Development of a novel risk score to predict 1-year mortality following donation after circulatory death heart transplant.

Abstract

Donation after circulatory death (DCD) has expanded the US heart donor pool and yielded similar outcomes for recipients as donation after brain death. However, little work has analyzed what factors contribute to 1-year mortality in DCD nor created a clinical risk score. This work fills these gaps. All heart-only, adult DCD heart transplants from the United Network for Organ Sharing database were analyzed. The data was split into a training cohort and a testing cohort. Univariable and multivariable analyses were used to identify 1-year mortality risk factors in the training cohort. These factors were then used to develop a clinical risk score and applied to the testing cohort. The risk score contained 12 recipient variables (age, diabetes, durable left ventricular assist device placemebnt, estimated glomerular filtration rate, prior cardiac surgery, retransplant, congenital diagnosis, malignancy, total bilirubin, extracorporeal membrane oxygenation at transplant, recent infection, and positive cytomegalovirus status) and 2 donor variables (compromised direct perfusion procurement donor and positive cytomegalovirus status), yielding a training concordance of 0.709. The risk score applied to the testing cohort displayed modest discrimination, with a concordance of 0.668. One point of the risk score had a hazard ratio of 1.16 (95% CI, 1.12-1.20; P = .00043). Many of the risk factors associated with 1-year mortality in DCD-heart transplant are also reported in donation after brain death heart transplant. However, factors related to the warm ischemic time in DCD, like in compromised direct perfusion procurement donor donors, also warrant consideration. This clinical risk score may be used to assess 1-year mortality risk for recipients of DCD hearts.