Risk Factors for New-Onset Diabetes Mellitus After Heart Transplantation: A Nomogram Approach

Abstract

New-onset diabetes mellitus after transplantation (NODAT) is a leading cause of morbidity and mortality after heart transplantation (HT), which still remains a clinical challenge. In this study, 522,708 follow-up records of HT were reviewed. After screening, 14,452 patients were analyzed when combined with immunosuppression records. We divided all patients into no-NODAT group, NODAT group, and preexisting diabetes group based on whether the patient had diabetes and the time when it occurred. Cox regression models were used to examine independent risk factors. A nomogram was established to predict the incidence of NODAT after HT. The machine learning method were used to confirm the prediction accuracy and reliability of the nomogram. Patients who experienced NODAT after HT had poor survival compared with those without NODAT. Tacrolimus, cyclosporine A (CsA), rapamycin, donor age, and recipient age at the time of transplant were significant predictors of NODAT. Tacrolimus had a more significant association with NODAT, followed by rapamycin and CsA. The nomogram method we adopted in this study had an accuracy of 63% in predicting the incidence of NODAT. The survival probability of HT recipients with NODAT showed a significant decreasing tendency. However, there was no difference in survival probability between patients with preexisting diabetes and patients with NODAT. Tacrolimus had a more significant association with NODAT than CsA and rapamycin.