Severe Primary Graft Dysfunction: Impact of the New UNOS Heart Allocation System

Abstract

Primary graft dysfunction (PGD) is a leading cause of early morbidity and mortality following heart transplantation (HTx). The purpose of this study was to determine the mortality risk and incidence of PGD pre- and post-implementation of the new UNOS heart allocation policy. A total of 426 consecutive adult patients underwent HTx from 1/2010 to 5/2019. Recipients of combined organ transplants were excluded. Definition of severe PGD was based on the 2014 ISHLT consensus statement which includes dependence on mechanical circulatory support, excluding requirement for intra-aortic balloon pump (IABP). The primary outcome was the incidence of PGD and overall 1-year survival. The cohort studied was predominantly male (70.8%), with a mean age of 53.4±12.8 years; 31.6% were supported with a left ventricular assist device as a bridge to HTx, and 38.4% were blood type O. There were 17 (4.0%) cases of severe PGD, all of which required extracorporeal membrane oxygenation support. Baseline characteristics pre-HTx were similar in those with and without PGD (Figure 1). There was no difference in the incidence of severe PGD following the introduction of the new allocation policy in October 18, 2018 (5.9 vs. 4.5%, p=0.70), but there was a significant increase in the use of post-transplant IABP (23.5 vs. 9.1%, p=0.009). There was no correlation between RADIAL score and incidence of severe PGD (p=0.41). Severe PGD was associated with reduced 30-day survival (97.6% [95% CI: 95.5-98.8] vs. 68.7% [40.5-85.6], p<0.001) and 1-year survival (92.4% [95% CI: 89.1-94.7] vs. 50.0% [95% CI: 24.5-71.1], p<0.001). (Figure 1). Patients with severe PGD had an 8.6-fold increased risk of death when adjusted for age (p<0.001). Incidence of severe PGD has not increased in the contemporary era with the new UNOS heart allocation policy. PGD is associated with poor 1-year survival and is not predicted by the RADIAL score, which underscores the need for a new risk prediction model.