Novel Strategy to Decrease Reperfusion Injuries and Improve Function of Cold-Preserved Livers Using Normothermic Ex Vivo Liver Perfusion Machine

Transcription

1 LIVER TRANSPLANTATION 22: , 2016 ORIGINAL ARTICLE Novel Strategy to Decrease Reperfusion Injuries and Improve Function of Cold-Preserved Livers Using Normothermic Ex Vivo Liver Perfusion Machine Babak Banan, 1 * Zhenyu Xiao, 1 * Rao Watson, 2,3 Min Xu, 1 Jianluo Jia, 1 Gundumi A. Upadhya, 1 Thalachallour Mohanakumar, 1,2 Yiing Lin, 1 and William Chapman 1 Departments of 1 Surgery and 2 Pathology and Immunology, School of Medicine, Washington University, St. Louis, MO; and 3 Department of Pathology and Laboratory Medicine, School of Medicine and Public Health, University of Wisconsin, Madison, WI Normothermic extracorporeal liver perfusion (NELP) can decrease ischemia/reperfusion injury to the greatest degree when cold ischemia time is minimized. Warm perfusion of cold-stored livers results in hepatocellular damage, sinusoidal endothelial cell (SEC) dysfunction, and Kupffer cell activation. However, the logistics of organ procurement mandates a period of cold preservation before NELP. The aim of this study was to determine the beneficial effects of gradual rewarming of cold-stored livers by placement on NELP. Three female porcine livers were used for each group. In the immediate NELP group, procured livers were immediately placed on NELP for 8 hours. In the cold NELP group, livers were cold-stored for 4 hours followed by NELP for 4 hours. In rewarming groups, livers were cold-stored for 4 hours, then gradually rewarmed in different durations to 388C and kept on NELP for an additional 4 hours. For comparison purposes, the last 4 hours of NELP runs were considered to be the evaluation phase. Immediate NELP livers had significantly lower concentrations of liver transaminases, hyaluronic acid, and b-galactosidase and had higher bile production compared to the other groups. Rewarming livers had significantly lower concentrations of hyaluronic acid and b-galactosidase compared to the cold NELP livers. In addition, there was a significant decline in international normalized ratio values, improved bile production, reduced biliary epithelial cell damage, and improved cholangiocyte function. Thus, if a NELP machine is not available at the procurement site and livers will need to undergo a period of cold preservation, a gradual rewarming protocol before NELP may greatly reduce damages that are associated with reperfusion. In conclusion, gradual rewarming of cold-preserved livers upon NELP can minimize the hepatocellular damage, Kupffer cell activation, and SEC dysfunction. Liver Transpl 22: , VC 2015 AASLD. Received July 22, 2015; accepted September 30, Normothermic extracorporeal liver perfusion (NELP) has created a new paradigm in the liver preservation field. Growing interest in using this system is based on studies that have shown the efficacy and superiority of warm perfusion over standard static cold storage of the liver. 1-3 NELP attenuates ischemia/reperfusion Abbreviations: ALP, alkaline phosphatase; ELISA, enzyme-linked immunosorbent assay; H & E, hematoxylin-eosin; HTK, histidine tryptophan ketoglutarate; INR, international normalized ratio; IRI, ischemia/reperfusion injury; LDH, lactate dehydrogenase; NELP, normothermic extracorporeal liver perfusion; SD, standard deviation; SEC, sinusoidal endothelial cell. These authors contributed equally to this work. This project was funded by the Barnes Jewish Hospital Foundation Project Award, Transplant Research Support. Potential conflict of interest: Nothing to report. Address reprint requests to William Chapman, M.D., F.A.C.S., Department of Surgery, School of Medicine, Washington University, 660 South Euclid Avenue, Campus Box 8109, St. Louis, MO Telephone: ; FAX: ; chapmanwi@wudosis.wustl.edu DOI /lt View this article online at wileyonlinelibrary.com. LIVER TRANSPLANTATION.DOI /lt. Published on behalf of the American Association for the Study of Liver Diseases VC 2015 American Association for the Study of Liver Diseases.

