Intravenous flumazenil following acute and repeated exposure to lorazepam in healthy volunteers: antagonism and precipitated withdrawal.

The effects of i.v. administered flumazenil (3.0 mg) were studied in healthy male subjects who received pretreatment with p.o. placebo or lorazepam. The duration of placebo or lorazepam (3.0 mg single p.o. daily dose) pretreatment before a flumazenil or placebo injection was 1, 3, 7 or 14 days in four sequential groups of subjects. Initial administration of lorazepam produced a classic sedative profile of effects on various psychomotor/behavioral performance, observer-rated and subject-rated measures. Tolerance to repeated daily administration of lorazepam was suggested by a progressive diminution of performance disrupting effects. In subjects pretreated with placebo, flumazenil increased subject-ratings of dizziness over preinjection ratings. Flumazenil produced an immediate reversal of lorazepam effects in subjects who were not tolerant to lorazepam (1- and 3-day pretreatment groups). Flumazenil did not precipitate withdrawal symptoms in subjects who received a single administration of lorazepam. Precipitated withdrawal symptoms were evident after 3 and 7 days of lorazepam pretreatment, and there was a tendency toward precipitated withdrawal symptoms (that included one panic attack) after 14 days of lorazepam pretreatment. Precipitated withdrawal was characterized by an elevation in subject-rated symptoms including dizziness, tenseness, tachycardia, perceptual disturbance and sweating. Symptoms were maximal immediately after injection, usually mild in severity and usually resolved within 1 hr. There was no evidence of precipitated withdrawal on psychomotor/behavioral performance or observer ratings. The present study provides the strongest human experimental evidence to date that flumazenil can precipitate withdrawal symptoms after a history of repeated benzodiazepine exposure.