusions of heparin, insulin, prostacycline, nutrition, and bile salts. The organ was perfused at 37 8C for a period of 24 hours following an initial 9-hour cold ischemia. Thereafter, it was then subjected to a classical liver split (left lateral, segments II and III; extended right lobe, segments I and IV-VIII). Results The initial perfusion revealed a homogeneously perfused graft. The perfusion was and remained stable throughout the planned 24-hour period (Fig. 1). Bile production started 2 hours after normothermic preservation was initiated. Second, acid-base physiology was maintained during the entire preservation period with the need of external additives. The initial drop of ph, bicarbonate, and base excess was regulated by the liver spontaneously indicating functional viability (Fig. 2). FIG. 2. Levels during normothermic perfusion for acid-base homeostasis. 702 LETTERS FROM THE FRONTLINE

LIVER TRANSPLANTATION, Vol. 23, No. 5, 2017 BROCKMANN ET AL. FIG. 5. Levels during normothermic perfusion for ALT. FIG. 3. Levels during normothermic perfusion for protein synthesis. aminotransferase (ALT) levels as a marker for cytosolic cellular damage revealed an initial increase reflecting the reperfusion injury following the start of normothermic perfusion (Fig. 5) until hour 2 of perfusion. Thereafter, no further increase, but a decrease was observed showing no additional cellular injury during FIG. 4. Levels during normothermic perfusion for lactate. the preservation period. Histology at the end of the perfusion period revealed necrosis, but not confluent as seen in other human graft research perfusions. All the described parameters are proof of viability during normothermic perfusion. Potential function following transplantation of such grafts remains to be validated. Clinical trials addressing this issue are underway. The perfusion setup is shown in Figs. 6-10. The split was performed by opening the anterior layer of the left part of the hepatoduodenal ligament and identifying the left hepatic artery. After ligation of the segment IV arterial branch, the left hepatic artery was dissected to the lateral segment. The parenchymal bridge between liver segments II and IV was divided and the ligamentum teres separated from adhesions to segment I. The peritoneal sheath on the right side of the umbilical fissure was incised and a total of 5 portal venous branches entering segment IV identified, ligated, and transected. Branches from the left portal vein to segment I were also divided, preserving branches from the portal venous bifurcation. Following identification and isolation of the left lateral segment inflow structures in this way, the liver was turned upside down (still perfused) in order to isolate the left hepatic vein; this was achieved by transecting the obliterated ductus venosum within the sulcus of Arantius. After successful isolation of the left hepatic vein, a plastic tube was passed between the left and middle hepatic veins, around the posterior aspect of the liver and, after returning the liver to a prone position, used as a hanging sling in order to facilitate parenchymal dissection; this was performed using a simple clamp-crushing technique and ligation of visible vessels (Fig. 2). Once the umbilical plate was reached (Fig. 3) this was LETTERS FROM THE FRONTLINE 703

BROCKMANN ET AL. LIVER TRANSPLANTATION, May 2017 FIG. 6. Normothermic liver perfusion of a DCD liver graft. The picture shows the state before perfusion (upper left corner) and after 15 minutes of a 24-hour perfusion period. Cannulae from left to right: IVC outflow, portal vein inflow, hepatic artery inflow, and bile collection tube (empty). transected using scissors. The right-hand side of the umbilical plate was closed using a running suture. This procedure resulted in 2 separate but perfused grafts after 28 minutes of surgery. Perfusion was paused in order to separate the grafts. The left hepatic artery was ligated to the right and thereafter divided. In the following the left portal vein and left hepatic vein were divided. The resulting FIG. 7. Hanging liver graft on perfusion during parenchymal dissection superior anterior view after 24 hours of normothermic perfusion. Approximately 70% of the parenchymal transection has been achieved using simple clamp dissection and ligation. Please note that the perfusate is fully heparinized. Cannula from left to right: portal vein, IVC, and hepatic artery. FIG. 8. Completed parenchymal dissection. The grafts still being perfused are now only attached by hepatic artery, portal vein, left hepatic vein, and the hilar plate. The left bile duct is located on top of the Mosquito clamp. Cannula from left to right: portal vein and IVC. 704 LETTERS FROM THE FRONTLINE