2 334 BANAN ET AL. LIVER TRANSPLANTATION, March 2016 injury (IRI), improves hemodynamics of the liver, and recovers grafts insulted with warm ischemia (eg, donation after cardiac death). 4-6 In addition, NELP reduces hepatocellular damage and maintains normal bile production 7,8 along with improving liver transplant survival in animal models. 5 Although ischemia leads to tissue injury via depletion of cellular energy sources, the impact of cold preservation induced graft injury is manifested mainly during the initial phase of reperfusion. 9,10 In other words, the process of transplant-induced IRI is initiated during the cold preservation period. Cold storage causes changes in the ultrastructure of Kupffer cells, and reperfusion leads to activation and subsequent release of cytotoxic metabolites that has been proposed to lead to microvascular injury that hinders liver viability. 11 Therefore, inhibition of reperfusion-induced Kupffer cell activation minimizes hepatocellular damage and improves allograft survival following transplantation. 12,13 Liver sinusoidal endothelial cells (SECs), which line the walls of sinusoids and separate hepatocytes from blood flowing through the liver, play an important role in the hepatic microcirculation. Ischemia leads to serious morphological changes in SECs during the cold preservation period, and subsequent warm reperfusion augments this process, leading to complete denudation of the SECs lining Significant SEC injury provokes graft malmicrocirculation, vasoconstriction, platelet activation, inflammatory cells infiltration, and Kupffer cell activation upon reperfusion Also, it has been shown that minimizing SEC damage during the cold preservation period ameliorates IRI and improves survival outcomes. 20 The temperature transition during NELP is important for improving graft viability following liver transplantation. Large temperature changes of more than 308C occur between organ procurement, cold preservation, and graft reperfusion. It has been shown that the ideal liver preservation happens when the liver is connected to the NELP machine immediately upon procurement. Even under these circumstances, a short period of cold ischemia time leads to significant SEC dysfunction and Kupffer cell activation. 21 Therefore, the beneficial effects of NELP can be hindered by cold storage of the liver before warm perfusion. However, the logistics of clinical liver procurement mandates a brief period of cold preservation before normothermic perfusion. In this study, we hypothesized that SEC dysfunction and Kupffer cell activation may in part be due to the rapid and large temperature changes that occur with perfusing warm blood in a cold-preserved liver. We show that gradual rewarming of a cold-preserved liver is a better option and protects hepatocytes and diminishes SEC dysfunction and Kupffer cell activation. Therefore, in clinical scenarios in which NELP is not available at the procurement site, livers will need to undergo a period of cold storage; a gradual rewarming protocol employed in this process before normothermic perfusion may greatly reduce injuries that are associated with reperfusion. MATERIALS AND METHODS Animals Eighteen female land race/farm pigs (30-35 kg) were purchased from Oak Hill Genetics (Ewing, IL). All animals were housed and maintained in accordance with the National Resource Council guidelines. All experimental procedures and protocols were approved by the Animal Studies Committee and Department of Comparative Medicine at the School of Medicine, Washington University, St Louis, MO. The animals were given access to food and water ad libitum and were restricted from solid food but not water 12 hours before the operation. Liver Procurement Animals were premedicated with mg/kg of buprenorphine and a drug cocktail (4 mg/kg of telazol, 2 mg/kg of ketamine, and 2 mg/kg of xylazine) intramuscularly. They were intubated and anesthetized with 1% isoflurane (induction and maintenance), 50% nitrous oxide, and oxygen, and were maintained on full anesthetic support. Through a midline abdominal approach, the access to the liver for standard dissection and isolation was performed, and the abdominal aorta was cannulated. The animal was heparinized (300 U/kg), and whole blood was collected in blood collection bags (Jorgensen Laboratories, Inc., Loveland, CO). The supraceliac and infrarenal aorta were then ligated, the portal vein was cannulated, and cardiac arrest was induced using KCl ( mg/kg). The livers were flushed with 3 L (1 L via aorta, 2 L via portal vein) of custodial histidine tryptophan ketoglutarate (HTK) solution (Dr. Franz Kohler Chemie Co., Bensheim, Germany) at 48C. The livers were prepared and cannulated on the back table. The cold storage before NELP did not exceed 15 minutes. Ex Vivo Normothermic Liver Perfusion System The circuit was primed with 700 ml of normal saline and 10,000 U heparin. Blood was then added ( ml) to the circuit. The NELP circuit was run for 30 minutes to allow the dialyzer to achieve equilibrium before placement of the liver onto the system (Table 1). The perfusion system (Fig. 1A) was a single recirculating circuit which consisted of the following: 1 centrifugal pump (CBBP50 Carmeda pediatric Bio- Pump; Medtronic Inc., Minneapolis, MN), a liver chamber that served as a blood reservoir, a membrane oxygenator/heat exchanger (CB3381 Carmeda Pediatric Minimax Plus; Medtronic Inc., Dublin, Ireland), 2 flow probes (Pediatric Bio-Probe; Medtronic Inc.), and a dialyzer (Fresenius 2008K). Perfusate temperature was controlled by a thermoelectric water pump (Polystat; Cole- Parmer, Vernon Hills, IL) connected to the integrated

3 LIVER TRANSPLANTATION, Vol. 22, No. 3, 2016 BANAN ET AL. 335 TABLE 1. Blood Gas Analysis and Chemistry at the Beginning of the Evaluation Phase in the Study Groups Immediate Cold 20-Minute 30-Minute 60-Minute 120-Minute Parameter NELP NELP Rewarming Rewarming Rewarming Rewarming ph pco 2, mm Hg po 2, mm Hg Hct, % Na 11, mmol/l K 1, mmol/l CL, mmol/l HCO3, mmol/l Ca 11, mg/dl Glucose, mg/dl NOTE: Data are given as mean 6 SD. The dialysis machine was connected to the system 30 minutes before connecting the livers to the machine perfusion. The dialysis machine insured stabilized blood chemistry, ph, and electrolytes throughout the perfusion sessions. Figure 1. (A) Study design and (B) liver perfusion circuit. Nonrewarming livers: in the immediate NELP group, livers were procured and immediately connected to the NELP and perfused for 8 hours. In the cold NELP group, livers were procured and stored at 48C for 4 hours followed by warm perfusion with NELP for an additional 4 hours. Rewarming livers: livers were procured and preserved the same as the cold NELP group and rewarmed in different durations and kept on NELP for a total of 4 hours of perfusion. The warm perfusion circuit contains 1 centrifugal pump and a dialyzer. The dialyzate was set to K 5 3 mmol/l, Ca 5 3 mmol/l, and flow of 300 ml/ minute. To simulate enterohepatic recirculation, produced bile was reinjected into the perfusate every hour. Blood samples were taken at hourly intervals.

