Quantification of benzodiazepine-induced topographic EEG changes by a computerized wave form recognition method: application of a principal component analysis.

Topographic EEG changes with medazepam and diazepam in normals were analyzed by the computerized wave form recognition method. A principal component (PC) analysis, using such EEG elements as wave percent-time (8 bands) and average amplitude (7 bands), resulted in a considerable reduction of variables (4 PCs). In O1, because of high positive loadings in the average amplitude in all bands and a decrease in the mean score with either drug, PC-1 represents a component which reacts in the form of diminution of average amplitude as a whole. In Fp1, C3 and O1, PC-2, with a bipolarity of alpha 2 versus beta 1 and beta 2 in the wave percent-time in loading profile, could be a component showing characteristic changes common to the 2 benzodiazepines. In C3, because of a significant difference in the mean score between the 2 drugs, PC-4 might be a between-drug difference component in which diazepam (medazepam) increases (decreases) slow activity. The relationship between the score at PC-4 in C3 and daytime sleepiness may signify that the slow components are associated with sedation. Based on the correlation at PC-2 in Fp1, a marked increase in beta 1 and beta 2 components (responder) rather means less sleepiness, and relative preservation of alpha 1 and alpha 2 (non-responder) more sleepiness.