Chronopharmacokinetic study with prolonged infusion of midazolam.

Physiological and temporal variation in the disposition of midazolam has been reported. In order to delineate the underlying mechanisms of these alterations, we infused in 5 healthy male volunteers for 26 hours midazolam at a rate of 0.025 mg/kg/h preceded by a bolus of 0.05 mg/kg. Thus, steady-state conditions were rapidly achieved. Plasma levels of midazolam were monitored on a 2-hourly basis during this period. In addition, the pharmacodynamic response to the new sedative/hypnotic benzodiazepine was characterised by a pencil tracking test, sedation index formed from visual analogue scales, and choice reaction time. In all subjects, small (4 to 16%) but clinically irrelevant fluctuations of steady-state plasma concentrations around 45 ng/ml were observed. During the night-time (11pm to 7am) plasma concentrations were slightly (p = 0.074) higher than during the daytime. Total plasma clearance varied from 563 to 823 ml/min. Plasma protein binding of midazolam was time independent. Since in only 1 of 5 subjects was a circadian rhythm observed, fluctuations in plasma midazolam concentrations under controlled and constant conditions are probably not of clinical significance.