4 336 BANAN ET AL. LIVER TRANSPLANTATION, March 2016 heat exchanger of the oxygenator, and heparin was infused at 500 U/hour. Before placement on NELP, the liver was flushed with 2 L of saline via the hepatic artery and portal vein. The liver was covered with a thin plastic sheet to prevent excess evaporation. The flows were maintained at L/minute for the hepatic artery and L/minute for the portal vein to achieve constant arterial pressures of mm Hg and portal pressures of mm Hg. The membrane oxygenator was supplemented with 95% O 2 and 5% CO 2. The flows and pressures were constantly monitored, and the hematocrit was kept between 20% and 25% by ultrafiltration of excess fluid by dialysis. The perfusate was supplemented with Clinimix E4.25/5 (Baxter, IL) at 7 ml/hour, prostacyclin (GlaxoSmithKline Inc., Verona, Italy) at 10 mg/hour, and insulin at 10 U/ hour. Assessment of the Liver Samples for blood gas analysis and hepatic enzymes were taken hourly. The oxygen content (C) was calculated using the formula: C a/v O 2 5 Hb S a/v O 2. Oxygen extraction ratio was then calculated using Fick s equation: (C a O 2 C v O 2 )/C a O 2. Liver biopsies were sent for hematoxylin-eosin (H & E) staining. A pathologist, blinded to the experimental conditions, reviewed the slides for hepatocyte morphology, necrosis, and level of IRI. Bile production was assessed every hour and was evaluated for concentrations of lactate dehydrogenase (LDH), alkaline phosphatase (ALP), bicarbonate, and glucose, as well as ph. International normalized ratio (INR; CoaguChek XS system; Roche Diagnostics, Berlin, Germany) and lactate (Lactate plus; Nova Biomedical, Runcorn, UK) levels were also measured at hourly intervals. Enzyme-linked Immunosorbent Assay (ELISA) b-galactosidase enzyme has been previously used as a marker of Kupffer cell activation. 21 In addition, concentration of hyaluronic acid enzyme has been used as a marker of SEC damage. 21,22 We assessed the Kupffer cells activation and SEC damage by measuring concentrations of these enzymes in the perfusate. Blood samples were taken at hourly intervals and used for ELISA according to the manufacturer s protocol for porcine b-galactosidase (My Biosource, San Diego, CA) and porcine hyaluronic acid (Echelon Biosciences, Inc., Salt Lake City, UT). Statistical Analysis GraphPad prism, version 5 (GraphPad Software, Inc., La Jolla, CA) software was used to generate graphs. A 2-way analysis of variance test for scatterplots, and t test for columnar graphs were carried out for direct comparison of the study groups. P value < 0.05 were considered to be significant. Data are presented as mean 6 standard deviation (SD). RESULTS Rewarming Technique Is Safe, Feasible, and Does Not Lead to Further Hepatocyte Injury Following Reperfusion Animals were randomly assigned to 6 groups (3 livers per group; Fig. 1B): in the immediate NELP group, livers were procured and immediately perfused (cold ischemia time < 15 minutes) with NELP for 8 hours. In the cold NELP group, livers were procured and stored at 48C for 4 hours followed by warm perfusion with NELP for 4 hours. In the rewarming groups, livers were procured and stored at 48C for 4 hours. Then, they were divided into 4 subgroups: they were connected to the NELP system and gradually rewarmed to 388C, over different time periods: 120-, 60-, 30-, and 20-minute rewarming. The total normothermic machine perfusion period after rewarming was 4 hours. Three livers were used for each group of this study. For comparison purposes, the data obtained during the 4 hours of machine perfusion of the cold NELP group and rewarming groups were compared to data from the last 4 hours of the immediate NELP group. Because the oxygen-hemoglobin dissociation curve shifts to the left in low temperatures, to overcome increased hemoglobin affinity toward oxygen, the blood was mildly acidified by the dialyzer, and ph was set to during the rewarming period. The oxygen extraction ratio was used as the marker of oxygen utilization by the livers during the evaluation phase. The O 2 extraction ratio was recovered above 50% in the first hour of warm perfusion in the cold NELP and 20- and 30-minute rewarming livers. However, 60- and 120-minute rewarming livers revealed lower extraction ratios by the end of the evaluation phase: and , respectively (Fig. 2A). Oxygen extraction ratio remained above 80% in the immediate NELP group during the evaluation phase (data not shown). To assess the hepatic vascular response to gradual rewarming, the NELP circuit was designed as a pressure-controlled system. Hence, the flows were adjusted to keep the pressure