Abstract 202: The Use of Bispectral Index EEG Monitoring as a Screening Tool for the Reliable Detection of Epileptic Activity in Post--Cardiac Arrest Patients

Abstract

Background: Assessment of prognosis in post[[Unable to Display Character: –]]cardiac arrest (CA) patients became very challenging since the introduction of therapeutic hypothermia (TH). Continuous EEG monitoring has been proposed to improve prognostication; however, its use is limited due to difficulties in readily interpretation. This emerges the need for a simple EEG montage. The bispectral index (BIS) monitor is a simplified EEG system, mainly calculating an index ranging from 0 (isoelectric EEG) to 100 (full consciousness) to provide information on hypnotic depth of anesthesia. The aim of the study was to validate the accuracy of simplified EEG monitoring in a CA - setting. Methods: BIS monitoring (BIS VISTA TM ) was applied to collect frontotemporal data in TH-treated CA patients. A standard 19 [[Unable to Display Character: –]] channel EEG was performed after return to normothermia. Afterwards, small EEG frames coincident with the time of full EEG registration were extracted from the BIS monitor. We asked 2 neurologists to indicate the presence of status epilepticus (SE), cerebral inactivity (CI), burst suppression (BS) or a diffuse slowing pattern (DS). In addition, these samples were analyzed by 2 inexperienced physicians, who were asked to indicate the presence of SE. Results: Thirty simplified EEG samples were analyzed. According to standard EEG, 11 patients showed a DS pattern, 3 had CI, 6 showed BS and 10 had an SE. Neurologists interpreted all samples with a high accuracy (sensitivity: 82% and specificity: 92%). Only 1 SE was missed by one neurologist. Interobserver reliability was high (kappa=0.843). High correlations were found for the comparison of full and simplified EEG for both neurologists (r=0.809). Further, the 2 inexperienced physicians identified SE with a sensitivity of 85% and specificity of 98%. Conclusion: Simplified EEG monitoring, using BIS, resulted in high accuracy of a simple classification system in post [[Unable to Display Character: –]] CA patients. Not only neurologists, but also treating physicians were capable to identify SE, which may play an important role in the early detection of SE. We suggest using BIS as a screening tool in post [[Unable to Display Character: –]] CA patients to save valuable time in the detection of SE, without replacing the need of full EEG monitoring for confirmation.