The effect of intravenous flumazenil 10mg on the electroencephalogram (EEG) was investigated in 7 volunteers in a placebo-controlled, randomised, double-blind, crossover study. The EEG was recorded between Fp1-M1 and Fp2-M2 and analysed using an aperiodic analysis technique. Two volunteers were excluded from the study because of significant asymmetry between baseline EEG recordings of the left and right hemisphere; in the remainder there were no changes in the beta-frequency range (12 to 30Hz) or in other ranges of the EEG during or after flumazenil or placebo administration, with regard to the parameters total number of waves per second or total amplitude per second. There were no changes in heart rate, respiratory rate or blood pressure after administration of flumazenil or placebo. Three volunteers reported feelings of 'pressure to move' during the initial 2 min of the flumazenil infusion. The pharmacokinetics of flumazenil were investigated in the same volunteers. Flumazenil 10mg was administered intravenously over 10 min; the data from 1 volunteer were excluded from this analysis because of blood sampling problems. The plasma concentration-time data of the remaining 6 volunteers were characterised by a biexponential function. The pharmacokinetic parameters were (mean +/- SD): initial volume of distributions, 16 +/- 5.7L; volume of distribution at steady-state, 64.8 +/- 12.5L; total body clearance, 53.8 +/- 1.2 L/h; distribution half-life, 4.1 +/- 1.3 min; and elimination half-life, 70.2 +/- 9.9 min. The authors conclude that flumazenil has no significant EEG effects. The rapid distribution and elimination of flumazenil may explain its previously reported short duration of action after intravenous anaesthesia with high doses of midazolam.