Use of bispectral electroencephalogram monitoring to assess neurologic status in unsedated, critically ill patients

Objective. To test whether spectral indices derived from the electroencephalogram (EEG), and especially the bispectral index (BIS), can be used as measures of neurologic status in unsedated, critically ill patients. Design. Prospective, observational study. Setting. Medical intensive care unit (ICU) of a university-affiliated teaching hospital. Patients. Thirty-one awake, unsedated critically ill adults were assessed in 108 separate sessions. Measurements and Main Results. In each session, severity of illness was assessed by the Acute Physiology and Chronic Health Evaluation (APACHE III). The APACHE III Acute Physiology Score was used to quantify the degree of physiologic derangement. Neurologic function was assessed using the APACHE III Neurologic Score, the Glasgow Coma Scale, the Reaction Level Scale, and the Modified Ramsay Sedation Scale. All indices were plotted against various spectral parameters of the EEG, including BIS, an empirical index of EEG activity that is scaled from 0 to 100. BIS was significantly (p < .05) correlated with neurologic score regardless of scoring system used and was more strongly correlated than any other EEG spectral parameter. Better neurologic function was associated with higher values of BIS. In multivariate analysis, the combination of BIS and relative power in the theta band of the EEG accounted for 38% of the variability in the Glasgow Coma Scale. Conclusions. BIS provides a reliable index of neurologic status in awake, unsedated, critically ill patients. Further research is needed to determine whether the effects of neurologic status and pharmacologic sedation upon EEG are additive, whether BIS can be used to assess pharmacologic sedation in the critically ill patient population, and whether such objective measures of neurologic status have prognostic value.