We conducted a retrospective study of 305 nondiabetic patients and 37 diabetic patients who underwent heart transplantation from July 1982 to May 1990. Actuarial survival was similar for both groups of patients at 1 year (76.4% versus 81.3%) and at 2 years (69.6% versus 73.0%). Because we were interested in long-term results, we further analyzed only those patients surviving more than 1 year after transplantation (214 nondiabetic patients and 29 diabetic patients). Mean follow-up for the nondiabetic patients was 31.8 +/- 16.2 months and for the diabetic patients, 32.9 +/- 4.1 months. The respective mean age in each group was 50.4 +/- 10.3 years and 51.6 +/- 9.1 years. No difference was observed between the nondiabetic patients and diabetic patients regarding the rejection rate per patient-month (0.040 +/- 0.041 versus 0.045 +/- 0.051 episodes per patient-month), the infection rate per patient-month (0.056 +/- 0.081 versus 0.081 +/- 0.102 episodes per patient-month), or renal function as evidenced by mean creatinine levels at 1, 2, and 3 years. Twelve patients were insulin-dependent before transplantation; and 1 year after transplantation, they required an insulin dose 2.12 times greater than the dose before operation. Coronary artery disease developed in 32.8% of the nondiabetic patients, compared with 31.0% of the diabetic patients by the fourth year of follow-up. Despite the need for increased insulin doses, the diabetic patients had similar long-term survival to that of the nondiabetic patients, without an increased risk of rejection, infection, renal dysfunction, or coronary artery disease. Our experience supports the feasibility of heart transplantation in selected diabetic recipients.