Discriminating Pathological from Physiological High Frequency Oscillations in Non-invasive Recordings: A Prospective Multicenter Study on Children with Drug Resistant Epilepsy and Healthy Controls
Abstract number :
3.041
Submission category :
1. Basic Mechanisms / 1C. Electrophysiology/High frequency oscillations
Year :
2024
Submission ID :
101
Source :
www.aesnet.org
Presentation date :
12/9/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Lorenzo Fabbri, BS – CookChildren's Health care System
Margherita Matarrese, PhD – Universita' Campus Bio-Medico di Roma
Steven Stufflebeam, MD – Athinoula A. Martinos Center for Biomedical Imaging
Phillip Pearl, MD – Boston Children’s Hospital
M. Scott Perry, MD – Jane and John Justin Institute for Mind Health, Neurosciences Center, Cook Children's Medical Center
Eleonora Tamilia, PhD – Boston Children’s Hospital, Harvard Medical School, Boston, MA, USA
Christos Papadelis, PhD – Cook Children's Health Care System
Samantha Laboy, MS – CookChildren's Health care System
Cynthia Keator, MD – CookChildren's Health care System
Linh Tran, MD – CookChildren's Health care System
Saleem Malik, MD – CookChildren's Health care System
Dave Shahani, MD – CookChildren's Health care System
Rationale: High-frequency oscillations (HFOs) are interictal estimators of the epileptogenic zone (EZ) in drug resistant epilepsy (DRE). However, their clinical significance is uncertain because the area generating HFOs often extends beyond the actual EZ. This is due to the presence of physiological HFOs generated by normal (non-epileptogenic) brain regions. Thus, discriminating physiological from pathological HFOs is critical to their use as epilepsy biomarkers. Prior studies attempted this discrimination using intracranial EEG (iEEG) but were restricted in their ability to cover the entire head and record data from healthy individuals. In this prospective multi-center study, we aim to discriminate physiological and pathological HFOs non-invasively with whole-head techniques [such as magnetoencephalography, (MEG), and high-density EEG, (HD-EEG)], by comparing data from children with DRE with a control cohort of typically developing (TD) children.
Methods: We analyzed MEG (306 sensors) and HD-EEG (256 channels) data from 47 healthy controls (11.6 ± 3.5 y; 22 males) and 54 children with focal or generalized/diffuse DRE (12.9 ± 3.5 y; 21 males). In focal DRE patients, the epileptogenic regions (ERs) were defined at the lobar level based on their surgical workup findings; while non-ERs were defined as those in the contralateral hemisphere. HFOs (ripples: 80-160 Hz) were automatically detected (followed by visual review) on MEG and HD-EEG, and their cortical generators were localized through source imaging. HFOs were grouped in four classes based on whether they were generated by the TD brain (HFOs-TD), the ER (HFOs-ER), the non-ER (HFOs-nonER) or the generalized/diffuse DRE brain (HFOs-Gen). To assess differences between HFO classes, we extracted a set of temporal, morphological, spatial, and spectral features (Fig 1). These features were compared between groups with Kruskal-Wallis test (Tukey-Kramer for multiple comparisons).
Results: Rates of HFOs were higher on HD-EEG than MEG for both TD (0.56 vs 0.08 HFOs/min, p< 0.001) and DRE (0.87 vs 0.16 HFOs/min, p< 0.001). In children with DRE, HFO rate on HD-EEG was negatively correlated with age (p=0.039, R=-0.28). For HD-EEG, several HFO features differed between HFO classes (
Basic Mechanisms