Back to Annual Meeting Program

Objective: Optimal use of split liver transplantation has the potential to significantly decrease pediatric waitlist mortality. Concerns over the risk to adult recipients of the split grafts have significantly limited their use. Our aim was to determine the adjusted-risk of graft failure in adult recipients of split livers using the United Network for Organ Sharing (UNOS).
Design: Prospective cohort study. All risk-factors suggestive on univariate analyses(p<0.2) were included in a proportional hazards model of graft survival with propensity modeling. Interaction/subset analyses were used to identify low-risk populations.
Setting: Transplantation-Centers
Patients: 62,252 first-time adult recipients of deceased donor liver transplants in the years 1995-2010 were analyzed, including 889 split livers (post-MELD era 2002-2010).
Interventions: None
Outcome: Graft Survival
Results: On adjusted-analysis, split livers are associated with an increased risk of graft failure relative to whole livers over the study period (HR 1.3 CI95%1.1-1.5,p<.001) and in the Pre-MELD era(HR1.4 CI95%1.2-1.7, p<.001). After MELD the adjusted risk of graft failure is HR1.17 (CI95%.98-1.4,p=.08) relative to whole livers. Interaction/subset analyses between split livers and other risk factors suggested an increased risk with split transplantation in patients listed as Status 1 and those with hepatocellular carcinoma (HCC). When these higher-risk recipients were excluded in the MELD era, split livers had comparable outcomes to whole livers (HR.97, CI95%.8-1.2, p=.76).
Conclusions: Split graft survival is comparable to that of whole livers when excluding recipients who are status 1 or who have HCC, who are at disproportionately high risk of graft failure when receiving splits. This study supports the use of split liver transplantation in the majority of adult recipients, an action that may significantly reduce pediatric waitlist-mortality and wait-time.