Prolonged elimination of midazolam after i.m. administration

The elimination pharmacokinetics of midazolam after i.m. administration was compared with combined i.m. and i.v. administration in a randomized study of 55 gynaecological patients in outpatient general anaesthesia. Group 1 (n = 40) received midazolam 0.1 mg/kg i.m. as premedication 45 min before induction of general anaesthesia with midazolam 0.3 mg/kg i.v. Group 2 (n = 15) received midazolam 0.1 mg/kg i.m. as premedication 45 min before induction of general anaesthesia with thiopentone 4 mg/kg. Serum midazolam concentration measurements were performed regularly post-induction for 7 h in each patient. The elimination half-life of midazolam after i.m. administration (Group 2) was 6.6 +/- 1.2 h (mean +/- s.e. mean), which was significantly longer (P less than 0.05) than the 3.9 +/- 0.3 h observed after the combined i.m. and i.v. administration of midazolam (Group 1), and significantly longer than 2.9 h obtained from a calculated i.v. administration curve. We postulate a slow i.m. depot release of midazolam, representing the rate-limiting step in the elimination of midazolam after i.m. administration.