within the physiologic ranges for the hepatic artery (75-85 mm Hg) and the portal vein (5-7 mm Hg). In all groups (Fig. 2B), the resistance was higher at the beginning of the evaluation phase and lower at the end of the experiments. Although the pressures were within physiologic ranges, comparison of the data showed significantly lower hepatic artery resistance in the immediate NELP group ( mm Hg/mL/minute; P ) and the 20-minute rewarming group ( mm Hg/mL/minute; P ) when compared to the cold NELP group ( mm Hg/mL/minute) at the end of the evaluation phase. Lactate was measured as a marker of anaerobic metabolism in the grafts. Our data indicated that gradual rewarming over 30 minutes led to a significant increase in lactate levels (Fig. 3A). Livers that

5 LIVER TRANSPLANTATION, Vol. 22, No. 3, 2016 BANAN ET AL. 337 Figure 2. (A) Oxygen extraction ratio and (B) vascular resistance during NELP. The oxygen extraction ratio was recovered in 20- and 30- minute rewarming groups after blood temperature reached the normothermic levels. The 60- and 120-minute rewarming livers did not reveal satisfactory oxygen extraction ratios during the evaluation phase. Generally, vascular resistance diminished in all study groups, and the reduction in resistance was significant in immediate NELP and 20-minute rewarming groups when compared to cold NELP livers. were rewarmed over 60 and 120 minutes finished the 4 hours of NELP runs with elevated lactate values ( and mmol/l, respectively), suggesting that after prolonged rewarming periods anaerobic metabolism continues despite perfusion with normothermic blood. Lactate levels dropped dramatically in the immediate NELP group in the first 30 minutes of warm perfusion. In the cold NELP group, although the mean starting lactate concentrations were elevated ( mmol/l), they decreased after 1 hour of warm perfusion. Liver enzymes were measured to evaluate hepatocyte damage. Immediate NELP group livers had the most stable hepatic enzyme levels throughout the experiments and finished the study with significantly lower concentrations of liver enzymes when compared to cold NELP livers (Fig. 3B): AST ( versus U/L; P < 0.001) and ALT (43 6 3versus U/L; P ). Comparison of ALT levels between the cold NELP group and rewarmed livers showed statistically significant lower ALT in the 20- minute rewarming ( U/L; P ), 30-minute rewarming ( U/L; P ), and 60-minute rewarming ( U/L; P ) groups, but not in the 120-minute rewarming group ( U/L; P > 0.05). Comparison of rewarmed livers with the immediate NELP group did not show statistically significant differences in ALT values. Comparing final AST concentrations between the cold NELP group and rewarmed livers did not reveal any significant differences. These data suggest that rewarming of a cold-preserved liver is safe and does not further extend the injuries to the hepatocytes from the cold preservation period. Rewarming Led to Lower Kupffer Cell Activation and SECs Damage b-galactosidase is a lysosomal enzyme that is mainly released by Kupffer cells following activation. 21,23 Using the ELISA assay to measure hourly b- galactosidase levels in the perfusate, we found that immediate NELP livers had low steady concentrations of b-galactosidase in the perfusate and completed the study with significantly lower enzyme levels when compared to cold NELP livers (23 6 9versus ng/ ml; P < 0.001; Fig. 4A). Rewarmed livers, however, revealed a different pattern of enzyme release when compared to the cold NELP group: in the 20-minute rewarming group, the enzyme peaked at 90 minutes of NELP ( ng/ml) and gradually decreased until the end of the experiment ( ng/ml; P < 0.001). In the 30-minute rewarming group, the enzyme peaked at 120 minutes of NELP and gradually decreased the end of the study ( ng/ml; P < 0.001). The 60-

6 338 BANAN ET AL. LIVER TRANSPLANTATION, March 2016 Figure 3. (A) Concentrations of lactate and (B) liver enzymes during the evaluation phase. Lactate levels diminished significantly in all study groups except 60- and 120-minute rewarming livers. Cold NELP livers had the highest concentrations of AST and ALT at the end of the evaluation phase. Figure 4. Concentrations of (A) b-galactosidase, (B) hyaluronic acid, (C) bile production, and (D) INR during evaluation phase. Kupffer cell activation and endothelial injury were significantly diminished by the rewarming technique. Production of coagulation factors and bile were significantly higher in immediate NELP and 20-minute rewarming groups. Long periods of rewarming led to lower bile production and higher INR levels in the 60- and 120-minute rewarming groups.

