How can we define expanded criteria for liver donors?

Abstract

A large and global imbalance between the supply of donor organs for liver transplantation and the pool of potential liver transplant recipients continues to fuel efforts to maximize utilization from existing donors, increase the overall number of donors, and identify new donor sources. Until recently, deceased donor liver transplantation was largely confined to donors with ongoing cardiac activity who were diagnosed with brain death. Donation after brain death (DBD) donors continue to supply about 95% of all livers used for transplantation in the US [[1]2005 Annual Report of the US Organ Procurement and Transplantation Network and the Scientific Registry for Transplant Recipients: Transplant Data 1995–2004: Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; United Network for Organ Sharing, Richmond, VA; University Renal Research and Education Association, Ann Arbor, MI, 2006.Google Scholar]. In recent years, donation of right hepatic hemiliver by living persons has been reported [2Lo C.M. Fan S.T. Liu C.L. et al.Adult-to-adult living donor liver transplantation using extended right lobe grafts.Ann Surg. 1997; 226: 261Crossref PubMed Scopus (529) Google Scholar, 3Wachs M.E. Bak T.E. Karrer F.M. et al.Adult living donor liver transplantation using a right hepatic lobe.Transplantation. 1998; 66: 1313Crossref PubMed Scopus (278) Google Scholar, 4Marcos A. Ham J.M. Fisher R.A. Olzinski A.T. Posner M.P. Single-center analysis of the first 40 adult-to-adult living donor liver transplants using the right lobe.Liver Transpl. 2000; 6: 296Crossref PubMed Scopus (261) Google Scholar]. These adult-to-adult living donor liver transplants place a healthy donor at risk for serious complications and death [5Middleton P.F. Duffield M. Lynch S.V. et al.Living donor liver transplantation–adult donor outcomes: a systematic review.Liver Transpl. 2006; 12: 24Crossref PubMed Scopus (216) Google Scholar, 6Beavers K.L. Sandler R.S. Shrestha R. Donor morbidity associated with right lobectomy for living donor liver transplantation to adult recipients: a systematic review.Liver Transpl. 2002; 8: 110Crossref PubMed Scopus (190) Google Scholar, 7Verbesey J.E. Simpson M.A. Pomposelli J.J. et al.Living donor adult liver transplantation: a longitudinal study of the donor’s quality of life.Am J Transplant. 2005; 5: 2770Crossref PubMed Scopus (100) Google Scholar, 8Pomfret E.A. Early and late complications in the right-lobe adult living donor.Liver Transpl. 2003; 9: S45Crossref PubMed Scopus (75) Google Scholar], and the number of such transplants has not had a major influence on overall liver donor availability [[1]2005 Annual Report of the US Organ Procurement and Transplantation Network and the Scientific Registry for Transplant Recipients: Transplant Data 1995–2004: Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; United Network for Organ Sharing, Richmond, VA; University Renal Research and Education Association, Ann Arbor, MI, 2006.Google Scholar]. Recently, many transplant programs have begun to perform transplants using livers from deceased donors with characteristics previously thought to be associated with an unacceptably high risk of graft failure. These include transplants using livers from, among others, older donors, and those whose hearts have stopped beating, termed donation after cardiac death (DCD), or non-heart-beating donor (see article by R. Deshpande & N. Heaton in this Forum). Efforts to expand the donor pool have included the development of novel surgical approaches to split a single liver to transplant two recipients. Typically, the extended right hemiliver (segments IV–VIII) graft is utilized for an adult recipient and the bisegment II and III is transplanted into a pediatric recipient [[9]Emond J.C. Whitington P.F. Thistlethwaite J.R. et al.Transplantation of two patients with one liver. Analysis of a preliminary experience with ‘split-liver’ grafting.Ann Surg. 1990; 212: 14Crossref PubMed Scopus (261) Google Scholar]. Some cases of splitting a liver into right and left lobes to transplant two adult recipients have also been reported [[10]Humar A. Khwaja K. Sielaff T.D. Lake J.R. Payne W.D. Technique of split-liver transplant for two adult recipients.Liver Transpl. 