Based on uncontrolled observations, we have proposed ursodeoxycholic acid (UDCA) as a novel therapeutic approach in primary biliary cirrhosis (PBC). To confirm and extend our original findings, we have designed a double-blind multicentre randomized clinical trial. An interim analysis was planned at 6 months, involving all subjects included in the trial, with a final analysis at 2 years. The UDCA-PBC trial began in June 1987 and will be completed in March 1990. Seventy patients were randomized to receive UDCA and 68 a placebo. The two groups were well matched with respect to age, sex, duration and prevalence of symptoms and histologic severity (50% of the UDCA group had stage III-IV disease vs. 37% of the placebo group). During the first 6 months of follow-up, six patients withdrew from the trial. At 6 months, the proportion of patients with jaundice was significantly lower (p less than 0.01) in UDCA recipients than in the placebo group. There was a similar decrease in the proportion of patients with pruritus and fatigue in both groups. The following laboratory test values were significantly lower in UDCA recipients than in the placebo group after 6 months of therapy: serum bilirubin, alkaline phosphatase, alanine aminotransferase (ALAT), aspartate aminotransferase (ASAT), gamma-glutamyltranspeptidase activities (p less than 0.001), cholesterol (p less than 0.003) and IgM levels (p less than 0.03). The results of this interim analysis confirm and extend the biochemical data provided by our previous pilot study. However the final analysis of the trial is necessary for a definitive assessment of the safety and efficacy of UDCA therapy in PBC.