Objective:

We present a case of using automated algorithms to analyze central-positive complexes (CPCs) and postictal generalized electroencephalographic suppression (PGES) in the context of ECT-induced seizures. Our goal is to characterize peri-ictal EEG changes in ECT-induced seizures to define their unique features as generalized seizures.

Background:

Generalized seizures induced by electroconvulsive therapy (ECT) have been proposed as a model to study epileptic seizures in humans. We recently described generalized, high-amplitude waveforms with maximum positive voltage over the vertex (CPCs) during ECT-induced seizures. Furthermore, we developed an automated algorithm for detecting postictal PGES, defined as electroencephalographic (EEG) activity of ≤10 mV following a generalized seizure. 

Design/Methods:

This investigation is a secondary analysis of data collected from Reconstructing Consciousness and Cognition Phase 2 study (ClinicalTrials.gov NCT02761330). A 53-year-old female with a diagnosis of major depressive disorder underwent 5 sessions of right unilateral ECT during concurrent 64-channel EEG recordings. Quantitative EEG analysis allowed for assessment of spatiotemporal properties and spectral content during the peri-ictal period. CPC and PGES durations, quantified by automated algorithms, were compared to those derived through visual interpretation by expert epileptologists.

Results:

CPCs occurred during Phase III of all recorded seizures, with a median duration of 41 seconds. CPC evolved from 4.1 to 3.2 Hz as the seizures progressed. Peak-amplitude CPC scalp topology was consistent across patient’s seizures, showing maximal positive polarity over the vertex and surrounding regions and maximal negative polarity over the subocular electrodes. PGES duration lasted 5-45 seconds (median of 22 seconds) following the seizure and the median time to responsiveness after the seizure was 18.3 minutes. The findings using the automated algorithm were comparable to manual visual interpretation.

Conclusions:

This case presents the utility of automated algorithms to analyze CPCs and PGES characteristics, many of which were comparable to those evaluated through visual EEG interpretation.