Fifty-five cases of ketoconazole-associated hepatic injury, reported to the Netherlands Centre for Monitoring of Adverse Reactions to Drugs, were analysed in detail. In 50 cases a causal relationship was considered likely, i.e. 'probable' (27 cases) or 'possible' (23 cases). Eighty-four % of individuals were women. Forty-six % of patients were over 50 years of age which suggests that, considering the lower prescription rate in this age group, the elderly are more vulnerable to ketoconazole. In 60% of all cases hepatic injury appeared within the first 6 weeks of therapy but in the group of 'probable'-cases the onset was mostly later. Jaundice was present in 44% of all cases but in 63% of the group of 'probable'-cases. Eosinophilia (10%), fever (6%) and rash (2%) were uncommon. Biochemically the pattern was hepatocellular in 54%, cholestatic in 16% and mixed cholestatic-hepatocellular in 30%. Histology (14 cases) showed a predominantly hepatocellular pattern in 57% with extensive centrilobular necrosis and mild to moderate bridging. In 43% cholestasis predominated. None of the cases had a fatal course. The incidence of symptomatic hepatic injury may be estimated at approximately 1:2000 but is probably higher. The mechanism of ketoconazole-induced hepatic injury seems to be based on metabolic idiosyncrasy although it is not excluded that in some patients an immunoallergic mechanism is causative.