Impact of Sharing O Heart With Non-O Recipients: Simulation in the United Network for Organ Sharing Registry.

Overview

abstract

BACKGROUND: Blood type O heart transplant recipients wait longer than non-O recipients and frequently require bridging left ventricular assist devices (LVADs). However, rarely has the effect of this disparity been shown in a large registry. This study used Markov simulation to clarify the outcome difference between O and non-O candidates and how allocation change could affect survival. METHODS: We reviewed the United Network for Organ Sharing registry for adults listed for heart transplantation from 2008 to 2015. Cumulative incidences of death or transplant and survival after listing were compared between O and non-O using propensity matching. A four-state Markov model-waiting without LVAD, waiting with LVAD, transplantation, and death-was created to simulate survival after listing. Sensitivity analysis was performed to see how the percentage of O hearts in non-O recipients would affect survivals. RESULTS: A total of 8,187 O and non-O candidates were included after matching. The cumulative incidence of transplantation was lower in O (p < 0.001), and death after listing was significantly higher (p < 0.001). During a median follow-up of 2.1 years, 69.6% of non-O candidates were transplanted, compared with 54.9% of O candidates (p < 0.001). Despite this disparity, 19% of non-O recipients received O hearts. Our simulated survival demonstrated that decreasing the O heart percentage in non-O recipients from the current 19% to 5% would provide similar survival in O and non-O after listing. CONCLUSIONS: Under the current strategy, there are death and transplant rate differences between O and non-O candidates. Our simulation-based allocation strategy might aid in mitigating this discrepancy across blood types.