2002; 8: 725Crossref PubMed Scopus (32) Google Scholar]. In recent years, as each of these forays into previously uncharted donor territory has been undertaken, reports have appeared suggesting that many liver transplant candidates could be served by these innovations. At the same time, experience has demonstrated that each option is associated with an incremental risk of graft failure when compared to whole liver transplants from young deceased donors. Beyond the issues mentioned above which affect graft function, there is an entirely separate concept of donor risk factors that concerns the possibility of transmissible disease (e.g., hepatitis virus, human immunodeficiency virus, or malignancy). These issues are discussed in the article by N. Muller & J. Fishman in this Forum. This article will review the dramatic evolution of donor characteristics that has occurred in recent years, describe factors for which the degree or level of risk has been quantified, report on the development of a method to combine multiple known risk factors into a continuous donor risk index, and conclude with some comments on the potential for integration of information about donor risk into decision making about the use of particular organs for a given liver transplant candidate. Throughout the 1980s and the first half of the 1990s, the typical donor was a young male who sustained brain death from cerebral trauma. These tragic situations eventuated in recovery and transplantation of multiple organs that routinely functioned immediately and sustained their respective recipients for many years thereafter. In the past decade, however, there have been dramatic shifts in deceased donor characteristics. For example, in 1994, only one deceased donor out of five was 50 years or older (Fig. 1). By 2004, the absolute number of such donors had increased by more than 150% and comprised more than one-third of all deceased liver donors. Traumatic causes of brain death have generally been considered to yield donor livers that are more likely to function well than those from donors whose cause of death is cerebrovascular accident or stroke. As with donor age, the distribution of cause of death among deceased donors in the US has shifted markedly. Whereas trauma was the cause of death in more than one-half of all donors ten years ago, they now comprise only 40%, while cerebrovascular causes have increased by 34% (Fig. 2). Recently, DCD donor liver transplantation has re-emerged and sharply increasing numbers have been reported in the past five years. Although they account for only about 4% of all deceased donor liver transplants nationally [[1]2005 Annual Report of the US Organ Procurement and Transplantation Network and the Scientific Registry for Transplant Recipients: Transplant Data 1995–2004: Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; United Network for Organ Sharing, Richmond, VA; University Renal Research and Education Association, Ann Arbor, MI, 2006.Google Scholar], DCD liver transplants may play an increasing role in satisfying donor needs (see article by R. Deshpande & N. Heaton in this Forum). Split liver transplantation has not become widespread in the US [[11]Renz J.F. Yersiz H. Reichert P.R. et al.Split-liver transplantation: a review.Am J Transpl. 2003; 3: 1323Crossref PubMed Scopus (86) Google Scholar]. In contrast, European centers have embraced this technique as a means to ameliorate the shortage of donor livers and split livers comprised about 5% of all transplants performed between 1988 and 2003 [[12]Burroughs A.K. Sabin C.A. Rolles K. et al.3-month and 12-month mortality after first liver transplant in adults in Europe: predictive models for outcome.Lancet. 2006; 367: 225Abstract Full Text Full Text PDF PubMed Scopus (281) Google Scholar]. Logistical issues in the US appear to account for some of the reluctance to split more livers. The Europeans have adopted a more flexible approach, particularly with respect to the allocation of the second graft created by the split procedure [[13]Gerling T. Karbe T. Kutemeier R. et al.The role of an organ exchange organization in increasing split-liver transplantation.Clin Transpl. 