In response to the growing demand for liver transplantation, there has been interest in expanding the donor pool through the use of non–heart-beating donors (NHBD). By some estimates, an additional 1,000 donors per year could be achieved from this source (1–3). The use of NHBD hepatic grafts has been approached with trepidation because of concerns that donor warm ischemic time when compounded by subsequent cold preservation and reperfusion injury could negatively impact graft function.
The adverse outcomes of prolonged donor warm ischemic time on graft survival are well recognized in animal models and human recipients of uncontrolled NHBD livers (4–6). However, the implications of shorter periods of donor warm ischemic time in controlled NHBD grafts are not well delineated. Data from several small single-center studies suggest that controlled NHBD grafts may experience more complications than grafts from heart-beating donors (HBD), such as shorter survival, higher rates of primary nonfunction, a greater frequency of rejection and cholestasis, and possibly an increased incidence of hepatic artery thrombosis (6–8). Although these findings are based on a small number of patients, they may stem from an increased generation of reactive oxygen species associated with donor warm ischemic time. Experimental evidence indicates that toxic oxygen intermediaries impact graft survival and function (9). Administration of a vitamin E derivative and the xanthine oxidase inhibitor allopurinol appear to reverse the effects of ischemia in NHBD animal models (10,11).
The differential sensitivity of cell populations within the liver to ischemia-reperfusion injury likely influences the type of complications observed after transplantation. Biliary ducts are known to be sensitive to ischemia-reperfusion injury, and complications occurring within this system may serve as an indicator of the additional stress donor warm ischemic time incurs on the NHBD graft (12). The sensitivity of the biliary tract has been observed in an NHBD model where histologic evidence of irreversible biliary tract damage followed 40 min of warm ischemic time, despite preservation of hepatocellular function (5). In the clinical setting, the association between intrahepatic strictures and prolonged cold ischemia has been well documented after transplantation with HBD grafts (13,14). Reduced glutathione levels in bile duct cells and alterations in actin filaments along the bile canaliculi provide insight into the sensitivity and mechanisms of injury to ischemia-reperfusion in the biliary tract (15,16).
Given the premise that the biliary system may act as a sensitive indicator of ischemia-reperfusion injury, a single-center study was undertaken to review biliary complications among NHBD recipients and to compare them to a population of HBD recipients.
PATIENTS AND METHODS
Sixteen livers were procured from 15 controlled and 1 uncontrolled NHBD between 1996 and 2001. The uncontrolled NHBD donor was excluded from further analysis. A group of 221 recipients of a first hepatic allograft from HBD performed from 1998 through 2000 served as a comparison population. Children and recipients of partial, split, or living-donor grafts were excluded.
The procurement technique was similar to that described by Casavilla et al. (6). Once a decision to withdraw care had been made by family members, an evaluation by the authors’ organ procurement organization was completed. The authors selectively chose donors who were, in general, younger than 45 years of age, hemodynamically stable, and had normal hepatic biochemical parameters. The procurements took place in the operating room. After the surgeons prepared and draped, they exited the operating room, 30,000 units of heparin was administered, and the medical team taking care of the patient performed the extubation. After cessation of cardiac activity, a 5-min wait period was observed before incision to ensure absence of spontaneous cardiopulmonary function. One member of the surgical team cannulated the infrarenal aorta for infusion of University of Wisconsin solution while another clamped the thoracic aorta. Visceral dissection, portal vein perfusion with University of Wisconsin solution, and bile duct irrigation were performed in situ. The recipients did not receive vasodilators or antioxidants and the University of Wisconsin solution was prepared with insulin and steroids, which is the authors’ practice with HBD. Donor warm ischemic time was defined as the time of extubation to aortic cross-clamp.
Many of the individuals selected to receive an NHBD graft were in need of a transplant because of rapid decline in their health or because they had hepatocellular carcinoma, with little chance of transplantation. If an NHBD graft was available, recipients were informed and had the opportunity to decline. Recipient operative technique was similar for both groups. The majority of the patients underwent venovenous bypass. Except for the last year of the study, hepatic reperfusion with portal and arterial blood flow occurred in a sequential fashion (n=1). Postoperative immunosuppression consisted of prednisone and cyclosporine A or tacrolimus-based therapy.