LIVER TRANSPLANTATION, Vol. 23, No. 5, 2017 BROCKMANN ET AL. FIG. 9. Left lateral split (segments II and II) weighing 283 g. From left to right: left hepatic vein (length 8 mm, diameter 9 mm), left hepatic artery (length 32 mm, diameter 3 mm), left portal vein (length 13 mm, diameter 12 mm), and left hepatic bile duct on the anterior inferior aspect of the portal vein (length 10 mm). venous defects in the main portal vein and the middle hepatic vein of the right graft were closed using 5-0 Prolene running sutures. Following a nonperfused period of 6 minutes, perfusion of the right extended graft was restarted. This yielded a left lateral segment weighing 283 g (Fig. 4) and an extended right, still perfused graft, weighing 1014 g (Fig. 5). Discussion At present, the optimal conditions for transplanting 2 split grafts require this to be carried out in 1 center simultaneously. This requires not only 2 operating rooms but also the surgical, anesthetic, and other personnel needed to run the procedures in parallel. This major logistic problem would potentially be overcome if either the right graft were to be normothermically stored until the first (pediatric) procedure is finished or, alternatively, shipped elsewhere while still avoiding the detriment of additional cold ischemia. Preperfusion biopsies showed mixed fatty changes in up to 60% of the hepatocytes. After perfusion, there was no evidence for hemorrhage or for extensive hepatocellular necrosis. Because this is a feasibility report for a novel surgical technique, ie, the ex situ normothermic liver split, including the presented functional data supporting graft viability, (4) the preperfusion and postperfusion histologies are not presented. Normothermic perfusion, if successfully applied in this way to liver splitting, might minimize the complications of cold ischemia, enabling safe transport and/or storage of the right graft. Further normothermic perfusion of the left split would be technically difficult due to the short vascular connections. In conclusion, this single case report shows that splitting is technically possible in the context of a fully anticoagulated perfusion system, although a successful outcome of transplantation of such a split liver remains to be demonstrated. This case demonstrates the important principle that a single liver can be split ex vivo without loss of oxygenated perfusion, thereby minimizing the ischemic insult and subsequent ischemia/reperfusion injury. This reduction in transplant-related graft injury may enable larger degrees of size mismatch than is currently acceptable and thereby enable more livers to be split and transplanted successfully. FIG. 10. Extended right split (segments I and IV-VIII) weighing 1014 g. Cannulae from left to right: common bile duct, IVC (drained toward the infrahepatic aspect), portal vein and hepatic artery via the celiac trunk (the aortic patch still present). Jens G. Brockmann, M.D., Ph.D. 1 Thomas Vogel, M.D. 2 Constantin Coussios, M.D., Ph.D. 3 Peter J. Friend, M.D., Ph.D. 4 1 Organ Transplant Center King Faisal Specialist Hospital and Research Center Riyadh, Saudi Arabia 2 Department of Surgery University Clinics Muenster Muenster, Germany LETTERS FROM THE FRONTLINE 705

BROCKMANN ET AL. LIVER TRANSPLANTATION, May 2017 3 Department of Engineering Science Institute of Biomedical Engineering and 4 Nuffield Department of Surgical Sciences Medical Sciences Division University of Oxford Oxford, UK REFERENCES 1) Broelsch CE, Emond JC, Whitington PF, Thistlethwaite JR, Baker AL, Lichtor JL. Application of reduced-size liver transplants as split grafts, auxiliary orthotopic grafts, and living related segmental transplants. Ann Surg 1990;212:368-375. 2) Rogiers X, Malago M, Gawad K, Jauch KW, Olausson M, Knoefel WT, et al. In situ splitting of cadaveric livers. The ultimate expansion of a limited donor pool. Ann Surg 1996;224: 331-339. 3) Broering DC, Kim JS, Mueller T, Fischer L, Ganschow R, Bicak T, et al. One hundred thirty-two consecutive pediatric liver transplants without hospital mortality: lessons learned and outlook for the future. Ann Surg 2004;240:1002-1012. 4) Imber CJ, St Peter SD, Lopez de Cenarruzabeitia I, Pigott D, James T, Taylor R, et al. Advantages of normothermic perfusion over cold storage in liver preservation. Transplantation 2002;73: 701-709. 5) Brockmann J, Reddy S, Coussios C, Pigott D, Guirriero D, Hughes D, et al. Normothermic perfusion: a new paradigm for organ preservation. Ann Surg 2009;250:1-6. 706 LETTERS FROM THE FRONTLINE