Noninvasive Mapping of Spike Onset Identifies the Epileptogenic Zone in Children with Refractory Epilepsy
Abstract number :
1.326
Submission category :
9. Surgery / 9B. Pediatrics
Year :
2023
Submission ID :
243
Source :
www.aesnet.org
Presentation date :
12/2/2023 12:00:00 AM
Published date :
Authors :
Presenting Author: Margherita A.G. Matarrese, M.Sc. – Campus Bio-Medico University of Rome
Eleonora Tamilia, Ph.D. – Boston Children’s Hospital; M Scott Perry, MD – Cook Children's Health Care System; Joseph R. Madsen, MD – Boston Children’s Hospital; Steven M. Stufflebeam, MD – Massachusetts General Hospital; Jeffrey Bolton, MD – Boston Children’s Hospital; Scellig S.D. Stone, MD – Boston Children’s Hospital; Phillip L. Pearl, MD – Boston Children’s Hospital; Simonetta Filippi, Prof. – Università Campus Bio-Medico di Roma; Christos Papadelis, PhD – Cook Children's Health Care System
Rationale:
For many children with drug resistant epilepsy (DRE) of focal onset, resective neurosurgery is the most efficacious treatment, but requires a biomarker that identifies the epileptogenic zone (EZ) with high precision. Our group has shown that spike-onset, identified via intracranial electroencephalography (iEEG), is a powerful biomarker but its delineation implies complications due to the intrinsic invasiveness. Hence, the availability of an interictal biomarker that delineates the EZ non-invasively is paramount. Here, we aim to map the propagation of spikes non-invasively via high-density EEG (HD-EEG) and magnetoencephalography (MEG), assess the concordance of the non-invasively localized spike-onset with the one derived from iEEG, and examine its ability to identify the EZ.
Methods:
We retrospectively analyzed interictal iEEG, HD-EEG and MEG data from 10 children with DRE who underwent resective neurosurgery with favorable outcome (Engel 1 at one year after surgery). Using electromagnetic source imaging, we mapped the spatiotemporal propagation of spikes in the source domain with wavelet Maximum Entropy of the Mean (wMEM) (Figure 1), and identified the anatomical gyrus where the spike-onset was located. We computed the duration, spatial displacement, and speed of each spike propagation. For each patient and modality, we defined the region that predominantly was the first to be active as spike-onset. We then estimated the mean distance of these noninvasive spike-onsets from resection and from the spike-onset defined invasively that served as gold standard.
Results:
We found a higher spike propagation rate in iEEG than HD-EEG (13.5 [3.8-29.4] vs. 4.3 [2.6-9.6] events/min; p=0.048). No difference in spike propagation rate was observed between MEG (6.2 [2.4-14.4] events/min) and both HD-EEG and iEEG (p>0.5). Also, no difference was observed across modalities in duration (iEEG: 74 ms [55-143 ms]; HD-EEG: 71 ms [55-125 ms]; MEG: 70 ms [64-108 ms]; p>0.5; Fig. 2A) and spatial displacement (iEEG: 5 cm [3-8 cm]; HD-EEG: 8 cm [6-13 cm]; MEG: 6 cm [3-8 cm]; p>0.5; Fig. 2B). The lowest propagation speed was observed with iEEG (~0.5 m/s), followed by MEG (~0.7 m/s) and HD-EEG (~1 m/s) (Fig. 2C). The spike-onset defined via MEG and HD-EEG was concordant with the iEEG-defined spike-onset in 70% and 40% of patients, respectively (Fig. 2D). The distance between the non-invasively localized spike-onset from the iEEG spike-onset were 3.2 cm [0.8-5.0 cm] for the HD-EEG and 1.1 cm [0.8-2.4 cm] for the MEG (Fig. 2E). Finally, the distance from resection was 0.9 cm (median) for the iEEG spike-onset and ~1.6 cm (median) for the spike-onsets defined via HD-EEG and MEG (Fig. 2F; p=0.15).
Conclusions:
Our data provide evidence that spike propagation, and its onset, can be mapped noninvasively with electromagnetic source imaging in children with DRE. The non-invasive mapping of such an interictal biomarker may augment the presurgical evaluation of children with DRE by reducing the need for invasive monitoring.
Funding: RO1NS104116-01A1 and R21NS101373-01A1 by NINDS.
Surgery