EEG and auditory evoked potential P300 compared with psychometric tests in assessing vigilance after benzodiazepine sedation and antagonism.

We have compared the EEG and auditory evoked wave P300 with psychometric tests in assessing vigilance after flumazenil antagonism of midazolam sedation in 12 healthy volunteers. Measurements were made before and after midazolam 0.1 and 0.2 mg kg-1 i.v., and immediately and 30, 60, 120 and 240 min after administration of flumazenil 1 mg. The sedative effects of midazolam and antagonism by flumazenil resulted in alterations in EEG, P300 and psychometric tests (syndrome short test, letter cancellation, choice reaction and recognition). However, 60 and 120 min after flumazenil a decrease in test performance indicating rebound sedation was seen only in P300 mapping. Thus P300 mapping was a sensitive method of detecting subtle differences in vigilance. Rebound sedation occurred even when midazolam 0.2 mg kg-1 was antagonized with an adequate dose of flumazenil. We suggest that it is advisable to supervise patients for at least 240 min after flumazenil antagonism of midazolam 0.2 mg kg-1.