The administration of propofol by infusion for maintenance of anesthesia has attracted much attention recently. We investigated the necessary infusion rate of propofol to maintain anesthesia for short surgical procedures without loss of the evident advantages of this substance. Forty unpremedicated female patients aged 18-59, scheduled for minor gynecological procedures, were randomly assigned to four groups. Anesthesia was induced with 2.0 mg/kg propofol i.v. and simultaneously an infusion of 0.05, 0.10, 0.15, or 0.20 mg propofol/kg per minute was started. The patients were breathing N2O/O2 with FIO2 33%. Additional propofol was administered as a bolus of 10 to 20 mg when the patients moved. With 0.05 mg propofol/kg per minute all patients required additional bolus injections of propofol; with 0.10 mg 8 patients, with 0.15 mg 5 patients, and with 0.20 mg 1 patient required bolus injection. Therefore, 0.15 mg/kg per minute can be considered as an approximate ED50 value. The total propofol consumption (infusion + bolus) increased from 0.102 +/- 0.028 (+/- SD) with the lowest infusion rate to 0.202 +/- 0.006 mg/kg per minute with the highest infusion rate and recovery time from 5.2 +/- 1.4 to 9.9 +/- 2.6 min. There was a significant correlation between propofol consumption and recovery time. After induction, arterial blood pressure decreased by systolic/diastolic 20/10-15 mmHg. With the low infusion rate, arterial pressure increased to its control value during operation; it remained at the postinduction value with high infusion rates. Side-effects: 10 patients had salivation that in some instances lead to coughing, 9 reported pain at the injection site during induction, and 9 reported dreams of a pleasant nature.(ABSTRACT TRUNCATED AT 250 WORDS)