There is a paucity of data examining the efficacy of valganciclovir (VGC) for cytomegalovirus (CMV) prophylaxis in kidney transplant patients, particularly with regard to utilization of a risk-stratified dosing regimen. Eighty adult African-American (AA) renal allograft recipients transplanted from November 3, 2001 to May 28, 2003 and followed for 22 +/- 8 months received VGC once daily for 90 d post-transplant dosed according to donor/recipient (D/R) serostatus: high risk (D+/R-) received 900 mg (n = 12); moderate risk (D+/R+, D-/R+) received 450 mg (n = 60); and low risk (D-/R-) received no prophylaxis (n = 8). Thymoglobulin or basiliximab was used for induction, and mycophenolate mofetil, prednisone, and either tacrolimus or sirolimus for maintenance immunosuppression. Only six patients (7.5%) developed symptomatic CMV infection diagnosed by pp65 antigenemia, three in the high-risk (25%) and three in the moderate-risk (5%) group (p = 0.02). All patients were on tacrolimus for at least 3 months prior to diagnosis. There were no cases of tissue-invasive disease, resistance to treatment, or recurrence. D+/R- serostatus was the only significant independent predictor for CMV infection using multivariate analysis (odds ratio 10.5; p = 0.04). Thymoglobulin induction was not associated with CMV infection. None of 43 patients who were exposed to sirolimus for >30 d developed CMV infection, vs. six of 37 who were not (p = 0.006). We conclude that VGC dosed according to D/R serostatus provides safe and effective CMV prophylaxis in AA renal allograft recipients.