2004; : 83PubMed Google Scholar]. Many recipient, donor, and transplant-related characteristics have been reported to influence outcomes after liver transplantation. Among these, older donor and recipient age, longer cold ischemia time, and surgical management of the graft (whole versus partial or split graft) are perhaps the best documented [12005 Annual Report of the US Organ Procurement and Transplantation Network and the Scientific Registry for Transplant Recipients: Transplant Data 1995–2004: Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; United Network for Organ Sharing, Richmond, VA; University Renal Research and Education Association, Ann Arbor, MI, 2006.Google Scholar, 12Burroughs A.K. Sabin C.A. Rolles K. et al.3-month and 12-month mortality after first liver transplant in adults in Europe: predictive models for outcome.Lancet. 2006; 367: 225Abstract Full Text Full Text PDF PubMed Scopus (281) Google Scholar, 14Feng S. Goodrich N.P. Bragg-Gresham J.L. et al.Characteristics associated with liver graft failure: the concept of a donor risk index.Am J Transplant. 2006; 6: 783Crossref PubMed Scopus (1416) Google Scholar, 15Ghobrial R.M. Gornbein J. Steadman R. et al.Pretransplant model to predict posttransplant survival in liver transplant patients.Ann Surg. 2002; 236: 315Crossref PubMed Scopus (146) Google Scholar, 16Berenguer M. Prieto M. San Juan F. et al.Contribution of donor age to the recent decrease in patient survival among HCV-infected liver transplant recipients.Hepatology. 2002; 36: 202Crossref PubMed Scopus (568) Google Scholar, 17Markmann J.F. Markmann J.W. Markmann D.A. et al.Preoperative factors associated with outcome and their impact on resource use in 1148 consecutive primary liver transplants.Transplantation. 2001; 72: 1113Crossref PubMed Scopus (157) Google Scholar, 18Deschenes M. Forbes C. Tchervenkov J. et al.Use of older donor livers is associated with more extensive ischemic damage on intraoperative biopsies during liver transplantation.Liver Transpl. 1999; 5: 357Crossref Scopus (41) Google Scholar, 19Furukawa H. Todo S. Imventarza O. et al.Effect of cold ischemia time on the early outcome of human hepatic allografts preserved with UW solution.Transplantation. 1991; 51: 1000Crossref PubMed Scopus (203) Google Scholar, 20Roberts J.P. Hulbert-Shearon T.E. Merion R.M. Wolfe R.A. Port F.K. Influence of graft type on outcomes after pediatric liver transplantation.Am J Transplant. 2004; 4: 373Crossref PubMed Scopus (95) Google Scholar]. While some have attempted to condense some of the variables (e.g., age and cold ischemia time) into a small number of subgroups, it appears that a more continuous spectrum of risk exists for each, which does not support their convenient division into arbitrary categories. In addition to these dominant variables, donor and recipient gender (and especially specific donor:recipient gender combinations), donor and recipient race or ethnicity, body mass index, cause of death, and DCD donor status have all been cited as factors significantly associated with graft outcome [21Zeier M. Dohler B. Opelz G. Ritz E. The effect of donor gender on graft survival.J Am Soc Nephrol. 2002; 13: 2570Crossref PubMed Scopus (196) Google Scholar, 22Rinella M.E. Alonso E. Rao S. et al.Body mass index as a predictor of hepatic steatosis in living liver donors.Liver Transpl. 2001; 7: 409Crossref PubMed Scopus (218) Google Scholar, 23Sawyer R.G. Pelletier S.J. Pruett T.L. Increased early morbidity and mortality with acceptable long-term function in severely obese patients undergoing liver transplantation.Clin Transpl. 1999; 13: 126Crossref Scopus (128) Google Scholar, 24Mateo R. Cho Y. Singh G. et al.Risk Factors for Graft Survival After Liver Transplantation from Donation After Cardiac Death Donors: An Analysis of OPTN/UNOS Data.Am J Transplant. 