The decision to place a T tube occurred after reperfusion and was made on the basis of the subjective quality of the liver as judged by inadequacy of bile production, instability during reperfusion, and size discrepancy between donor and recipient ducts. T-tube cholangiograms were routinely obtained at the time of transplantation, on postoperative day 7, and 3 months after transplant at the time of tube removal. For the purpose of this study, routine biliary imaging was not included in the analysis; rather, only those studies that were in addition to any regular peri- or postoperative studies were included. For patients who underwent transplantation without a T tube, any postoperative biliary imaging study or procedure was not considered routine and was included in the analysis.
Postoperative data were examined for complications that could be related to donor warm ischemia, in particular, primary nonfunction (PNF), rejection, and major biliary complications. Major biliary complications were defined as anastomotic strictures, ischemic type strictures, choledocholithiasis, or biliary cast syndrome. Hepatic artery thrombosis was evaluated by Doppler ultrasound, magnetic resonance angiography, angiography, or a combination of these modalities.
Means were compared with Student t test and proportions were compared with Fischer’s or chi-square test where appropriate. Survival rates were estimated with the Kaplan-Meier method and compared with the log-rank test. A value of P <0.05 was considered significant.
Characteristics of the NHBD and HBD donors are summarized in Table 1. NHBD were younger than HBD. They were less likely to have suffered an intracranial hemorrhage or cerebrovascular accident as a cause of brain injury; however, devastating head trauma or anoxia was more common. Use of cardiopulmonary resuscitation was greater among NHBD, but there was similar use of inotropic support before procurement. NHBD warm ischemic time ranged from 11 to 34 min, with a mean of 20.4±6 min. These grafts experienced significantly less cold ischemic time than HBD grafts.
Recipient Pretransplant Characteristics
Recipients of NHBD and HBD grafts were similar. In both groups, the majority of patients were male patients (2:1 vs. 2.1:1), with a viral cause of their liver disease (60.0% vs. 51.1%), and were United Network for Organ Sharing (UNOS) status 2B (73.3% vs. 61.8%). However, NHBD recipients had a greater incidence of hepatocellular carcinoma at the time of transplantation (33.3% vs. 9.5%, P <0.01).
Mean length of follow-up was 819±588 days among recipients of NHBD grafts and 690±345 days among recipients of HBD grafts. Mean length of stay after transplantation was similar between NHBD and HBD recipients (21.3±30.2 days vs. 16.6±16.6 days, P =0.57). Graft survival did not differ between the two groups at 1 (71.8% vs. 85.4%, P =0.23) and 3 years (71.8% vs. 73.9%, P =0.68). Patient survival at 1 year (79.0% vs. 90.9%, P =0.16) and 3 years (79.0% vs. 77.7%, P =0.8) was also not statistically different (Fig. 1).
Three NHBD recipients have died, one from recurrent hepatocellular carcinoma, another from multisystem organ failure and biliary sepsis related to diffuse intrahepatic strictures, and a third who did not undergo retransplantation after PNF associated with an irretrievable neurologic event. One recipient of an NHBD graft has undergone retransplantation for progressive diffuse intrahepatic strictures.
PNF occurred among 8 of 221 HBD recipients and all underwent retransplantation. Hepatic artery thrombosis was not detected among NHBD recipients, whereas this complication was detected in seven HBD recipients. The number of patients with biopsy-proven rejection within the first 90 days after transplant did not differ between the two groups (Table 2).
Total bilirubin and aspartate aminotransferase levels were similar at 7, 30, and 90 days after transplantation. Mean alkaline phosphatase was higher in NHBD recipients at 30 and 90 days posttransplant (320±335 IU/mL vs. 193±249 IU/mL, P =0.18; and 324±457 IU/mL vs. 140±169 IU/mL, P =0.17), although this did not reach statistical significance.
Analysis of biliary complications.
