Employing a randomized study design, the intra- and post-operative effects of three sedative agents were compared in 30 female patients under regional anaesthesia for lower abdominal surgery. After establishing epidural blockade to a level of T4, 10 patients were sedated with propofol 2.79 +/- 0.4 mg kg-1 h-1, 10 with midazolam 0.08 +/- 0.01 mg kg-1 h-1, and a further 10 patients with methohexitone 1.8 +/- 0.2 mg kg-1 h-1 (mean dose +/- SEM). An additional 10 patients did not receive any sedative agent and served as a control group. There was excellent haemodynamic stability in all groups. They were comparable with respect to respiratory rate, SpO2, and PaCO2. Testing of vigilance showed significant differences between peri-operative and baseline scores and those found in the recovery room in all three groups with intra-operative sedation; in the midazolam group vigilance was still impaired the day following surgery. There were significant differences in the post-operative vigilance scores between patients sedated during surgery and the control group; in the methohexitone group even 24 h later. The ability to concentrate was particularly impaired in patients treated with methohexitone. Patients treated with midazolam or methohexitone needed longer to recall their date of birth than those sedated with propofol.