2006; 6: 791Crossref PubMed Scopus (214) Google Scholar]. Hepatic steatosis and donor hypernatremia have both been reported to be significantly associated with liver graft failure [25Verran D. Kusyk T. Painter D. et al.Clinical experience gained from the use of 120 steatotic donor livers for orthotopic liver transplantation.Liver Transpl. 2003; 9: 500Crossref PubMed Scopus (223) Google Scholar, 26Totsuka E. Dodson F. Urakami A. et al.Influence of high donor serum sodium levels on early postoperative graft function in human liver transplantation: effect of correction of donor hypernatremia.Liver Transpl. 1999; 5: 421Crossref Scopus (130) Google Scholar], although large registry analyses have generally not supported these findings, perhaps due to an inadequate level of granularity in the data collected regarding these factors [[14]Feng S. Goodrich N.P. Bragg-Gresham J.L. et al.Characteristics associated with liver graft failure: the concept of a donor risk index.Am J Transplant. 2006; 6: 783Crossref PubMed Scopus (1416) Google Scholar]. Additionally, although some reports do not demonstrate a significant adverse association, inadequacies in sample size (e.g., type II error) and other methodological shortcomings may be present. It is also important to note that the magnitude of risk posed by some factors may change over time. For example, as for use of split organs, the adverse outcomes that are now observed may lead to modifications in practice or changes in utilization that may ameliorate some of the negative impact of a split liver transplant. However, the underlying biology responsible for the suboptimal outcomes associated with use of a partial graft is unlikely to completely disappear. Along with increasing utilization of donated livers from individuals with characteristics that are outside of previously defined, albeit arbitrary, boundaries, greater awareness of the impact on outcome has developed. On the foundation of the many reports of associations between specific donor characteristics and post-transplant graft failure cited above, quantitative estimations of organ quality relative to index or ideal donor organs have been more recently developed and reported. For example, in the area of deceased donor kidney transplantation, the relative risk of graft failure based entirely upon donor characteristics has been established. Using these relative risks, a dichotomous classification using the terminology expanded criteria donor (ECD) or non-ECD was derived [[27]Port F.K. Bragg-Gresham J.L. Metzger R.A. et al.Donor characteristics associated with reduced graft survival: an approach to expanding the pool of kidney donors.Transplantation. 2002; 74: 1281Crossref PubMed Scopus (614) Google Scholar]. In the realm of liver transplantation, donor variables predictive of graft failure are expected to differ from those identified for kidney donors. Combinatorial regression equations can easily accommodate continuous scales for continuous variables, so that forced dichotomization of ECD or non-ECD liver donors can be avoided. Additionally, concerns about selective transplantation practices suggest that the impact of donor factors may be further tempered by the choice of recipient into whom such organs are placed. With these principles in mind, the foundation was laid to create a continuous donor risk index comprised of liver donor-specific factors that accounts for the potential confounding effect of recipient and/or transplant characteristics [[14]Feng S. Goodrich N.P. Bragg-Gresham J.L. et al.Characteristics associated with liver graft failure: the concept of a donor risk index.Am J Transplant. 2006; 6: 783Crossref PubMed Scopus (1416) Google Scholar]. Seven characteristics were identified as significantly associated with liver allograft failure in a multivariable Cox regression model of time to graft failure or death [[14]Feng S. Goodrich N.P. Bragg-Gresham J.L. et al.Characteristics associated with liver graft failure: the concept of a donor risk index.Am J Transplant. 