A comparative pharmacokinetic and pharmacodynamic investigation was carried out on the interaction between the hypnotic midazolam and 2 different macrolide-type antibiotics, clarithromycin and azithromycin. In an open randomized crossover study of 3 phases 12 healthy volunteers received either clarithromycin (250 mg twice a day for 5 days), azithromycin (500 mg once a day for 3 days) or no pretreatment. On the last day of antibiotic treatment they ingested 15 mg midazolam. Plasma samples were collected for midazolam analysis up to 24 h and pharmacodynamic performance measured by a series of tests up to 12 h. Pretreatment with clarithromycin caused large and statistically significant changes in both the pharmacokinetic and pharmacodynamic parameters of midazolam compared to control. For example, the AUC was increased from 248.84-888.75 hng/ml (factor of 3.57, p < 0.0001) and the mean duration of sleep increased from 135.4 min to 281.3 min (p < 0.05). No statistically significant effect was found with azithromycin in any test. It is concluded that a drug interaction exists between midazolam and clarithromycin which could be of clinical importance. No such effect is present with azithromycin.