Intramuscular administration of lidocaine or bupivacaine alters the effect of midazolam from sedation to hypnosis in a dose-dependent manner.

We examined the sedative/hypnotic interaction between the administration of intravenous (i.v.) midazolam and intramuscular (i.m.) lidocaine or bupivacaine. Women undergoing gynecological surgery (n = 150) were randomly assigned to 15 dose groups of 10 patients each. Fifty patients received one of five predetermined doses of midazolam for the calculation of its median effective dose (ED50). The remaining patients (n = 100) received i.v. midazolam 0.1 mg/kg following an i.m. injection of either bupivacaine, lidocaine, or saline (control). Three minutes after the i.v. dose, the loss of response to verbal command was evaluated. The ED50 of midazolam was 0.226 mg/kg (95% confidence interval [CI] 18-027; p = 0.03). Both bupivacaine and lidocaine enhanced the effect of midazolam in a dose-dependent fashion. The hypnotic ED50 for bupivacaine and lidocaine was 0.7 mg/kg (95% CI 0.5-1.0) and 3.32 mg/kg (95% CI 2.2-11.7), respectively. The slopes of the dose-response curves were significantly different (p < 0.01). Local anesthetics that are well within the range of clinical use for regional blocks or local infiltration can bring the effect of midazolam from the sedative into the hypnotic range.