QUANTITATIVE EEG CORRELATIONS WITH BRAIN GLUCOSE METABOLIC-RATE DURING ANESTHESIA IN VOLUNTEERS

Background: To help elucidate the relationship between anesthetic-indu ced changes in the electroencephalogram (EEG) and the concurrent cereb ral metabolic changes caused by anesthesia, positron emission tomograp hy data of cerebral metabolism obtained in volunteers during anesthesi a were correlated retrospectively with various concurrently measured E EG descriptors. Methods: Volunteers underwent functional brain imaging using the (18)fluorodeoxyglucose technique; one scan always assessed awake-baseline cerebral metabolism (n = 7), and the other scans assess ed metabolism during propofol sedation Cn = 4), propofol anesthesia Cn = 4), or isoflurane anesthesia (n = 5). The EEG was recorded continuo usly during metabolism assessment using a frontal-mastoid montage. Pow er spectrum variables, median frequency, 95% spectral edge, and bispec tral index (BIS) values subsequently were correlated with the percenta ge of absolute cerebral metabolic reduction (PACMR) of glucose utiliza tion caused by anesthesia. Results: The percentage of absolute cerebra l metabolic reduction, evident during anesthesia, trended median frequ ency (r = - 0.46, P = 0.11), and the spectral edge (r = - 0.52, P = 0. 07), and correlated with anesthetic type (r = - 0.70, P < 0.05), relat ive beta power (r = - 0.60, P < 0.05), total power (r = 0.71, P < 0.01 ), and bispectral index (r = - 0.81, P < 0.001). After controlling for anesthetic type, only bispectral index (r = 0.40, P = 0.08) and alpha power (r = 0.37, P = 0.10) approached significance for explaining res idual percentage of absolute cerebral metabolic reduction prediction e rror. Conclusions: Some EEG descriptors correlated linearly with the m agnitude of the cerebral metabolic reduction caused by propofol and is oflurane anesthesia. These data suggest that a physiologic link exists between the EEG and cerebral metabolism during anesthesia that is mat hematically quantifiable.