Localization of the Epileptogenic Zone Using High-frequency Oscillation and Phase-amplitude Coupling Measured Under Isoflurane Anesthesia
Abstract number :
1.285
Submission category :
3. Neurophysiology / 3G. Computational Analysis & Modeling of EEG
Year :
2024
Submission ID :
1285
Source :
www.aesnet.org
Presentation date :
12/7/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Tomoko Maehara, MD Candidate – Kagoshima University
Hiroshi Uda, MD/PhD – Wayne State University
Naoto Kuroda, MD – Wayne State University
Ethan Firestone, MD-PhD Candidate – Wayne State University
Keisuke Hatano, MD/PhD – Wayne State University
Mio Maeda, MD Candidate – Kagoshima University
Dominik Choromanski, MD – Children’s Hospital of Michigan
Michael Cools, MD – Children's Hospital of Michigan
Aimee F. Luat, MD – Central Michigan University
Ryosuke Hanaya, MD/PhD – Kagoshima University
Eishi Asano, MD/PhD – Wayne State University
Rationale: Complete removal of the epileptogenic zone is crucial for achieving seizure freedom in patients with drug-resistant focal epilepsy. When a patient achieves long-term seizure freedom, it is generally considered that the epileptogenic zone has been sufficiently removed. In clinical practice, intraoperative intracranial EEG (iEEG) is utilized to delineate surgical boundaries. iEEG biomarkers include the occurrence rate of high-frequency oscillation (HFO) and the modulation index (MI), which quantifies the strength of phase-amplitude coupling between HFOs and delta waves. We aimed to assess the diagnostic value of these biomarkers measured under Isoflurane anesthesia, specifically determining their accuracy in identifying the extraoperative iEEG-defined seizure onset zone (SOZ), the resection of which resulted in seizure freedom.
Methods: We studied 62 children (aged 1-19 years; 31 girls) with drug-resistant focal epilepsy who achieved an ILAE class 1 outcome following extraoperative iEEG recording. Using 5-minute iEEG recordings under Isoflurane anesthesia, open-source detectors computed HFO rates at ≥80 Hz as well as MI, reflecting the strength of coupling between HFOs at ≥80 Hz and slow waves at 3-4 Hz. We employed logistic regression analysis per patient to determine whether HFO rates or MI could classify the SOZ from non-epileptic sites, defined as areas outside the resected regions. We evaluated the classification accuracy of each per-patient logistic regression model using the area under the curve (AUC) of the receiver operating characteristic analysis.
Results: We analyzed 558 SOZ artifact-free electrode sites (frontal lobe: 19%; temporal lobe: 48%; parietal lobe: 26%; occipital lobe: 7%) and 3828 non-epileptic artifact-free electrode sites (frontal lobe: 39%; temporal lobe: 26%; parietal lobe: 25%; occipital lobe: 10%). The iEEG biomarker values were consistently higher in the SOZ than in the non-epileptic sites across four brain lobes (Figure 1). The logistic regression model incorporating HFO rates classified the SOZ with a mean AUC of 0.78 (95% CI: 0.74 to 0.82), whereas that incorporating MI had a mean AUC of 0.73 (95% CI: 0.69 to 0.77).
Conclusions: HFO and MI measured under Isoflurane anesthesia may provide valuable information for localizing the epileptogenic zone.
Funding: JSPS JP22J23281 and JP22KJ0323 (N.K.), NIH NS064033 (E.A.).
Neurophysiology