A Bayesian re-analysis of hypothermic oxygenated perfusion versus static cold storage in the NIHP2019 heart transplantation trial.

The NIHP2019-trial compared hypothermic oxygenated perfusion (HOPE) with static cold storage (SCS) during heart transplantation and investigated clinical outcomes. This post hoc Bayesian re-analysis estimated the probability of clinically meaningful overall treatment effects. In total, 204 patients were analyzed at 12 months (HOPE n = 101, SCS n = 103). Bayesian models with minimally informative priors were used to derive relative risks (RR), absolute risk differences (ARD), and posterior probabilities. In these Bayesian models, HOPE was associated with lower rates of the primary composite endpoint (RR 0.71, 95%CrI 0.49-0.99; ARD -13.8%, 95%CrI -26.8; -0.4%). The probability of any benefit was 97.8%, and the probability of a clinically relevant effect exceeded 90%. For all-cause mortality, the probability of benefit was 90.5%, with a >85% probability of a clinically relevant effect. The results were robust across sensitivity analyses. These findings demonstrate clinically meaningful benefits with HOPE and provide a complementary interpretation of the NIHP2019-trial's results.