7 LIVER TRANSPLANTATION, Vol. 22, No. 3, 2016 BANAN ET AL. 339 minute rewarming livers peaked at 60 minutes of warm perfusion ( ng/ml) and completed the experiments with remarkably lower enzyme concentrations ( ng/ml; P < 0.001). Interestingly, the 120-minute rewarming group had the lowest overall enzyme levels among rewarmed groups: enzyme levels peaked at 120 minutes of NELP and significantly decreased to ng/ml (P < 0.001) by the end of the study. These data indicate that gradual rewarming of a cold-preserved liver significantly reduces Kupffer cell activation after starting warm perfusion. Hyaluronic acid is a matrix polysaccharide enzyme that is known to be associated with inflammation 24 and has been used as a marker of SEC damage. 21,22 Using ELISA assays, we measured concentrations of hyaluronic acid in the perfusate at hourly intervals. Immediate NELP livers had constantly lower concentrations throughout the experiments and completed the study with significantly lower hyaluronic acid concentrations when compared to the cold NELP group ( versus ng/ml; P < 0.001; Fig. 4B). In the cold NELP and rewarmed groups, enzyme release was diminished during warm perfusion nevertheless. Generally, rewarmed livers had significantly lower enzyme release by the end of the experiments when compared to the cold NELP group; however, 60- and 120-minute rewarming livers revealed the lowest concentrations: 20-minute rewarming ( ng/ ml; P > 0.05), 30-minute rewarming ( ng/ml; P < 0.001), 60-minute rewarming ( ng/ml; P < 0.001), and 120-minute rewarming ( ng/ ml; P < 0.001). These data suggest that the slower rate of rewarming period leads to a significant reduction in cold-induced SEC damage. Fast Rewarming Led to Better Bile and Coagulation Factors Production When Compared to Slow Rewarming Immediate NELP livers started making bile almost immediately upon placement on NELP and maintained remarkably high bile production during the study. At the end of the experiments, immediate NELP livers had significantly higher bile production when compared to the cold NELP group: versus ml/hour (P < 0.001; Fig. 4C). In rewarming livers, a short period of rewarming yielded better bile production when compared to long rewarming periods. However, comparison of bile production of cold NELP livers at the end of the experiments with 20- and 30-minute rewarming did not show statistically significant differences: 20-minute rewarming ( ml/hour; P > 0.05) and 30-minute rewarming ( ml/hour; P > 0.05). The 60- and 120-minute rewarming livers produced remarkably lower bile during warm perfusion when compared to the cold NELP group: ml/hour (P ) and ml/hour (P < 0.001), respectively. Production of coagulation factors was assessed by measuring INR. The immediate NELP group started the experiments with INR of and completed the study with significantly lower INR when compared to the cold NELP group ( versus ; P < 0.001; Fig. 4D). Among the gradual rewarming livers, only the 20-minute rewarming groups completed the experiments with significantly lower INR compared to the cold NELP group: 20-minute rewarming ( ; P < 0.001), 30-minute rewarming ( ; P > 0.05), 60-minute rewarming ( ; P > 0.05), and 120-minute rewarming ( ; P > 0.05). Short Period of Rewarming Preserved the Function of Cholangiocytes and Led to Less Biliary Epithelial Injury Biliary epithelial cell function was evaluated by assessing ph, bicarbonate, and glucose concentrations in produced bile by the livers. In addition, the degree of damage to the epithelial cells was evaluated by measuring ALP and LDH concentrations in the bile. 25,26 Our data revealed overall higher cholangiocyte function as well as less damage in the 20- and 30-minute rewarming liver group when compared to the cold NELP group. At the end of the evaluation phase, there were no significant differences in biliary ph: (P > 0.05) for 20-minute rewarming and (P > 0.05) for the 30-minute rewarming and the cold NELP group ( ). However, the results showed remarkable differences in glucose and bicarbonate concentrations when compared to the cold NELP group (Fig. 5A). Biliary glucose values were as follows: immediate NELP ( mg/dl; P ), 20-minute rewarming ( mg/dl; P ), 30-minute rewarming ( mg/dl; P > 0.05), and cold NELP ( mg/dl) groups. Biliary bicarbonate concentrations were as follows: immediate NELP ( mmol/l; P ), 20-minute rewarming ( mmol/l; P ), 30-minute rewarming ( mmol/l; P ), and cold NELP ( mmol/l) groups. Markers of epithelial injury were significantly lower in the immediate NELP and 20-minute rewarming groups when compared to cold NELP livers. However, the last 2 groups revealed significantly higher biliary concentrations of ALP and LDH at the end of the evaluation phase (Fig. 5B): ALP levels were U/L (P < 0.001) for immediate NELP, U/L (P ) for 20-minute rewarming, U/L (P > 0.05) for 30-minute rewarming, U/L (P ) for 60-minute rewarming, U/L (P < 0.001) for 120-minute rewarming, and U/L for cold NELP groups. LDH levels were U/L (P < 0.001) for immediate NELP, U/L (P ) for 20-minute rewarming, U/L (P > 0.05) for 30-minute rewarming, U/L (P ) for 60-minute rewarming, U/L (P ) for 120-minute rewarming, and U/L for cold NELP groups. These data suggest that gradual rewarming over minutes can decrease biliary epithelial damage in cold-preserved livers.