2006; 6: 783Crossref PubMed Scopus (1416) Google Scholar]. These included three donor demographic characteristics (age, race, and height), two relating to cause of donor death (cerebrovascular accident or causes other than cerebrovascular accident, trauma, or anoxia), one related to type of donor death (DCD), and use of a split/partial graft. Compared to a reference group of donors less than 40 years of age, increasing age was associated with a significant and monotonic increase in the risk of graft failure. For example, donor age was an extremely important risk factor for graft failure (relative risk 1.53 and 1.65, for donor age 61–70 and >70, respectively; both P < 0.0001). Livers from black donors had a 19% higher risk of graft failure compared to those from white donors (relative risk 1.19; P < 0.0001). Although two parameters reflecting donor size were assessed, the association of height was stronger than, and independent of, the association of weight. Compared to trauma as a cause of death, cerebrovascular accident and other causes of death (not trauma, cerebrovascular accident, or anoxia) were associated with 16% and 20% higher risks of graft failure, respectively (both P < 0.02). DCD status and split/partial grafts were associated with a 51% and 52% higher risk of graft failure (P < 0.001). Two transplant-related factors, cold ischemia time and sharing outside of the local donor service area, were also found to be significantly associated with increased risk of graft loss [[14]Feng S. Goodrich N.P. Bragg-Gresham J.L. et al.Characteristics associated with liver graft failure: the concept of a donor risk index.Am J Transplant. 2006; 6: 783Crossref PubMed Scopus (1416) Google Scholar]. Each additional hour of cold ischemia time was associated with an additional 1% increased risk of graft loss (P = 0.008). For organ allocation in the US, there are 58 donation service areas that are grouped into 11 geographical regions. If the donor hospital and the recipient’s transplant center are located in the same donation service area, the transplant is considered to have occurred in the local area. Transplants that occur outside the local area but within that donation service area’s region are termed regionally shared, and those that are used for transplants beyond the region are termed nationally shared. Compared to grafts transplanted within the local area, regionally shared grafts had an 11% increase in risk of graft loss (P = 0.002), while grafts that were transplanted beyond the region (nationally shared) had a 28% increased risk of graft loss (P < 0.0001), even when adjusted for cold ischemia time. It is likely that unmeasured factors associated with local and/or regional programs demurring on particular organs are responsible for the effect for which geographic sharing unit is a surrogate. Taken together, these factors can be incorporated into a single equation to generate the combinatorial overall donor risk index for each donor, given the values for each of the underlying covariates (Table 1). The donor risk index for each particular combination is directly related to a corresponding predicted rate of graft survival, which varies from 86.4% at one year for donors less than age 40 years with no other risk factors to 23.9% for DCD or split grafts from donors over age 70 years. As an example, consider a brain-dead liver donor of white race, age 64 years, with a height of 170 cm, who died of a traumatic brain injury, and whose liver was transplanted locally after a cold ischemia time of 8 h. The calculated donor risk index would be 1.53, meaning that there is a 53% higher risk of graft failure than a reference donor under age 40 with the same characteristics. If that same donor had died of a stroke instead of a traumatic injury, the donor risk index would rise to 1.77 and further to 1.88 if the cold ischemia time were 14 h instead of 8 h. Each of these calculations can be made with a modern spreadsheet program using the formula shown in Table 1.Table 1Calculated donor risk index and one year graft survival and 95% confidence interval estimates for specified donor profiles (1998–2002) [from Feng S., et al. Am J Transplant 2006;6:783–90]Donor profileN (%)Range of calculated DRIaCalculation: donor risk index=exp[(0.154 if 40⩽age<50)+(0.274 if 50⩽age<60)+(0.424 if 60⩽age<70)+(0.501 if 70⩽age)+(0.079 if cause of