Flumazenil reduces midazolam-induced cognitive impairment without altering pharmacokinetics.

OBJECTIVES: Intravenous midazolam is used as an in vivo biomarker of hepatic cytochrome P450 (CYP) 3A activity. Midazolam is a central nervous system depressant and can produce cognitive impairment. The purpose of this study was 2-fold: (1) to determine whether administration of intravenous flumazenil given before intravenous midazolam minimizes cognitive impairment and (2) to determine whether flumazenil pretreatment has an effect on midazolam pharmacokinetics during hepatic CYP3A phenotyping. METHODS: Eleven healthy subjects (8 men) received intravenous flumazenil (0.005 mg/kg) or placebo followed 7 minutes later by intravenous midazolam (0.025 mg/kg) in a randomized, double-blind crossover study. Plasma midazolam concentrations were obtained before dosing and at 5, 30, 60, 120, 240, 300, and 360 minutes after dosing and were assayed by liquid chromatography-tandem mass spectrometry. Midazolam pharmacokinetics were determined by noncompartmental methods. The two 1-sided tests procedure was used to compare area under the curve (AUC) between study phases. Data were log-transformed before analysis, and bioequivalence criteria were applied. Digit symbol substitution tests, performed before dosing and at 5, 30, 60, 120, 240, 300, and 360 minutes after dosing, were used to measure cognition. General linear modeling was used to compare scores between study phases. RESULTS: Midazolam AUC extrapolated to infinity [AUC(0-infinity)] between phases was bioequivalent. The AUC ratio (flumazenil plus midazolam/midazolam) was 0.99, with a 90% confidence interval of 0.98 to 1.00. Statistically significant differences(P