Abstracts

Rationale: Postictal generalized EEG suppression (PGES) is commonly observed in patients after generalized convulsive seizures (GCS). It may be a causative factor in sudden unexpected death in epilepsy (SUDEP). The objective of this study is to investigate the incidence, risk factors, and clinical correlates of PGES, in so doing, to further understand its significance in the pathogenesis of SUDEP. Methods: We retrospectively reviewed the video-EEG of 109 consecutive patients with151 GCS during long term video-EEG monitoring. We determined the incidence, duration, and clinical correlates of PGES. We also investigated whether factors such as age, sex, seizure type (primary vs. secondarily GCS), total seizure duration, duration of tonic and clonic phases, and seizure termination (abrupt vs. gradual) influenced PGES. Results: PGES was observed in 64 of 109 (58.7%) patients and in 98 of 151 (64.9%) GCS. Average duration of PGES was 42.4 ± 19.1s. Statistical analysis showed that patients with PGES had no significant difference in age, gender, total seizure duration, total convulsive duration, clonic phase, seizure type, and seizure termination, as compared to those without PGES. However, PGES was observed in 97 of 123 (78.9%) GCS with tonic phase, and was observed only in 1 of 28 (3.6%) GCS without tonic phase. In addition, tonic phase was significantly prolonged in patients with (23.4 s) PGES than in those without (12.8 s) PGES (P=0.00086). Furthermore, regression analysis showed that a seizure's tonic phase proved to be an independent predictor of PGES. A one second increase in tonic phase duration was associated with a 0.06 increase in log odds of PGES (odds ratio =1.1, P=0.00055). Clinically, 97.3% patients were comatose during PGES, whereas only 26.7 % patients without PGES were comatose immediately after seizure termination. Postictal tachycardia and tachypnea associated with increased respiratory effort and coarse breathing sounds were commonly observed in patients with GCS, and such seizure related respiratory compromise appeared more prominent in patients with PGES than in those without PGES. Conclusions: PGES is a common EEG feature in patients with GCS. The presence and duration of a tonic phase is an independent predictor of PGES. Postictal coma, loss of brain stem auto-resuscitative reflexes, and respiratory compromise accompany patients with PGES. Given that 7 of 8 patients in video-EEG recorded SUDEP cases died in prone position, it is likely that the convergence of postictal coma, respiratory compromise and peripheral hypoventilation (i.e. prone position) play an important role in patients who suffer SUDEP.