Antagonism of midazolam sedation by flumazenil: a placebo-controlled study in patients recovering from intravenous anaesthesia with high doses of midazolam.

Twenty adult surgical patients were anaesthetized with high-dose midazolam and alfentanil by infusion, vecuronium, and intubated and ventilated with 50% N2O in O2. The midazolam and alfentanil infusions were stopped at the end of surgery. Residual neuromuscular blockade and ventilatory depression were antagonized and the patients extubated. In the recovery room, patients were randomly allocated to receive either flumazenil 1 mg of placebo i.v. Before, and until 2 h after injection, patients were asked to perform psychomotor tests. In addition, sedation, comprehension and orientation were scored. The flumazenil (n = 10) and the placebo (n = 10) groups were comparable. Prior to injection all patients were heavily sedated. After flumazenil all were awake within 2-3 min, but fell asleep again 15-60 min later. The improvement in test scores was sustained for a longer time. After placebo, patients awoke in 1-2 h. At 60 and 120 min, test scores in the two groups were similar. Heart rate, blood pressure and respiration rate did not change. No side-effects were observed or reported. It is concluded that flumazenil is an effective and safe antagonist of high dose midazolam, with a rapid onset but a short duration of action.