8 340 BANAN ET AL. LIVER TRANSPLANTATION, March 2016 Figure 5. Biliary measures of (A) ph, glucose, and bicarbonate and (B) concentrations of ALP and LDH in the bile at the end of the evaluation phase. Gradual rewarming in 20 minutes significantly improved markers of cholangiocyte function. Immediate NELP and 20-minute rewarming groups had significantly lower ALP and LDH at the end of the evaluation phase when compared to the cold NELP group. Gradual Rewarming Led to Significant IRI Reduction After Warm Perfusion With NELP H & E staining of immediate NELP livers showed little evidence of necrosis or IRI in the liver sections. Cold NELP liver sections showed confluent necrosis with mild IRI, but tissue obtained from rewarmed livers uniformly had little IRI or hepatocellular necrosis (Fig. 6). DISCUSSION In this study, we compared the outcomes of warm perfusion of cold-preserved livers with a gradual rise in temperature in different time periods. This is the first study of its kind that investigates strategies to minimize damage to SECs and Kupffer cells following graft reperfusion. We demonstrated that liver injuries resulting from warm perfusion after cold storage were significantly reduced by a gradual rewarming technique. The idea of rewarming cold-preserved organs originated from studies that had shown the efficacy of oxygenated perfusion of cold-stored livers at either hypothermic or gradual rewarming to subnormothermic temperatures before graft reperfusion We postulated that a gradual increase in the perfusate temperature will maintain graft integrity and decrease reoxygenation injury in grafts with ischemic injuries. Given that the extent of reperfusion injury is dependent on the duration of the cold preservation period, 30 maintaining a lower temperature during the initial phase of reperfusion minimizes oxidative stress and enables replenishment of cellular energy sources which may protect the graft against tissue injuries during a second warm ischemia insult that happens during implantation of the graft into the recipient. 35 A recent study using an ex vivo porcine heart model of machine perfusion demonstrated that gradual rewarming of blood to normothermic temperature in 30 minutes improves graft viability and function. 36 In another study, porcine livers were preserved in cold storage for 18 hours and perfused with oxygenated HTK at either hypothermic (48C) or subnormothermic (208C) conditions. In the latter group, the livers were gradually rewarmed from 48C to 208C in 90 minutes. Results from this study showed that gradually rewarmed livers had significantly lower cellular enzyme loss and histological markers of injury accompanied by improved bile production. 29 In our study, we gradually rewarmed the livers to normothermic temperatures and showed that rewarming of coldpreserved livers (in minutes) is not only safe, but it also improves synthetic functions of the livers with less injuries to the SECs and Kupffer cells. Hepatocyte function is reduced at low temperatures; therefore, in hypothermic and subnormothermic machine perfusion systems, the liver can be perfused with solutions without oxygen carriers. Because of a substantial increase in the metabolism rate at

9 LIVER TRANSPLANTATION, Vol. 22, No. 3, 2016 BANAN ET AL. 341 Figure 6. H & E staining of liver sections obtained at the end of the evaluation phase (magnification 3 400): (A) immediate NELP, (B) 20-minute rewarming, (C) 30-minute rewarming, (D) cold NELP, (E) 60-minute rewarming, and (F) 120-minute rewarming. In the immediate NELP and 20-minute rewarming groups, liver biopsies revealed no evidence of IRI at the end of the evaluation phase. The 30-minute rewarming livers showed no zone 3 abnormalities. The cold NELP group had confluent zone 3 necrosis without bridging consistent with mild IRI. The 60- and 120- minute rewarming livers uniformly revealed no evidence of IRI at the end of the evaluation phase. normothermic conditions, livers should be perfused with an oxygen carrier. We achieved this with the use of donor blood as the perfusate in the NELP circuit. In low temperatures, the oxygen-hemoglobin curve shifts to the left. This change increases the affinity of hemoglobin toward the oxygen molecule and subsequently lowers the oxygen delivery to the tissue. To balance this physiologic phenomenon, we acidified the perfusate by higher acid/bicarbonate ratios of the dialysate fluid and decreased the blood ph to However, although acidification of the blood theoretically shifts the curve back to the right, this counterbalance was not efficient enough to neutralize hemoglobin affinity toward O 2 during the rewarming period. Hence, because of lower oxygenation of the liver tissues (as evident by lower oxygen extraction ratio) in 60- and 120-minute rewarming livers, the anaerobic metabolism rate was substantially high during the rewarming period (as evident by high lactate levels). Therefore, 60- and 120-minute rewarming livers did not produce satisfactory bile quantities and had remarkably higher INR and degrees of injury to the biliary epithelial cells at the end of the evaluation phase. Previous studies have shown the adverse effects of cold preservation on procured livers. In 1 study, porcine livers underwent 60 minutes of warm ischemia followed by either 24 hours of NELP or a combination of 1 hour of cold storage and 23 hours of NELP. One hour of cold preservation led to remarkable hepatocellular damage, Kupffer cell activation, and SEC dysfunction. 