Tacrolimus is associated with fewer acute rejections than cyclosporine, but a greater risk of new onset diabetes mellitus. When compared to no tacrolimus among nondiabetics in a large patient registry, it is associated with improved graft survival. The current study used the same patient registry to compare more correctly graft survival between nondiabetic renal transplant recipients initially immunosuppressed with either of the two most frequently used calcineurin inhibitors, tacrolimus or modified cyclosporine (CsA).We examined data provided by the United States Renal Data System (USRDS) on all first, single-organ, renal transplants to nondiabetic recipients that occurred during the years 1996 to 2000. Importantly, we then limited the study to patients on CsA (n = 7,867) or tacrolimus (n = 3,082) as the initial agent. Patients with both or neither were excluded. We used Cox proportional hazards regressions to estimate the tacrolimus-related relative risk of graft failure, controlling for other significant donor, recipient, and transplant characteristicsWe found that tacrolimus patients had graft failure rates equivalent to those of CsA patients (hazard ratio= 1.031, P = 0.631)Although tacrolimus is being used with increasing frequency, analyses of the USRDS data show no net advantage in the ultimate transplantation outcome, graft survival. Given the higher acquisition price of tacrolimus compared to CsA and the similar risk of graft failure, further studies should be conducted to define those patient groups for which tacrolimus might be cost-effective.