Quantitative parameters of EEG suppression and epileptiform activity as a prediction of neurological outcome after out-of-hospital cardiac arrest

Introduction. The prognosis after cardiac arrest (CA) is mainly determined by the extent of hypoxic-ischaemic encephalopathy developed under circulatory arrest. EEG characteristics associated with poor outcome include suppression1 and generalized epileptiform activity.2 Burst-suppression ratio (BSR) for the quantification of suppressed EEG and wavelet subband entropy (WSE)3 for the epileptiform activity were monitored in this study. Methods. We prospectively studied 30 comatose adult patients, resuscitated from a witnessed ventricular fibrillation CA, and treated with hypothermia (338C, midazolam, fentanyl, and pancuronium for 24 h). EEG recording started on arrival to ICU. BSR and WSE were scored offline. Neuron-specific enolase (NSE), protein S-100b, and transcranial Doppler (TCD) were measured. Within a 6 month follow-up period, the neurological outcome was assessed as good or poor by the five-category Pittsburgh Outcome Scale that defines cerebral performance. Results. Twenty-one patients had a best-achieved cerebral performance of 1 or 2 (good outcome). BSR was lower in the good prognosis group (18.21 vs 64.95, P¼0.01 Mann-Whitney U-test). WSE was higher in the group with good prognosis (0.84 vs 0.82, P¼0.05) 24-48 h after CA (Table 6). Discharges (19% vs 67%) and status epilepticus (0% vs 44%) were indicative of poor outcome. NSE, S-100b, and pulsatile index of TCD were lower (P,0.01) in the good outcome group (Table 7). Discussion. The quantitative parameters of EEG, BSR, and WSE differed significantly between the groups with good and poor