EEG reactivity testing for prediction of good outcome in patients after cardiac arrest.

Abstract

To determine the additional value of EEG reactivity (EEG-R) testing to EEG background pattern for prediction of good outcome in adult patients after cardiac arrest (CA). In this post hoc analysis of a prospective cohort study, EEG-R was tested twice a day, using a strict protocol. Good outcome was defined as a Cerebral Performance Category score of 1-2 within 6 months. The additional value of EEG-R per EEG background pattern was evaluated using the diagnostic odds ratio (DOR). Prognostic value (sensitivity and specificity) of EEG-R was investigated in relation to time after CA, sedative medication, different stimuli, and repeated testing. Between 12 and 24 hours after CA, data of 108 patients were available. Patients with a continuous (n = 64) or discontinuous (n = 19) normal voltage background pattern with reactivity were 3 and 8 times more likely to have a good outcome than without reactivity (continuous: DOR, 3.4; 95% confidence interval [CI], 0.97-12.0; p = 0.06; discontinuous: DOR, 8.0; 95% CI, 1.0-63.97; p = 0.0499). EEG-R was not observed in other background patterns within 24 hours after CA. In 119 patients with a normal voltage EEG background pattern, continuous or discontinuous, any time after CA, prognostic value was highest in sedated patients (sensitivity 81.3%, specificity 59.5%), irrespective of time after CA. EEG-R induced by handclapping and sternal rubbing, especially when combined, had highest prognostic value. Repeated EEG-R testing increased prognostic value. EEG-R has additional value for prediction of good outcome in patients with discontinuous normal voltage EEG background pattern and possibly with continuous normal voltage. The best stimuli were clapping and sternal rubbing.