Major biliary complications occurred in 5 (33.3%) NHBD recipients and 21 (9.5%) HBD recipients (P <0.01). The frequency of these complications is demonstrated in Table 3. All NHBD grafts were anastomosed to the recipient by means of choledochocholedochostomy, whereas 211 of the HBD grafts were reconstructed with choledochocholedochostomy. T tubes were used in 10 of the 15 NHBD recipients and 133 of the 211 HBD recipients. The detection of major biliary complications among the five NHBD recipients occurred within the first 3 months of transplantation during routine postoperative T-tube cholangiograms or during radiologic evaluation for elevated hepatic enzymes.
A comparison of NHBD recipients with and without adverse biliary events demonstrates that biliary complications associated with the use of NHBD grafts were not inconsequential and resulted in multiple diagnostic and therapeutic endoscopic and percutaneous radiologic procedures (4.4±1.7 vs. 0.2±0.7, P <0.01), reoperation for anastomotic stricture (1 vs. 0), and retransplantation (1 vs. 0). Length of hospital stay within the first year of transplant was longer among those with biliary complications, but this did not reach statistical significance (77.0±70.1 days vs. 22.9±19.4 days, P =0.16). Of the four patients with diffuse intrahepatic strictures, one has required retransplantation, another died of multisystem organ failure related to recurrent bouts of biliary sepsis, a third has percutaneous transhepatic biliary tubes, and the fourth experiences intermittent elevation of hepatic enzymes. The fifth patient experienced cast syndrome and died of recurrent hepatocellular carcinoma.
Recipients of NHBD and HBD grafts with major biliary complications incurred similar degrees of morbidity and mortality. Length of hospitalization within the first year of transplant (77.0±70.0 days vs. 32.6±20.8 days, P =0.23), number of biliary endoscopic or radiologic procedures (4.4±1.7 vs. 5.0±4.2, P =0.74), grafts lost (three vs. six), retransplantation (one vs. one), and deaths (two vs. five) did not differ. These two groups also had a similar incidence of donor and recipient factors that have been implicated in the development of biliary complications, particularly intrahepatic strictures. These include hepatic artery thrombosis (zero vs. one), rejection (one vs. three), ABO incompatibility (zero vs. zero), and cold ischemic time (367.8±215.6 min vs. 429.3±92.7 min, P =0.56). The only detectable difference between these two groups was the exposure of the NHBD grafts to donor warm ischemic time (18.8±6.1 min vs. 0 min).
Allografts obtained from NHBD have been cited as an important and underused source with which to ameliorate the shortage of donor organs. However, in the current era of transplantation, the number of NHBD hepatic allografts used for transplantation is relatively small. Between 1993 and 2001, the UNOS database recorded only 191 liver transplants from NHBD donors. Furthermore, published outcomes are limited and the results are varied.
Since 1995, the University of Pennsylvania has performed 15 liver transplants from controlled NHBD donors. The authors have been concerned about the potential for allograft damage secondary to prolonged donor warm ischemia. The authors arbitrarily set a maximum donor age of 45 years to limit age-related susceptibility to ischemia and have attempted to place these grafts in relatively stable patients with limited cold ischemic times (17). This is reflected by the younger donor age and shorter cold ischemic times of the authors’ NHBD group. Given the uncertainty of long-term outcome using NHBD grafts, the authors have chosen to place them in recipients who might otherwise have little chance of receiving a transplant under the allocation rules that were in place during this study. This is reflected by the high percentage of patients who had a history of hepatocellular carcinoma.
PNF and biliary complications in addition to increased episodes of rejection may reflect manifestations of allograft damage from prolonged warm ischemia. The incidence of primary nonfunction in the authors’ population is not higher than the group that received grafts from HBD. The University of Wisconsin has reported an increased incidence of PNF among their NHBD grafts, but these occurred during their initial experience (7).
Increased ischemic time has been reported to result in increased episodes of rejection; however, episodes of biopsy-proven rejection were not different between the authors’ two groups, and were similar to what has been described by D’Alessandro et al. (7). Interestingly, Reich et al. (8) noted that 75% of their NHBD recipients developed one to two episodes of rejection, more than twice the incidence of their recipients of HBD grafts.
