Introduction: In patients with recurrent seizures, the postictal window can be associated with a spectrum of behavioral and physiological alterations. Analyzing electrographic changes in the postictal window may help better characterize the epileptogenic brain tissue prior to epilepsy surgery. The purpose of this study was to describe postictal changes in functional connectivity networks among patients undergoing intracranial evaluation.
Methods: Intracranial EEG recordings from 30 patients were analyzed. For each seizure, 15 minutes of preictal and postictal EEG activity were examined. Functional connectivity networks were derived using a band-limited spectral coherence approach, and measures of network organization (e.g., clustering coefficient, characteristic path length, mean edge weight) were computed. For each seizure, shifts in network measures from the preictal to postictal window were expressed as normalized percent change. Surgical outcomes were classified as seizure-free or seizure-persistent at =2 post-operative years.
Results: In total, 84 seizures were analyzed. Postictal network alterations were maximal in the beta (13-30 Hz) and alpha (8-13 Hz) frequency bands. In the beta band, the average percent change in the global clustering coefficient across seizures was 31.7%. Similarly, the average percent change in edge weight was 25.9%. These overall trends appeared to be related to surgical outcome. Among patients with unfavorable (seizure-persistent) outcomes, the average peri-ictal change in clustering coefficient was 50.3% compared to 16.8% in the seizure-free group (p < 0.001), a pattern that was conserved across multiple network measures. In the seizure-persistent group, significant postictal changes often persisted for the entire 15-minute time window.
Conclusions: The postictal window was associated with a shift towards increased network connectivity and node clustering. The magnitude of peri-ictal shifts was correlated with the frequency band analyzed and the surgical outcome of the patient. We hypothesize that postictal deactivation of subcortical relay nuclei may contribute to the differences observed across surgical outcome groups.
Patient Care: Our research has the potential to inform predictive models of epilepsy surgery outcome. By identifying features of postictal network connectivity that tend to correlate with favorable vs. unfavorable epilepsy surgery outcomes, we may begin to implement this knowledge into prospective research aiming to predict which patients will benefit from surgery.
Learning Objectives: - Understand the concept of network connectivity and epileptogenic networks in the pathophysiology of epilepsy
- Appreciate the significance of postictal electrographic activity in characterizing epileptic networks