Abstracts

Rationale: Prolonged postictal EEG suppression (PES) following generalized tonic-clonic seizures (GTCS) has been proposed as a quantifiable biomarker for Sudden Unexpected Death in Epilepsy (SUDEP) and a possible marker of postictal autonomic dysfunction. Decreased heart rate variability (HRV) in patients with drug-resistant epilepsy (DRE) may also serve as an autonomic biomarker for SUDEP. It remains unclear whether PES as quantified by scalp EEG can reliably predict postictal autonomic dysfunction and SUDEP. We compared PES quantified from intracranial EEG with simultaneous scalp EEG; additionally, we assessed the association of PES with changes in HRV and SUDEP outcome.

Methods: Postictal EEG suppression was defined by attenuation < 50 μV on intracranial and < 10 μV on scalp EEG. Three EEG experts reviewed the scalp EEGs together and agreed on characteristics of PES for each patient. Intracranial PES was quantified by an algorithm using the Hilbert transform. Root mean square of successive differences of interbeat intervals (RMSSD), normalized low frequency (LF, 0.04–0.15 Hz) and high frequency (HF, 0.15–0.4 Hz) power, and LF/HF ratio (LHR) were computed to quantify HRV. Generalized estimating equation models were used to assess periictal changes in HRV accounting for within-subject repeated measures, and multivariate regression analysis for correlations between PES, HRV, and SUDEP.

Results: 63 patients with DRE (31 female, mean [range] age, 35.5 [15-65] years) and 6 SUDEP patients (all male, mean [range] age, 33 [12-56] years) were included. PES was identified for 74/75 seizures on intracranial EEG and 26/30 seizures on simultaneous scalp EEG. The mean duration of intracranial PES was shorter than scalp PES (mean 23.31 vs 35.43 s, t=-2.36, p< 0.05). The intracranial PES was generalized ( >90% of contacts) in 48 seizures and regional (< 90% of contacts) in 27 seizures. The mean duration of regional PES was shorter than generalized PES on intracranial EEG (mean 2.90 vs 34.68 s, t=7.86, p< 0.0001). The mean decrease in postictal RSSMD compared to preictal value was 0.33 s (95% CI 0.27-0.39 s). The mean increase in LHR over the 40 minutes before and after seizure was 5.4 (95% CI 2.5-8.2). In multivariate regression analysis, SUDEP outcome was not predicted by duration (p=0.72) or the extent (generalized vs regional, p=0.239) of PES. Postictal RMSSD was lower in patients with generalized PES compared to regional PES (t=-2.16, p< 0.05).

Conclusions: Postictal EEG suppression, regional or generalized, is a common finding on intracranial recordings after GTCS. Although, periictal HRV changes can predict the occurrence of GTCS, they do not correlate with duration of PES. These findings do not support PES as an independent predictor of SUDEP in patients with DRE. Postictal HRV changes correlate with the extent of PES indicating further investigation is warranted to identify and intervene on brain areas involved in postictal cortical/subcortical dysfunction and autonomic dysregulation.

Funding: Please list any funding that was received in support of this abstract.: AES/EF RTFC (630111), IFCN Research Scholarship, and ACNS Fellowship Award (BE).