Abstracts

Transient Slow Oscillations Support Interictal Spike Networks in Drug-resistant Focal Epilepsy

Abstract number : 2.103
Submission category : 3. Neurophysiology / 3G. Computational Analysis & Modeling of EEG
Year : 2023
Submission ID : 509
Source : www.aesnet.org
Presentation date : 12/3/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Hongyi Ye, BEng – Epilepsy Center, Second Affiliated Hospital, School of Medicine, Zhejiang University

Lingqi Ye, MD – Epilepsy Center, Second Affiliated Hospital, School of Medicine, Zhejiang University; Lingli Hu, MS – Epilepsy Center, Second Affiliated Hospital, School of Medicine, Zhejiang University; Shuang Wang, MD, PhD – Epilepsy Center, Second Affiliated Hospital, School of Medicine, Zhejiang University; Cong Chen, MD – Epilepsy Center, Second Affiliated Hospital, School of Medicine, Zhejiang University

Rationale: Interictal epileptic spikes are a classic epilepsy biomarker that often co-occur across brain regions, forming spike networks. Despite the well-elucidated clinical relevance of spike networks, the factors prompting spike propagation remain poorly understood. We postulated that pre-spike slow oscillations (SOs) in delta and theta ranges might play a role in facilitating spike propagation.

Methods: Intracranial and scalp electroencephalography (EEG) were simultaneously recorded from 22 patients with drug-resistant focal epilepsy. Selected intracranial channels located in the grey matter were categorized as the irritative zone (IZ) or the normal zone (NOZ). Interictal spikes were classified as propagating or non-propagating spikes depending on whether they co-occurred in different brain regions. We investigated if propagating spikes exhibited enhanced SOs before them. The time window for pre-spike analysis was initially set from 1250 to 250 ms preceding the spike peak. The amplitude and phase synchronization of pre-spike SOs, computed via the Hilbert transform, were analyzed during non-rapid eye movement (NREM) sleep and wakefulness. We further explored dynamic changes of pre-spike SOs using a sliding window approach.

Results: Analysis of pre-spike SOs was performed on scalp EEG during NREM sleep and on intracranial EEG during wakefulness. Scalp EEG data revealed that propagating spikes had a higher amplitude of pre-spike SOs compared to non-propagating spikes (delta band: p = 0.02, theta band: p = 0.01). Within the IZ, the amplitude of pre-spike SOs was significantly higher in the propagating spike group than in the non-propagating spike group (p < 0.001 in delta and theta band), and this phenomenon began at approximately 1100 ms before spikes. In addition, similar results were also observed in the NOZ (delta band: p = 0.005, theta band: p = 0.07), with this phenomenon beginning at approximately 1050 ms before spikes. Surprisingly, the dynamic changes of pre-spike SOs differed between propagating and non-propagating spikes. Compared to the interictal baseline, SOs preceding propagating spikes showed a trend of increase in both amplitude and phase synchronization as they approached the spike onset, while those of SOs preceding non-propagating spikes remained almost unchanged.

Conclusions: The propagation of interictal spikes is facilitated by transient but widely distributed pre-spike SOs. Furthermore, pre-spike SOs may induce spike generation and spike propagation through different mechanisms.

Funding: This work was supported by the National Natural Science Foundation of China (grant numbers: 81971207, 81971208).

Neurophysiology