A National Registry Analysis of Aortic versus Dual in Situ Perfusion for Retrieval of the DBD Liver

Abstract

Listen

Translate article

Introduction In situ perfusion of preservation fluid during donation after brain death (DBD) liver retrieval can be conducted via the aorta alone, or aorta and portal vein (dual perfusion). There is considerable disagreement in the literature with regards to the comparative efficacy of each perfusion route for both normal and expanded criteria liver donors, and the few existing studies are disadvantaged by low patient numbers and short periods of follow-up. Materials/Methods DBD whole liver transplants (initial) in Australia were included from 2007-2016, and stratified by aortic (n = 957) or dual (n = 425) perfusion routes. Data points were obtained from the Australia and New Zealand (ANZ) Liver Transplant Registry, the ANZ Organ Donation Registry, and a national survey of senior donor surgeons. University of Wisconsin (UW) solution was given via the aorta and/or portal vein, followed by organ transport in the same fluid. Missing data was handled by multiple imputations. Graft and patient survival were compared using Kaplan-Meier curves and Cox proportional hazards. Causes of graft loss, including primary non-function (PNF), hepatic artery (HAT) and portal vein thrombosis (HAT), biliary complications (BC) and acute rejection (AR) were compared using logistic regression. Results Baseline characteristics between study groups were similar, except for a lower mean cold ischemic time (CIT; 6.3 vs 7.0 hrs), mean secondary warm ischemic time (SWIT; 37.8 vs 45.4 mins), and median recipient MELD score (14 vs 18) in the dual-perfused patient cohort compared to the aortic-only perfusion group (p < 0.001). Actuarial 5-year graft and patient survivals in aortic and dual perfusion cohorts were 80.1% vs 84.6% (p = 0.066, univariate log-rank test), and 82.6% vs 87.8% (p = 0.026, univariate log-rank test), respectively. Multivariate Cox proportion hazards models, accounting for CIT, SWIT, MELD, and other donor/recipient factors with a p-value < 0.1 in univariate analyses, showed that graft survival after aortic vs dual perfusion was not significantly different (HR 0.81, 95% CI 0.60-1.11, p = 0.188). Similarly, overall patient survival was not different between the aortic and dual groups (HR 0.74, 95% CI 0.52-1.05, p = 0.087). There were no significant differences between aortic and dual perfusion groups with respect to causes of graft loss, including PNF, HAT, PVT, BC, and AR. Discussion After accounting for confounders, there were no significant differences in causes of graft loss, graft survival, and patient survival between liver transplants performed after aortic-only or dual in situ liver perfusion at retrieval. Subgroup analyses will need to be conducted to compare high-risk donors. Conclusion The retrieval technique employed does not impact outcomes for standard risk donors. Future RCTs should focus on the efficacy of either technique in liver donors with a high donor risk index, and also consider the impact on other organs, in particular the pancreas. Royal Australasian College of Surgeons.