21 However, in our study, warm perfusion of cold-stored livers without the rewarming technique led to significant graft dysfunction as evidenced by high INR, low biliary ph and bicarbonate, and reduced bile production. In contrast, rewarming of livers in 20 and 30 minutes led to significant bile production, INR reduction, and lower markers of biliary epithelial damage (Figs. 4 and 5). We also measured concentrations of tumor necrosis factor a as a marker of Kupffer cell activation but did not find considerable differences between the study groups (data not shown). This may be due to dysregulated immune responses on the machine perfusion settings that disrupt classical cell signaling pathways. It has been shown that perfusion of the liver with low perfusate pressures prevents endothelial damage, independent of the oxygen content of the perfusate. 37 Furthermore, reduction of hepatic artery pressure and reduced vascular resistance at constant flow rates during machine perfusion has been linked to increased graft viability. 38 Our results extend these findings and confirm that reduction of vascular resistance and SEC damage in cold-preserved livers is achievable by the gradual rewarming technique (Fig. 2B). Our results are also consistent with previous studies that have shown the high sensitivity of SECs to cold temperatures. 19 Moreover, we show that to reduce endothelial cell damage upon liver perfusion, a brief period of rewarming with oxygenated blood is necessary. Although all coldpreserved livers had reduced markers of SEC damage through the evaluation phase, longer periods of rewarming (60 and 120 minutes) showed even greater reduction in SEC damage as evident by the remarkably low concentrations of hyaluronic acid by the end of the evaluation phase (Fig. 4B). The function of cholangiocytes has been previously tested by measuring biliary concentrations of bicarbonate, glucose, and ph. Moreover, high biliary bicarbonate concentrations have been previously linked to better hepatocyte function. 39 Because bile production is the result of an intact interrelationship between hepatocytes, sinusoidal cells, and biliary epithelial cells, bile production has been used as a marker that integrates all of these measures for an indicator of graft viability. Our results presented in Fig. 5 indicate that rewarming of cold-stored livers for minutes significantly improves hepatocyte and cholangiocyte function and lowers injuries associated with cold preservation of the liver. In summary, our results clearly demonstrate that warm preservation of the liver is superior to standard cold storage. Furthermore, our results show that a brief period of cold ischemia followed by a rewarming period of minutes before NELP can significantly

10 342 BANAN ET AL. LIVER TRANSPLANTATION, March 2016 reduce SEC injury and Kupffer cell activation leading to improvement in liver graft function. REFERENCES 1. 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: Schlegel A, Kron P, Graf R, Dutkowski P, Clavien PA. Warm vs. cold perfusion technique to rescue rodent liver grafts. J Hepatol 2014;61: Xu H, Berendsen T, Kim K, Soto-Gutierrez A, Bertheium F, Yarmush ML, Hertl M. Excorporeal normothermic machine perfusion resuscitates pig DCD livers with extended warm ischemia. J Surg Res 2012;173:e83-e Vogel T, Brockmann JG, Coussios C, Friend PJ. The role of normothermic extracorporeal perfusion in minimizing ischemia reperfusion injury. Transplant Rev (Orlando) 2012;26: 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: Izamis ML, Tolboom H, Uygun B, Berthiaume F, Yarmush ML, Uygun K. Resuscitation of ischemic donor livers with normothermic machine perfusion: a metabolic flux analysis of treatment in rats. PLoS One 2013;8: e op den Dries S, Karimian N, Sutton ME, Westerkamp AC, Nijsten MW, Gouw AS, et al. Ex vivo normothermic machine perfusion and viability testing of discarded human donor livers. Am J Transplant 2013;13: Imber CJ, St Peter SD, de Cenarruzabeitia IL, Lemonde H, Rees M, Butler A, et al. Optimization of bile production during normothermic preservation of porcine livers. Am J Transplant 2002;2: Menger MD, Lehr HA, Messmer K. Role of oxygen radicals in the microcirculatory manifestations of postischemic injury. Klin Wochenschr 1991;69: Lemasters JJ. V. Necrapoptosis and the mitochondrial permeability transition: shared pathways to necrosis and apoptosis. Am J Physiol 1999;276(pt 1):G1-G Southard JH, Lindell S, Ametani M, Richer JP, Vos AW. Kupffer cell activation in liver preservation: cold storage vs machine perfusion. Transplant Proc 2000;32: Takei Y, Marzi I, Kauffman FC, Currin RT, Lemasters JJ, Thurman RG. Increase in survival