Efficacy without tolerance or rebound insomnia for midazolam and temazepam after use for one to three months.

Midazolam (15 mg) was compared with temazepam (30 mg) in a randomized, double-blind, parallel group study. An initial screening period was followed by 3 days of placebo baseline, 4 to 12 weeks of nightly oral use of the medication and a 4-day placebo withdrawal period. One hundred seventy-five patients with chronic insomnia participated in this multicenter outpatient study. Because the elimination half-life of midazolam, a new trizolobenzodiazepine hypnotic, is short (1.3-2.2 hr) compared to temazepam's (12-16 hr), more problems with tolerance and rebound insomnia were expected to occur. Hypnotic efficacy (increased total sleep time, decreased wake time, and decreased sleep latency) was demonstrated for both medications over the entire 3-month period without the development of tolerance. In fact, if anything, efficacy increased with time on medication, suggesting possible facilitation or "inverse tolerance" effect. On withdrawal, sleep was improved compared with baseline, suggesting partial resolution of the insomniac condition rather than rebound insomnia. These effects were both statistically and clinically significant for midazolam, with 16% to 50% improvement in sleep measures. The results of this study suggest that patients with chronic insomnia may benefit from 30 to 90 days of treatment. A three-factor model that separates pharmacologic from behavioral and psychologic effects of hypnotics was proposed to explain these results in part.