PREDICTA: A Model to Predict Primary Graft Dysfunction After Adult Heart Transplantation in the United Kingdom

Background: Primary graft dysfunction (PGD) is a major cause of morbidity and mortality post-heart transplantation. The rates of PGD across Europe are higher than North America possibly because of the increasing use of extended criteria donors because of organ shortage. Our aim was to derive a novel scoring system based on data collected nationally in the UK over a 3-year period. We compared this scoring system to a previously validated (RADIAL) score in a contemporary cohort of patients. Methods: Medical records of all adult patients who underwent heart transplantation between October 1, 2012 and September 30, 2016 in the 6 UK heart transplant centers were analyzed. Preoperative donor and recipient characteristics, intraoperative details and post-transplant complications were compared between the PGD and non-PGD groups using the International Society of Heart and Lung Transplant definition. Multivariable logistic regression was used to build the predictive model. An area under receiver operating characteristics curve was used to test the novel scoring system (PREDICTA) versus the RADIAL score. Results: Six hundred and thirteen heart transplants were included in the study. There were 233 patients who had PGD. The variables included in the model were recipient diabetes mellitus, preoperative mechanical circulatory support (short-term ventricular assist devices/extracorporeal membrane oxygenation), implant time, donor age, and bypass time >180 minutes. The C statistic of the PREDICTA score was 0.704 versus 0.547 for the RADIAL score indicating an acceptable discriminatory value. Conclusion: The PREDICTA score is a novel scoring tool with improved ability to predict the development of PGD compared with the RADIAL score. Its application in the prevention and early management of PGD needs further evaluation.