Multimodal Outcome Prognostication After Cardiac Arrest and Targeted Temperature Management: Analysis at 36°C.

Abstract

Targeted temperature management (TTM) represents the standard of care in comatose survivors after cardiac arrest (CA) and may be applied targeting 33° or 36°C. While multimodal prognostication has been extensively tested for 33°C, scarce information exists for 36°C. In this cohort study, consecutive comatose adults after CA treated with TTM at 36°C between July 2014 and October 2016 were included. A combination of neurological examination, electrophysiological features, and serum neuron-specific enolase (NSE) was evaluated for outcome prediction at 3months (mortality; good outcome defined as cerebral performance categories (CPC) score of 1-2, poor outcome defined as CPC 3-5). We analyzed 61 patients. The presence of two or more predictors out of, unreactive electroencephalogram (EEG) background, epileptiform EEG, absent pupillary and/or corneal reflex, early myoclonus, bilaterally absent cortical somatosensory evoked potentials, and serum NSE>75μg/l, had a high specificity for predicting mortality (positive predictive value [PPV]=1.00, 95% CI 0.87-1.00) and poor outcome (PPV=1.00, 95% CI 0.80-1.00). Reactive EEG background was highly sensitive for predicting good outcome (0.95, 95% CI 0.74-0.99). Prediction of outcome after CA and TTM targeting 36°C seems valid in adults using the same features tested at 33°C. A reactive EEG under TTM appears highly sensitive for good outcome.