Standardized EEG interpretation in patients after cardiac arrest: Correlation with other prognostic predictors

Abstract

Introduction: Standardized EEG patterns according to the American Clinical Neurophysiology Society (ACNS) ("highly malignant", "malignant" and "benign") demonstrated good correlation with outcome after cardiac arrest (CA). However, this approach relates to EEGs after target temperature management (TTM), and correlation to other recognized outcome predictors remains unknown.

Objectives: To investigate the relationship between categorized EEG and other outcome predictors, during and after TTM, at different temperatures.

Methods: In a prospective adult CA registry between 01.2014 and 06.2017, EEG at day one and two after CA were reclassified into pre-defined categories. Correlations between EEG and clinical, biochemical, neurophysiological outcome predictors, and prognosis (CPC at three months; good: 1-2), were assessed.

Results: Of 203 CA episodes, 31.5% were managed targeting 33 °C, 60.6% targeting 36 °C, and 7.9% with spontaneous temperature. "Highly malignant" EEG was found in 36.7% of patients at day one (predicting poor prognosis with 91% specificity -95%CI: 83%-97%-, and 63% sensitivity -95% CI: 53%-72%), and 27.1% at day two. "Benign" EEG occurred in 19.2% at day one (sensitivity to good prognosis: 35% -95%CI: 26%-46%-, positive predictive value: 89% -95% CI: 75%-97%), and in 33.2% at day two. Categorized EEG showed robust correlations with all prognostic predictors. Results were similar between EEGs recorded at day one or two, and, especially for poor prognosis, across TTM targets.

Discussion: Standardized EEG categorization after CA shows strong correlation with other outcome predictors, without marked variation across EEG recording time or TTM targets, underscoring its prognostic role in a multimodal approach.

Keywords: ACNS nomenclature; anoxic-ischemic encephalopathy; multimodal prognostic approach after cardiac arrest; outcome.