Study objective: To evaluate the potential of rifaximin, an oral nonabsorbed (< 0.4%) structural analog of rifampin, to induce human hepatic and/or intestinal cytochrome P450 (CYP) 3A enzymes, with use of a known CYP3A probe, midazolam.
Design: Prospective, randomized, open-label, two-period, crossover study.
Setting: Clinical research center.
Subjects: Twenty-seven healthy adult volunteers.
Intervention: During the first treatment period, subjects received a single dose of either intravenous midazolam 2 mg over 30 minutes or oral midazolam 6 mg on day 0. From days 3-10, they received rifaximin 200 mg every 8 hours. On days 6 (after the 9th dose of rifaximin) and 10 (after the 21st dose of rifaximin), subjects received a concomitant single dose of intravenous or oral midazolam. After a 15-day washout period, subjects were crossed over to the other formulation of midazolam, and the treatment schedule was repeated, with the second treatment period starting on day 26 and single-dose administration of midazolam on days 26, 32, and 36. Serial plasma samples were collected for pharmacokinetic analyses.
Measurements and main results: The pharmacokinetic parameters of single-dose intravenous or oral midazolam were determined alone and after coadministration of rifaximin for 3 and 7 days. Rifaximin coadministration did not alter the measured pharmacokinetic parameters for midazolam or its major metabolite, 1'-hydroxymidazolam. The 90% confidence intervals for the maximum concentration and area under the concentration-time curve from time zero extrapolated to infinity (bioavailability) were all within 80-125% for intravenous and oral midazolam. Therefore, no drug interaction was observed between rifaximin and midazolam. Coadministration of midazolam and rifaximin was well tolerated.
Conclusion: Overall, 3-7 days of rifaximin 200 mg 3 times/day did not alter single-dose midazolam pharmacokinetics. Rifaximin also does not appear to induce intestinal or hepatic CYP3A activity.