Differing manifestations of hepatitis C and tacrolimus on hospitalized diabetes mellitus occurring after kidney transplantation

PURPOSE: Previous studies suggest the association of recipient hepatitis C seropositivity (HCV +) and use of tacrolimus (TAC) with post-transplant diabetes mellitus (PTDM) may differ by manifestations of type I or type II diabetes, but this has not been assessed in the era of current immunosuppression. METHODS: We performed a retrospective cohort study of 10,342 Medicare primary renal transplantation recipients without evidence of diabetes at the time of listing in the United States Renal Data System between January 1, 1998 and July 31, 2000, followed until December 31, 2000. Outcomes were hospitalizations for a primary diagnosis of diabetic ketoacidosis (DKA) or hyperglycemic hyperosmolar syndrome (HHS). Cox regression analysis was used to calculate adjusted hazard ratios (AHR) for time to DKA or HHS, stratified by diabetes status at the time of transplant. RESULTS: In Cox regression analysis, use of TAC at discharge was independently associated with shorter time to DKA (AHR, 1.88; 95% CI, 1.05-3.37, p = 0.034) but not HHS. In contrast, recipient HCV+ was independently associated with shorter time to HHS (AHR, 3.90; 1.59-9.60, P = .003), but not DKA. There was no interaction between TAC and HCV+ for either outcome. CONCLUSION: These results confirm earlier findings that TAC and HCV+ may mediate the risk of PTDM through different mechanisms, even in the modern era.