The influence of renal impairment on the clearance of the new HMG-CoA reductase inhibitor cerivastatin was evaluated. A single oral dose of 300 microg cerivastatin was given to 18 patients with different degrees of renal impairment and 6 healthy controls. Concentrations of total cerivastatin, its fraction unbound, and the total concentrations of the active metabolites M1 and M23 were measured in plasma. Serum concentrations of unbound cerivastatin were calculated for each individual from the concentration of total cerivastatin and cerivastatin's fraction unbound at t = 2.5 hours. In contradiction to what had been expected, renal impairment significantly influenced the pharmacokinetics of cerivastatin. The best correlation to the AUC and Cmax of unbound cerivastatin was found with serum albumin concentration. Also, serum albumin concentration was the only factor significantly correlated to t 1/2 of cerivastatin. Significant but slighter correlation with the AUC and Cmax of unbound cerivastatin was also observed for creatinine clearance and cerivastatin's fraction unbound, while no correlation was observed with total plasma protein. No significant correlation of creatinine clearance, serum albumin concentration, fu, or total plasma protein concentration with the AUC and Cmax of total cerivastatin or the AUC, Cmax or t 1/2 of M1 and M23 was observed. The authors conclude that low serum albumin concentration rather than low creatinine clearance predicts the pharmacokinetics of cerivastatin in renal impairment.