The high incidence of intrahepatic ischemic type biliary strictures found in the authors’ NHBD recipients has not been previously reported. These complications were responsible for a significant degree of morbidity resulting in reoperation, multiple endoscopic and percutaneous biliary interventions, retransplantation, and death. Many of these patients demonstrated progressive stricturing and beading of the intrahepatic biliary system. Intrahepatic strictures have been attributed to a number of causes, including ABO incompatibility, hepatic artery occlusion, use of Euro-Collins preservation solution, prolonged cold ischemic time, rejection, and recurrent primary sclerosing cholangitis (13,14,18–20). None proved different when compared to the authors’ HBD patients with strictures, but given the small number of NHBD recipients, it is possible that some were confounding factors in the authors’ observed complications. Donor warm ischemic time was the only factor that differed between NHBD and HBD recipients with biliary complications. The identification of diffuse intrahepatic biliary stricturing among NHBD recipients is not unique to the authors’ center. D’Alessandro et al. (7) noted that one patient required retransplantation for this complication, but the incidence was similar to HBD recipients.
Clinical and experimental data demonstrate that the biliary tract is sensitive to ischemia-reperfusion injury and, as such, an increased incidence of intrahepatic strictures might be expected among NHBD grafts. Although the authors are unaware of any clinical studies directly implicating warm ischemic time as causing intrahepatic strictures, indirect evidence is provided by a decrease in nonanastomotic biliary strictures after reperfusion of the liver simultaneously through the portal vein and hepatic artery as compared with sequential portal and arterial reperfusion (21). More direct evidence is derived from a porcine model of NHBD hepatic transplantation, where warm ischemia was found to result in irreversible damage to the biliary system (5). In this model, donor hepatic xanthine levels appear to predict survival after transplantation, whereas l-arginine administration during liver procurement prevented liver and biliary tract damage (22,23). Bile duct cells have lower levels of glutathione than hepatocytes and thus may be more susceptible to reoxygenation injury (15). The authors surmise that the combination of donor warm ischemia and subsequent cold ischemia-reperfusion injury increase the generation of oxygen radicals within the biliary epithelium to a greater extent than in HBD grafts. Although the authors do not treat their NHBD recipients with antioxidants before hepatic reperfusion, the Wisconsin group administers vitamin E, prostaglandin E1, and N-acetylcysteine.
Although 1-year graft failure and patient deaths were more frequent among NHBD recipients compared with HBD recipients, the difference was not statistically different. It is possible that the small number of patients in the authors’ NHBD group “underpowers” the study and thus a difference in graft and patient survival may exist. Any differences in patient or graft survival is unlikely to be related to the recipient factors, because both groups were similar in all aspects examined except for the incidence of hepatocellular carcinoma. D’Alessandro et al. (7) have found decreased graft survival (53.8% vs. 80.9%) but not patient survival at 3 years when NHBD recipients are compared with a group of HBD recipients, although the cause of half of the graft losses was death with function. The Pittsburgh group has reported six recipients of controlled hepatic allografts with an actuarial 1-year patient and graft survival of 50% (6). Two patients experienced PNF and underwent retransplantation and another died from a myocardial infarction with a functioning graft. There were no incidences of graft failure or patient death in eight NHBD recipients reported by Reich et al. (8).
NHBD offer the potential to expand the donor pool of hepatic allografts. Most of the authors’ recipients of NHBD allografts have had successful outcomes, with a 3-year survival similar to HBD recipients. However, the authors urge caution, because warm ischemia may have an important effect on biliary complications. Larger series and further follow-up will be needed to determine the exact incidence of ischemic type biliary complications.
1. Institute of Medicine, National Academy of Sciences. Non-heart-beating organ transplantation: Medical and ethical issues in procurement. Washington, DC: National Academy Press, 1997.
2. Koogler T, Costarino AT. The potential benefits of the pediatric nonheartbeating organ donor. Pediatrics 1998; 101: 1049.
