OBJECTIVE: The aim of this study was to assess the safety and efficacy of high-dose ursodeoxycholic acid (UDCA, 28-32 mg/kg/day) in patients with primary biliary cirrhosis (PBC) who had shown an incomplete response to the standard dose (13-15 mg/kg/day). METHODS: A total of 25 patients with PBC who had been on UDCA (13-15 mg/kg/day) therapy for 24-141 months and had shown persistent elevation of ALP activity at least two times the upper limit of normal were enrolled. The dose of UDCA was increased to 30 (28-32) mg/kg/day and given for 1 yr. RESULTS: A significant but marginal improvement in serum ALP activity (707+/-52 vs 571+/-32, p = 0.001) was noted at 1 yr of treatment with high-dose UDCA. However, levels of total bilirubin (1.1+/-0.2 vs 1.0+/-0.2, p = 0.1), AST (58+/-9 vs 54+/-1, p = 0.1), albumin (4.1+/-0.7 vs 4.0+/-0.08, p = 0.1), or Mayo risk score (4.13+/-0.3 vs 4.12+/-0.3, p = 0.2) remained essentially unchanged. Normalization of liver tests did not occur in any patient, and adverse events were not recorded in any case. CONCLUSIONS: Although UDCA at a dose of 28-32 mg/kg/day is well tolerated, this dosage does not seem to benefit most patients with PBC responding incompletely to a dose of 13-15 mg/kg/day. The results of this pilot study would seem to discourage further controlled trials of high-dose UDCA in suboptimal responders to the standard dose of UDCA.