Role of Donor Hemodynamic Trajectory in Determining Graft Survival in Liver Transplantation From Donation After Circulatory Death Donors

Donation after circulatory death (DCD) donors show heterogeneous hemodynamic trajectories following withdrawal of life support. Impact of hemodynamics in DCD liver transplant is unclear, and objective measures of graft viability would ease transplant surgeon decision making and inform safe expansion of the donor organ pool. This retrospective study tested whether hemodynamic trajectories were associated with transplant outcomes in DCD liver transplantation (n587). Using longitudinal clustering statistical techniques, we phenotyped DCD donors based on hemodynamic trajectory for both mean arterial pressure (MAP) and peripheral oxygen saturation (SpO(2)) following withdrawal of life support. Donors were categorized into 3 clusters: those who gradually decline after withdrawal of life support (cluster 1), those who maintain stable hemodynamics followed by rapid decline (cluster 2), and those who decline rapidly (cluster 3). Clustering outputs were used to compare characteristics and transplant outcomes. Cox proportional hazards modeling revealed hepatocellular carcinoma (hazard ratio [HR] = 2.53; P = 0.047), cold ischemia time (HR = 1.50 per hour; P = 0.027), and MAP cluster 1 were associated with increased risk of graft loss (HR = 3.13; P = 0.021), but not SpO(2) cluster (P = 0.172) or donor warm ischemia time (DWIT; P = 0.154). Despite longer DWIT, MAP and SpO(2) clusters 2 showed similar graft survival to MAP and SpO(2) clusters 3, respectively. In conclusion, despite heterogeneity in hemodynamic trajectories, DCD donors can be categorized into 3 clinically meaningful subgroups that help predict graft prognosis. Further studies should confirm the utility of liver grafts from cluster 2.