Cardiovascular disease is a major cause of morbidity and mortality following kidney transplantation. Many cardiovascular risk screening tools have been developed, but they have never been compared in kidney transplant patients. Our aim was to evaluate the predictive value of three risk assessment tools on a single transplant population. All kidney transplants were identified from 2003-2010 (n=710) at our institution. Retrospective chart review was performed using the clinical looking glass program and electronic medical records. Preoperative information necessary for scoring each patient on the Revised Cardiac Risk Index (RCRI), the Patient Outcomes in Renal Transplantation (PORT) model, and the Gupta cardiac risk calculator, as well as postoperative cardiac complications defined respectively by each tool, were collected. Models were compared using a Spearman's rank correlation coefficient. Prediction accuracy was evaluated using the C-statistic (area under the ROC curve). RCRI preoperative risk factors were significantly correlated with cardiac outcomes (Spearman, 0.21; CI, 0.19-0.30; p=9.3e-9) and the model was fairly predictive (ROC, 0.60; p=0.0001). PORT preoperative risk factors were also significantly correlated with cardiac outcomes (Spearman, 0.19; CI, 0.13-0.25; p=3.7e-7) and the model was a good predictor (ROC, 0.71; p=1.2e-5). Gupta risk factors were not correlated (Spearman, -0.01, p=0.89) and the model was a poor predictor (ROC, 0.49, p=0.86). In summary, the PORT model was the most accurate predictor of cardiac risk. Kidney transplant patients have a unique physiological state. Evaluating their risk of major adverse cardiac events necessitates a unique scoring system.

« Back to 2015 American Transplant Congress