On comparing adult living donor liver transplantation to adult deceased donor liver transplantation: What have we learned?

In the present issue of Liver Transplantation, Thuluvath and Yoo retrospectively compare 2-year patient and graft survival of 764 living donor liver transplant (LDLT) adults to that of 1,470 deceased donor liver transplant (DDLT) recipients, utilizing the database of the United Network for Organ Sharing (UNOS).1 Using a case-control method of 2 DDLT to 1 LDLT recipients at or over the age of 18 years, between 1998-2001, matched for age ± 5 years, sex, ethnicity, diagnosis, and year of transplant ± 2 years, 4% of the LDLT study population were unmatched.1 Great effort was extended to gather as much recipient and donor data for comparison, and yet the comparison of complications to understand reasons for outcome was hindered by lack of data. For example, 90% of surgical complications were not reported in the UNOS database. Thus, comparison of anything other than graft loss was problematic. To further confound the graft survival analysis, 10% of the LDLT grafts were reported as left-lobe or left-lateral segment grafts (the percentage of each not specified) used for U.S. adults. Equally important is the time period studied, when the learning curve of a developing surgical therapy2, 3 of adult LDLT is a significant confounding variable that is not included in the Cox regression analysis of graft survival. Despite the difficulties of choosing the “perfect” cohorts from the UNOS database, with it's unavoidable bias4 and the inadequacy of models based on retrospective data to account for developments in diagnosis and treatment,5 the present authors confirm the findings of other investigators using the UNOS database. The findings are that LDLT recipients receive significantly younger donor grafts (P < .001), with significantly shorter cold ischemia time (P < .001) and better ABO matching (P < .001), and they are “less sick” (P < .001) than their contemporaneous DDLT recipients and achieve the same 2-year patient survival of 80%.6, 7 Like the present study, the UNOS data analysis by Freeman et al. and Brown et al. show a lower LDLT graft survival compared to the contemporaneous DDLT group but without significance.6, 7 The recent Organ Procurement and Transplantation Network / Scientific Registry of Transplant Recipients (SRTR) analysis shows that the unadjusted living donor recipient graft survival is slightly lower at 1 to 3 years compared to deceased donor recipient graft survival, only to become 15% greater at 5 years.7 A prospective single center study comparing living donor liver transplant to deceased donor liver transplant outcomes over 6 years, with a unified immunosuppression and transplant care protocol by an experienced multidisciplinary transplant team, reported similar findings to the present study of younger age of living donor and living recipient, as well as LDLT recipients having lower risk score at transplant.8 LDLT, living donor liver transplant; DDLT, deceased donor liver transplant; UNOS, United Network for Organ Sharing; MELD, model for end-stage liver disease. The findings of similar patient and graft survival at 5 years, with less histologic hepatitis C recurrence in the LDLT recipients compared to DDLT recipients in our prospective study,8 allows me to be more optimistic in continuing LDLT in hepatitis C recipients. However, I still agree with Thuluvath and Yoo that “Patients, especially those with HCV, should be offered LDLT on the basis of medical necessity and not on the basis of donor availability.”1 Thuluvath and Yoo conclude “LDLT is a reasonable option for patients who are unlikely to receive DDLT in a timely fashion.”1 What does this mean? Who are these patients? The increasing mortality on the liver transplant waiting list that pushed many of us to begin living liver donor programs for our patients declined by 11% in 2002, with a median time from listing to transplant of 128 days for adults with a model for end-stage liver disease (MELD) score of 21 to 30.7 Candidates for liver transplant with a MELD score range of 14 to 25 may not have a high probability of receiving a deceased donor liver transplant prior to suffering morbidity from their native liver disease. Also, the critical importance of matching adequate liver volume to the degree of decompensation of recipient disease2, 3, 9 would support the most lifetime benefit and lowest risk of posttransplant mortality of LDLT for candidates with MELD scores of 14 to 25, as suggested by Freeman et al.6 Finally, the recently formed National Institutes of Health consortium of nine U.S. transplant centers with a large-enough experience of living donor liver transplantation will address risk-adjusted outcomes and uniform reporting of postoperative complications and preoperative risk factors to compare living donor liver transplant to whole-organ transplant. It is hoped that the limitations of the present retrospective database comparative models and single-center studies will be overcome by these future standardized and critically overseen efforts.