Generalized epileptiform discharges in postanoxic encephalopathy: Quantitative characterization in relation to outcome

Abstract

Objective Electrographic status epilepticus is observed in 10-35% of patients with postanoxic encephalopathy. It remains unclear which electrographic seizure patterns indicate possible recovery, and which are a mere reflection of severe ischemic encephalopathy, where treatment would be futile. We aimed to identify quantitative electroencephalography (EEG) features with prognostic significance. Methods From continuous EEG recordings of 47 patients with generalized electrographic status epilepticus after cardiac arrest, 5-min epochs were selected every hour. Epochs were visually assessed and categorized into seven categories, including epileptiform discharges. Five quantitative measures were extracted, reflecting background continuity, discharge frequency, discharge periodicity, relative discharge power, and interdischarge waveform correlation. The best achieved outcome within 6 months after cardiac arrest was categorized as "good" (Cerebral Performance Category 1-2, i.e., no or moderate neurologic disability) or "poor" (CPC 3-5, i.e., severe disability, coma, or death). Results Ten patients (22%) had a good outcome. Status epilepticus in patients with good outcome started later (45 vs. 29 h after cardiac arrest, p < 0.001), more often ceased for at least 12 h (90% vs. 16%, p = 0.02), and was less often treated with antiepileptic drugs (30% vs. 73%, p = 0.02). Status epilepticus in patients with a good outcome always evolved from a continuous background pattern, as opposed to evolution from a discontinuous background pattern in 14 patients (38%) with a poor outcome. Epileptiform patterns of patients with good outcome had higher background continuity (1.00 vs. 0.83, p < 0.001), higher discharge frequency (1.63 vs. 0.90 Hz, p = 0.002), lower relative discharge power (0.29 vs. 0.40, p = 0.01), and lower discharge periodicity (0.32 vs. 0.45, p = 0.04). Significance Our results can be used to identify patients with possible recovery. We speculate that quantitative features associated with poor outcome reflect low neural network complexity, resulting from extensive ischemic damage.