time of liver transplants by protease inhibitors and a calcium channel blocker, nisoldipine. Transplantation 1990;50: Kukan M, Vajdova K, Horecky J, Nagyova A, Mehendale HM, Trnovec T. Effects of blockade of Kupffer cells by gadolinium chloride on hepatobiliary function in cold ischemia-reperfusion injury of rat liver. Hepatology 1997; 26: Stolz DB, Ross MA, Ikeda A, Tomiyama K, Kaizu T, Geller DA, Murase N. Sinusoidal endothelial cell repopulation following ischemia/reperfusion injury in rat liver transplantation. Hepatology 2007;46: Caldwell-Kenkel JC, Currin RT, Tanaka Y, Thurman RG, Lemasters JJ. Reperfusion injury to endothelial cells following cold ischemic storage of rat livers. Hepatology 1989;10: McKeown CM, Edwards V, Phillips MJ, Harvey PR, Petrunka CN, Strasberg SM. Sinusoidal lining cell damage: the critical injury in cold preservation of liver allografts in the rat. Transplantation 1988;46: Jaeschke H. Molecular mechanisms of hepatic ischemiareperfusion injury and preconditioning. Am J Physiol Gastrointest Liver Physiol 2003;284:G15-G Selzner N, Rudiger H, Graf R, Clavien PA. Protective strategies against ischemic injury of the liver. Gastroenterology 2003;125: Ikeda T, Yanaga K, Kishikawa K, Kakizoe S, Shimada M, Sugimachi K. Ischemic injury in liver transplantation: difference in injury sites between warm and cold ischemia in rats. Hepatology 1992;16: Ikeda A, Ueki S, Nakao A, Tomiyama K, Ross MA, Stolz DB, et al. Liver graft exposure to carbon monoxide during cold storage protects sinusoidal endothelial cells and ameliorates reperfusion injury in rats. Liver Transpl 2009;15: Reddy S, Greenwood J, Maniakin N, Bhattacharjya S, Zilvetti M, Brockmann J, et al. Non-heart-beating donor porcine livers: the adverse effect of cooling. Liver Transpl 2005;11: Copple BL, Banes A, Ganey PE, Roth RA. Endothelial cell injury and fibrin deposition in rat liver after monocrotaline exposure. Toxicol Sci 2002;65: Liu W, Sch ob O, Pugmire JE, Jackson D, Zucker KA, Fry DE, Glew RH. Glycohydrolases as markers of hepatic ischemia-reperfusion injury and recovery. Hepatology 1996;24: Laurent TC, Fraser JR. Hyaluronan. FASEB J 1992;6: Op den Dries S, Sutton ME, Karimian N, de Boer MT, Wiersema-Buist J, Gouw AS, et al. Hypothermic oxygenated machine perfusion prevents arteriolonecrosis of the peribiliary plexus in pig livers donated after circulatory death. PLoS One 2014;9:e Vajdova K, Smrekova R, Kukan M, Lutterova M, Wsolova L. Bile analysis as a tool for assessing integrity of biliary epithelial cells after cold ischemia--reperfusion of rat livers. Cryobiology 2000;41: Fuller BJ, Busza AL, Proctor E. Possible resuscitation of liver function by hypothermic reperfusion in vitro after prolonged (24-hour) cold preservation--a 31P NMR study. Transplantation 1990;50: Minor T, Saad S, K otting M, Nagelschmidt M, Paul A. Endischemic oxygen persufflation to improve viability of marginally preserved donor livers. Transpl Int 1998; 11(suppl 1):S400-S Minor T, Efferz P, Fox M, Wohlschlaeger J, L uer B. Controlled oxygenated rewarming of cold stored liver grafts by thermally graduated machine perfusion prior to reperfusion. Am J Transplant 2013;13: Rauen U, Polzar B, Stephan H, Mannherz HG, de Groot H. Cold-induced apoptosis in cultured hepatocytes and liver endothelial cells: mediation by reactive oxygen species. FASEB J 1999;13: Minor T, K otting M. Gaseous oxygen for hypothermic preservation of predamaged liver grafts: fuel to cellular homeostasis or radical tissue alteration? Cryobiology 2000;40: Dutkowski P, Furrer K, Tian Y, Graf R, Clavien PA. Novel short-term hypothermic oxygenated perfusion (HOPE) system prevents injury in rat liver graft from non-heart beating donor. Ann Surg 2006;244: Minor T, Stegemann J, Hirner A, Koetting M. Impaired autophagic clearance after cold preservation of fatty livers correlates with tissue necrosis upon reperfusion and is reversed by hypothermic reconditioning. Liver Transpl 2009;15: Dutkowski P, Graf R, Clavien PA. Rescue of the cold preserved rat liver by hypothermic oxygenated machine perfusion. Am J Transplant 2006;6(pt 1): Vajdova K, Graf R, Clavien PA. ATP-supplies in the coldpreserved liver: A long-neglected factor of organ viability. Hepatology 2002;36:

11 LIVER TRANSPLANTATION, Vol. 22, No. 3, 2016 BANAN ET AL White CW, Hasanally D, Mundt P, Li Y, Xiang B, Klein J, et al. A whole blood-based perfusate provides superior preservation of myocardial function during ex vivo heart perfusion. J Heart Lung Transplant 2015;34: Schlegel A, Rougemont Od, Graf R, Clavien PA, Dutkowski P. Protective mechanisms of end-ischemic cold machine perfusion in DCD liver grafts. J Hepatol 2013;58: Shigeta T, Matsuno N, Obara H, Kanazawa H, Tanaka H, Fukuda A, et al. Impact of rewarming preservation by continuous machine perfusion: improved post-transplant recovery in pigs. Transplant Proc 2013;45: Sutton ME, op den Dries S, Karimian N, Weeder PD, de Boer MT, Wiersema-Buist J, et al. Criteria for viability assessment of discarded human donor livers during ex vivo normothermic machine perfusion. PLoS One 2014; 9:e