Authors :
Presenting Author: Taruna Yadav, PhD – Yale University
Bogdan Litvinov, MD – Yale University; George Culler, MD – Dartmouth-Hitchcock Medical Center; Avisha Kumar, NA – Yale University; Yinchen Song, PhD – Dartmouth-Hitchcock Medical Center; Benjamin Brinkmann, PhD – Mayo Clinic; Boney Joseph, MBBS – Mayo Clinic; Zan Ahmad, NA – Yale University; Nisali Gunawardane, MD – Yale University; Courtney Yotter, NA – Yale University; Imran Quraishi, MD – Yale University; START Clinical Trial Research Consortium, NA – Yale University, Dartmouth, Mayo Clinic; Barbara Jobst, MD PhD – Dartmouth-Hitchcock Medical Center; Gregory Worrell, MD PhD – Mayo Clinic; Hal Blumenfeld, MD PhD – Yale University
Rationale: Transient loss of consciousness in seizures can severely affect quality of life and mortality in epilepsy patients. A thorough analysis of brain activity patterns in seizures with and without behavioral impairment can provide useful insights into selecting effective neurostimulation parameters for patients. Therefore, the goal of this work is to study the electrophysiological characteristics of behaviorally impaired and spared mesial temporal lobe seizures using intracranial EEG (icEEG).
Methods:
A cohort of 24 patients with mesial temporal lobe epilepsy undergoing intracranial EEG at three sitesv (Yale, Mayo Clinic, and Dartmouth) were studied. Video behavior during seizures was retrospectively reviewed to distinguish seizures with or without impaired consciousness, based on the ability to respond to questions or commands during seizures. EEG signals were aligned to seizure onset and offset to extract data from the ictal and postictal periods. Electrical noise and high-frequency artifacts were visually identified and removed. Fast Fourier Transform was applied to compute EEG power in different frequency bands and averaged across electrodes in anatomical regions of interest within and outside the temporal lobe. Percentage changes in ictal and postictal EEG power relative to preictal baseline (30 sec) were compared for impaired vs. spared seizures.
Results:
We found that impaired seizures were significantly longer than spared seizures (Mean ± SEM: 103.2 ± 7.56 s vs. 57.96 ± 3.95 s, p < 0.05, Wilcoxon rank sum test). During the ictal period, average EEG power for impaired seizures was significantly greater than that of spared seizures across all frequency bands (Delta 0.5-4 Hz, Theta 4-8 Hz, Alpha 8-13 Hz, Beta 13-25 Hz, and Gamma 25-50 Hz, p < 0.05, Wilcoxon rank sum test) for both ipsilateral and contralateral mesial temporal electrodes. Time-course analysis shows that increases in EEG power were observed immediately after seizure onset on the ipsilateral side compared to 30sec after onset on the contralateral side. In the postictal period, impaired seizures had higher EEG power than spared seizures mainly in the delta frequency band.