Single electroencephalographic patterns as specific and time-dependent indicators of good and poor outcome after cardiac arrest

Objective: To evaluate the prognostic value of single EEG patterns recorded at various time-frames in postanoxic comatose patients. Methods: This retrospective study included 30-min EEGs, classified according to the definitions of continuity of background activity given by the American Clinical Neurophysiology Society. Isoelectric pattern was distinguished from other suppressed activities. Epileptiform patterns were considered separately. Outcome was dichotomised based on recovery of consciousness as good (Glasgow Outcome Scale [GOS] 3-5) or poor (GOS 1-2). Results: We analysed 211 EEGs, categorised according to time since cardiac arrest (within 12 h and around 24, 48 and 72 h). In each time-frame we observed at least one EEG pattern which was 100% specific to poor or good outcome: at 12 h continuous and nearly continuous patterns predicted good outcome and isoelectric pattern poor outcome; at 24 h isoelectric and burst-suppression predicted poor outcome; at 48 and 72 h isoelectric, burst-suppression and suppression (2-10 lV) patterns predicted poor outcome. Conclusions: The prognostic value of single EEG patterns, defined according to continuity and voltage of background activity, changes until 48-72 h after cardiac arrest and in each time-frame there is at least one pattern which accurately predicts good or poor outcome. Significance: Standard EEG can provide time-dependent reliable indicators of good and poor outcome throughout the first 48-72 h after cardiac arrest. (C) 2016 Published by Elsevier Ireland Ltd on behalf of International Federation of Clinical Neurophy-siology.