This is the first multicenter, randomized, open-label study to compare the efficacy and safety of cyclosporine A microemulsion (CsA-ME) (Neoral, Novartis, Basel, Switzerland ) with C2 monitoring versus tacrolimus in de novo liver transplant recipients. Patients were stratified according to hepatitis C virus status and randomized to receive CsA-ME (n= 250) or tacrolimus (n= 245) with steroids, with or without azathioprine. The primary endpoint was the incidence of biopsy-proven acute rejection (BPAR) at 3 months. Secondary endpoints included death or graft loss and safety evaluations at 6 months. The incidence of BPAR at 3 months was 26% in the CsA-ME group and 24% in the tacrolimus group (not significant). At 6 months, 89% of patients receiving CsA-ME and 88% of patients receiving tacrolimus were alive with a functioning graft. Among the hepatitis C virus-positive patients, there was no difference in BPAR, but death or graft loss was more frequent in those receiving tacrolimus (15% vs. 6%, P <0.05). Diabetes mellitus (14% vs. 7%, P <0.02) and diarrhea (29% vs. 14%, P <0.001) were significantly more often reported in patients receiving tacrolimus. The incidence of hypertension was similar in both groups. At 6 months, the median total cholesterol was 4.7 mmol/L (2.9-7.4 mmol/L) in the CsA-ME arm versus 4.3 mmol/L (2.5-6.4 mmol/L) in the tacrolimus arm; the median serum creatinine was 106 micromol/L (52-238 micromol/L) in the CsA-ME arm versus 103 micromol/L (44-477 micromol/L) in the tacrolimus arm. Efficacy is equivalent with CsA-ME using C2 monitoring or tacrolimus in liver transplant recipients. The incidence of adverse events is comparable except for a significantly higher incidence of diabetes mellitus and diarrhea in the tacrolimus group. Both agents are effective primary immunosuppressants in liver transplant recipients.