3. Nathan HM, Jarrell BE, Broznik B, et al. Estimation and characterization of the potential renal organ donor pool in Pennsylvania: Report of the Pennsylvania statewide donor study. Transplantation 1991; 51: 142.
4. Takada Y, Taniguchi H, Fukunaga K, et al. Hepatic allograft procurement from non-heart-beating donors. Transplantation 1997; 63: 369.
5. Garcia-Valdecasas JC, Tabet J, Valero R, et al. Evaluation of ischemic injury during liver procurement from non-heart-beating donors. Eur Surg Res 1999; 31: 447.
6. Casavilla A, Ramirez C, Shapiro R, et al. Experience with liver and kidney allografts from non-heart-beating donors. Transplantation 1991; 59: 197.
7. D’Alessandro AM, Hoffmann RM, Knechtle SJ, et al. Liver transplantation from controlled non-heart-beating donors. Surgery 2000; 128: 579.
8. Reich DJ, Munoz SJ, Rothstein KD, et al. Controlled non-heart-beating donor liver transplantation. Transplantation 2000; 70: 1159.
9. Connor HD, Gao W, Nukina S, et al. Evidence that free radicals are involved in graft failure following orthotopic liver transplantation in the rat: An electron paramagnetic resonance spin trapping study. Transplantation 1992; 54: 199.
10. Kusumoto K, Morimoto T, Minor T, et al. Allopurinol effects in rat liver transplantation on recovery of metabolism and free radical-induced damage. Eur Surg Res 1995; 27: 285.
11. Yagi T, Sadamori H, Nakagawa H, et al. Hepatropic soluble vitamin E derivative EPC-K1 prevents warm ischemia/reperfusion injury of non-heart-beating donor liver transplantation in pigs. Transplant Proc 1997; 29: 1390.
12. Cutrin JC, Cantino D, Biasi F, et al. Reperfusion damage to the bile canaliculi in transplanted human liver. Hepatology 1996; 24: 1053.
13. Colonna JO, Shaked A, Gomes AS, et al. Biliary strictures complicating liver transplantation: Incidence, pathogenesis, management, and outcome. Ann Surg 1992; 216: 344.
14. Sanchez-Urdazpal L, Gores GJ, Ward EM, et al. Ischemic-type biliary complications after orthotopic liver transplantation. Hepatology 1992; 16: 49.
15. Noack K, Bronk SF, Kato A, et al. The greater vulnerability of bile duct cells to reoxygenation injury than to anoxia. Transplantation 1993; 56: 495.
16. Benkoel L, Dodero F, Hardwigsen J, et al. Effect of ischemia-reperfusion on bile canalicular f-actin microfilaments in hepatocytes of human liver allograft. Dig Dis Sci 2001; 46: 1663.
17. Deschenes M, Forbes C, Tchervenkov J. Use of older donor livers is associated with more extensive ischemic damage on intraoperative biopsies during liver transplantation. Liver Transpl Surg 1999; 5: 357.
18. Campbell WL, Sheng R, Zajko AB, et al. Intrahepatic biliary strictures after liver transplantation. Radiology 1994; 191: 735.
19. Li S, Stratta RJ, Langnas AN, et al. Diffuse biliary tract injury after orthotopic liver transplantation. Am J Surg 1992; 164: 536.
20. Rull R, Valdecasas JC, Grande L, et al. Intrahepatic biliary lesions after orthotopic liver transplantation. Transpl Int 2001; 14: 129.
21. Sankary HN, McChesney L, Frye E, et al. A simple modification in operative technique can reduce the incidence of nonanastomotic biliary strictures after orthotopic liver transplantation. Hepatology 1995; 21: 63.
22. Net M, Valero R, Almenara R, et al. Hepatic xanthine levels as viability predictor of livers procured from non-heart-beating donor pigs. Transplantation 2001; 71: 1232.
23. Valero R, Garcia-Valdecasas JC, Net M, et al. L-arginine reduces liver and biliary tract damage after liver transplantation from non-heart-beating donor pigs. Transplantation 2000; 70: 730.