BACKGROUND: Many studies indicated that the predictive accuracy of propofol TCI may be compromised by premedication with benzodiazepine which has been shown to reduce markedly the induction dose. This study was designed to examine the influence of midazolam premedication on certain parameters of treatment using the propofol TCI. METHODS: One hundred and sixty ASA I or II patients undergoing elective surgery were randomly allocated to two groups according to premedication: Group 1, glycopyrrolate 0.2 mg; Group 2, glycopyrrolate 0.2 mg and midazolam 0.07 mg/kg IM 1hr before induction. Each group divided to four subgroups (n=20 for each subgroup) according to expected target propofol concentration (4~7 microgram/ml for group 1 and 3~6 microgram/ml for group 2). Anesthesia induction within 3 min was considered as successful. Induction dose and time, success rate of induction, calculated concentration when successful induction, context sensitive decrement time when awakening concentration was 1.2 microgram/ml and side effects were checked. RESULTS: Successful induction rate was higher in group 2 (53.3% vs 77.8% at target concentration of 5 microgram/ml, P<0.05). Mean target concentration of propofol were lower in group 2 (5.18 vs 3.87 microgram/ml, P<0.05). Induction time and dose were decreased 48.4% and 36.8% at target concentration of 4 g/ml, respectively in group 2 (P<0.05). Vital signs, average pain score and incidence of pain showed no differences between groups, but incidence of apnea was significantly increased in group 2 (P<0.05). CONCLUSION: Group 2 showed better quality of propofol induction using a TCI in terms of induction time, induction dose and lower selected target without significant vital sign changes, but showed increased incidence of apnea compared with group 1.