The pharmacokinetics of diacerein following a single oral dose of 50 mg was studied in 12 healthy volunteers, 10 patients with a mild liver cirrhosis (Child Pugh's grade A), and 6 patients with a more severe liver cirrhosis (Child Pugh's grade B to C). Statistical analysis using a Kruskal-Wallis test showed no significant differences between the three groups for the following parameters: median Cmax was 3.9 mg l-1 for the cirrhotic patients group I (CPI) and 3.2 mg l-1 for the cirrhotic patients group II (CPII) versus 3.2 mg l-1 for the healthy volunteers (HV); median t1/2 was 4.9 h for CPI and 4.3 h for CPII versus 4.3 h for HV; median Cl/F was 2.1 l h-1 for CPI and 2.5 l h-1 for CPII versus 1.6 l h-1 for HV; median Vdss/F was 12.6 l for CPI and 14.0 l for CPII versus 13.21 for HV. The urinary parameters were comparable. It was concluded that, from a pharmacokinetic point of view, no reduction in the initial dosage of diacerein need be proposed in liver cirrhosis.