Continuous EEG Monitoring Predicts a Clinically Meaningful Recovery Among Adult Inpatients

PURPOSE: Continuous EEG (cEEG) monitoring is primarily used for diagnosing seizures and status epilepticus, and for prognostication after cardiorespiratory arrest. The purpose of this study was to investigate whether cEEG could predict survival and meaningful recovery. METHODS: The authors reviewed inpatient cEEG reports obtained between January 2013 and November 2015 and recorded demographics, preadmission modified Rankin Scale, history of preexisting epilepsy, Glasgow Coma Scale for those admitted to the intensive care unit, and EEG data (posterior dominant rhythm, reactivity, epileptiform discharges, seizures, and status epilepticus). Associations between clinical outcomes (death vs. survival or clinically meaningful recovery [inpatient rehabilitation, home-based rehabilitation, or home] vs. other [death, skilled nursing facility]) and cEEG findings were assessed with logistic regression models. P < 0.05 was considered significant. RESULTS: For 218 cEEG reports (197 intensive care unit admits), the presence of at least a unilateral posterior dominant rhythm was associated with survival (odds ratio for death, 0.38; 95% confidence interval, 0.19-0.77; P = 0.01) and with a clinically meaningful outcome (odds ratio, 3.26; 95% confidence interval, 1.79-5.93; P < 0.001); posterior dominant rhythm remained significant after adjusting for preadmission disability. Those with preexisting epilepsy had better odds of a meaningful recovery (odds ratio, 3.31; 95% CI, 1.34-8.17; P = 0.001). Treated seizures and status epilepticus were not associated with a worse mortality (P = 0.6) or disposition (P = 0.6). High Glasgow Coma Scale (≥12) at intensive care unit admission was associated with a clinically meaningful recovery (odds ratio, 16.40; 95% confidence interval, 1.58-170.19; P = 0.02). CONCLUSIONS: Continuous EEG findings can be used to prognosticate survival and functional recovery, and provide guidance in establishing goals of care.