Authors :
Presenting Author: Jose Ramirez Guerrero, BS – University of Alabama at Birmingham
Helen Brinyark, BS – University of Alabama at Birmingham
Caila Coyne, MEng – University of Alabama at Birmingham
Mahtab Rostamihosseinkhani, MD – University of Alabama at Birmingham
Hayder Al Hashim, MB – University of Alabama at Birmingham
Arie Nakhmani, PhD – University of Alabama at Birmingham
Benjamin Cox, MD – University of Alabama at Birmingham
Rachel Smith, PhD, MS, BS – University of Alabama at Birmingham
Rationale:
30% to 40% of patients with epilepsy experience drug-resistant epilepsy (DRE). DRE can be treated by surgical resection of the specific region in the brain where the seizure starts, or the seizure onset zone (SOZ). Identifying SOZs can be a costly and time-consuming process because of the lack of clinical validated biomarkers. Consequently, the surgical success rate is highly variable, and many patients continue to have seizures after surgical intervention. Single pulse electrical stimulation (SPES) is one of the techniques used to identify SOZs. SPES is administered to different regions of the brain to generate cortico-cortical evoked potentials (CCEPs) to aid the clinician to identify SOZs through intracranial EEG (iEEG) recordings. However, it is unknown whether the networks are significantly altered after stimulation which may have repercussions on how the SOZs are being localized clinically. In this study, we have systematically characterized how brain activity changes immediately before and after SPES to determine the degree to which SPES alters brain networks.
Methods: We collected one hour of iEEG data from 8 patients, both before and after the SPES procedure. We selected three 10-minute windows: anytime within the hour before SPES, and the first and last 10-minute windows of the hour long recording after SPES. Functional connectivity analysis of all three clips was computed based on the method described by Kramer et al. (2009), with permutation resampling for significance testing, to quantify connectivity before and after SPES. We obtained one connectivity matrix per clip, and the difference between each connectivity matrix before and after SPES was calculated. We then applied graph theory metrics to identify changes in the nodes and edges within the brain network, with a special focus on the SOZ nodes.
Results: At the individual level, different connectivity patterns were found acutely following SPES. In some patients, connectivity was stronger right after stimulation and in others, connectivity was weaker. In some patients, connectivity levels continued to increase or decrease compared to the baseline, whereas in others, the connectivity networks renormalized by the end of the hour. To test for significance of this effect, we calculated the median connectivity difference value for each patient and compared these values across the 3 different conditions. Interestingly, no significant median connectivity difference (p-value < 0.05) was found between the three conditions. Additionally, graph theory metrics like eigenvector centrality and shortest-path algorithms confirmed minimal changes in node influence.