Eight healthy volunteers received a single 5-mg intravenous dose of the imidazobenzodiazepine derivative midazolam on three occasions in random sequence: a, control, with no other treatment; b, during coadministration of cimetidine, 300 mg every 6 hr; c, during coadministration of ranitidine, 150 mg every 12 hr. Midazolam kinetics in each trial were determined from multiple plasma midazolam levels measured by gas chromatography for 24 hr after each dose. High pressure liquid chromatography analysis of plasma also verified compliance with cimetidine (mean level, 0.61 microgram/ml) and ranitidine (mean level, 0.36 microgram/ml) regimens. Analysis of variance indicated no significant differences in mean values for trials a, b, and c in midazolam elimination half-life (2.25 vs 2.02 vs 2.05 hr), volume of distribution (2.13 vs 2.14 vs 2.16 L/kg) or total clearance (10.8 vs 12.2 vs 12.3 ml . min-1 . kg-1). In a second study, six subjects received a 15-mg oral dose of midazolam on three occasions identical to those described above. Again, there were no significant differences among trials a, b, and c in midazolam peak plasma level (90 vs 95 vs 117 ng/ml), time of peak level (0.65 vs 1.45 vs 0.90 hr after dose), elimination half-life (3.04 vs 3.38 vs 3.30 hr), or apparent oral clearance (16.2 vs 14.3 vs 13.8 ml . min-1 . kg-1). Thus the usual therapeutic doses of cimetidine or ranitidine do not significantly alter the kinetics of intravenous or oral midazolam in healthy individuals.