Abstract
Bispectral Index (BIS) has been used to monitor level of "sedation" based on the electroencephalogram (EEG). Patients evaluated for surgery to control a seizure disorder undergo Wada testing, during which one hemisphere is rendered functionally inactive after injecting a short-acting barbiturate. We surmised that the BIS values would reflect these functional changes. Eight epileptic patients were enrolled. A full array of 21 EEG electrodes and 2 BIS XP (Quatro) strips over each frontal region of the scalp were applied. The EEG was continuously recorded. BIS values from each hemisphere were recorded every minute. Angiography was performed by advancing a catheter into each internal carotid artery. Amobarbital or methohexital was injected until the patient developed a hemiparesis. The EEG confirmed a significant lateralized cortical effect of the barbiturate. Repeated measures analysis of variance was used to analyze the differences between the BIS values from monitor electrode strips placed on the left (left BIS) and the right (right BIS) sides of the head as well as the differences in the left and right BIS values before and after each injection of the barbiturate. Injection of barbiturate into either the left or right internal carotid artery produced a significant change on the 21-electrode EEG. However, there was no difference between left BIS to right BIS values (P = 0.84). With repeated injections of barbiturates, some patients became sedated. At these times, both left BIS and right BIS values decreased together before and after injection of barbiturate. The BIS monitor was unable to distinguish significant hemispheric EEG and clinical functional changes except when the patient became sedated.