519

Introduction: Electroencephalographic (EEG) monitoring is recommended during therapeutic hypothermia (TH), but availability, interpretation, and clinical utilization vary widely. We report simultaneous monitoring with simplified EEG using the bispectral index (BIS) and full montage continuous EEG (cEEG) in a cohort of cardiac arrest (CA) survivors during TH. Methods: 91 adults were monitored with cEEG and BIS (Covidien VISTA real-time fronto-temporal EEG display) during TH. Demographic and clinical data were prospectively collected. Both EEG systems were reviewed blinded to the other using standardized definitions with the cEEG as gold standard. Good outcome (GO) represents discharge cerebral performance category (CPC) 1-2. Results: Initial rhythm was VT/VF in 45%, asystole in 31%, and PEA in 16%. Time to ROSC was 19 (12-33) minutes, age was 60 (48-69) years, 74% were male. BIS monitoring was started 3.8 (range 1.5-9.5) hours and cEEG 14.5 (range 3-52) hours after ROSC. Seizures (SZ) occurred in 10 (11%) patients with onset 31 (range=15-56) hours post-ROSC. PEDS without progression to SZ occurred in 7 patients with onset 23 (range 11-44) hours post ROSC, and 74 patients (81%) had no SZ or PEDS. Seizures and PEDS were generalized in all patients. 80% of SZ patients were treated with benzodiazepines, fosphenytoin, levetiracetam, or valproic acid beginning 10 (3-20) hours after onset, and controlled in 25% of treated patients. GO occurred in 38% of the total cohort, 65% who met HACA criteria, but 0% of SZ and 14% of PEDS patients. Conclusions: Epileptiform activity occurred in 18.7% and SZ in 10% of TH patients after cardiac arrest, with onset delayed nearly 24 hours. All activity was generalized, and detectable with simplified frontotemporal EEG monitoring. Anti-epileptic medications were started late. Although SZ were controlled in 25% of those treated, no patient with SZ or PEDS made a good recovery.