Given the prominent role of CYP3A in the metabolism of drugs, it is important to identify whether new chemical entities will affect this enzyme system and produce clinically relevant drug interactions. This study evaluated concomitant administration of intravenous [14C N-methyl] erythromycin (3 microCi) (erythromycin breath test; EBT) and 2 mg oral midazolam as probes of systemic and of systemic plus presystemic CYP3A activity, respectively. Twelve males received the probes in a two-period crossover fashion: one period included the probes on two occasions, 5 days apart; in the second period, 200 mg ketoconazole was given orally 2 hours prior to the probes. The within-subject CV for EBT (%14CO2/h) and midazolam AUC0-last was 4.9% and 16.9%, respectively. Ketoconazole reduced %14CO2/h by 43% and increased midazolam AUC0-last by approximately fivefold. In a nonrandomized third period (N = 5), ketoconazole was given simultaneously with midazolam (no EBT); midazolam AUC0-last was similar whether ketoconazole was given 2 hours prior to or simultaneously with the midazolam. The low midazolam dose was generally well tolerated; mild sedation was occasionally seen. Concurrent administration of the EBT and oral midazolam is a sensitive and reproducible tool to screen new chemical entities for potentially important CYP3A interactions.