death=anoxia)+(0.145 if cause of death=cerebrovascular accident)+(0.184 if cause of death=other)+(0.176 if race=African-American)+(0.126 if race=other)+(0.411 if DCD)+(0.422 if partial/split graft)+(0.066×((170-height)/10))+(0.105 if regional share)+(0.244 if national share)+(0.010×(8-cold time))].95% CI for adjusted-year survival estimatesNo risk factorsUnder 406814 (34.0%)Ref.84.8–86.440–491174 (5.9%)1.1782.9–86.950–59653 (3.3%)1.3279.0–84.860–69299 (1.5%)1.5377.7–85.970+140 (0.7%)1.6561.3–76.3Cause of death-other or cause of death-stroke or race-blackUnder 402683 (13.4%)1.16–1.2081.8–84.640–492128 (10.6%)1.35–1.4079.8–83.150–592293 (11.5%)1.52–1.5877.2–80.560–691445 (7.2%)1.77–1.8473.4–77.870+655 (3.3%)1.91–1.9972.4–78.7(Cause of death-other + race-black) or (cause of death-stroke + race-black)Under 40276 (1.4%)1.38–1.4371.7–81.340–49365 (1.8%)1.61–1.6773.7–82.050–59290 (1.5%)1.81–1.8973.7–82.860–69132 (0.7%)2.11–2.1963.4–78.470+58 (<0.5%)2.27–2.3771.6–91.0DCD or partial/split graftUnder 40344 (1.7%)1.51–1.5275.4–83.840–4934 (<0.5%)1.76–1.7845.6–78.050–5917 (<0.5%)1.98–2.0173.0–100.060–696 (<0.5%)2.30–2.3333.3–100.070+3 (<0.5%)2.49–2.5223.9–100.0(DCD or partial/split graft) + at least one other factorUnder 40129 (0.6%)1.74–3.3064.9–80.540–4951 (<0.5%)2.03–3.8560.7–84.750–5920 (<0.5%)2.29–4.3449.0–91.560–6914 (<0.5%)2.66–5.0446.4–97.070+0 (0.0%)2.88–5.45N/Aa Calculation: donor risk index = exp[(0.154 if 40 ⩽ age < 50) + (0.274 if 50 ⩽ age < 60) + (0.424 if 60 ⩽ age < 70) + (0.501 if 70 ⩽ age) + (0.079 if cause of death = anoxia) + (0.145 if cause of death = cerebrovascular accident) + (0.184 if cause of death = other) + (0.176 if race = African-American) + (0.126 if race = other) + (0.411 if DCD) + (0.422 if partial/split graft) + (0.066 × ((170-height)/10)) + (0.105 if regional share) + (0.244 if national share) + (0.010 × (8-cold time))]. Open table in a new tab At the time when a donor liver is offered for a candidate on the liver transplant waiting list, a choice must be made to either accept the risk of transplantation with that particular organ or to decline the offer in favor of waiting for a subsequent donor, which may or may not have a more favorable profile. This decision requires prognostic information about the donor graft being offered as well as knowledge of the risk of death from progressive liver disease if the offered graft is declined. The donor risk index provides substantial data regarding the first issue, and published reports show that the risk of death on the waiting list can be reasonably estimated using the Model for End-stage Liver Disease [[28]Wiesner R.H. McDiarmid S.V. Kamath P.S. et al.MELD and PELD: application of survival models to liver allocation.Liver Transpl. 2001; 7: 567Crossref PubMed Scopus (725) Google Scholar]. However, there is still one piece of the puzzle missing. The donor risk index is estimated for the average waiting candidate. If particular elements of the donor risk index have a unique or differential effect on outcome in particular subgroups of recipients, then the appropriate choice may be less straightforward than outlined here. For example, an especially adverse effect of older donor age has been described in recipients with hepatitis C [[16]Berenguer M. Prieto M. San Juan F. et al.Contribution of donor age to the recent decrease in patient survival among HCV-infected liver transplant recipients.Hepatology. 2002; 36: 202Crossref PubMed Scopus (568) Google Scholar], and donor age is an important component of the donor risk index. Analyses currently under way that are focused on careful and systematic examination of potentially important interactions between donor risk index and recipient characteristics, incorporating many of the concepts embodied in an accompanying article (see article by P. Burran & R. Porte in this Forum) and those of other reports [[29]Haydon G.H. Hiltunen Y. Lucey M.R. et al.Self-organizing maps can determine outcome and match recipients and donors at orthotopic liver transplantation.Transplantation. 2005; 79: 213Crossref PubMed Scopus (30) Google Scholar], are the subject of a forthcoming publication.