Abstract:

Objective: To analyze stereo-electroencephalography (SEEG) data from patients with drug-resistant epilepsy (DRE) and explore the differences in graph theory indices of brain network between the epileptogenic zone (EZ) and the non-ictal zone (NIZ). Methods: Reviewed data from 11 patients who underwent SEEG implantation at Beijing Tiantan Hospital, Capital Medical University from August 2022 to December 2023.Based on the SEEG and structural imaging data, we calculated the epileptogenicity index and constructed epileptogenic map to differentiate the EZ from the NIZ.We then used Granger causality analysis to calculate functional adjacency matrices for both regions during interictal and epileptic periods, combining graph theory indices such as global efficiencies (Eg_lob), local efficiencies (E_loc), clustering coefficients (Cp), characteristic path length (Lp), normalized clustering coefficients (γ), normalized characteristic path length (λ), and small-world parameter (σ).We analyzed changes in the graph theory indices of patients in interictal and epileptic periods. Results: Compared with the interictal period, both γ (t=-3.730, P=0.005) and λ (t=-6.436, P=0.001) decreased in the NIZ during the epileptic period, while the differences of γ and λ in the EZ during the epileptic period were not statistically significant (P>0.05, for all).In terms of network efficiency, compared with the interictal period, Eg_lob (t=5.360, P=0.002; t=5.577, P=0.001) and E_loc (t=4.450, P=0.003; t=4.038, P=0.005) in both the EZ and NIZ increased during the epileptic period, while Lp decreased (t=-5.127, P=0.002; t=-3.912, P=0.005). Conclusions: During the epileptic period, both the EZ and NIZ exhibit increased synchronization across the whole brain network.Changes in graph theory indices, particularly the γ and λ may serve as the potential biomarkers for distinguishing the EZ and NIZ in epilepsy patients.

Key words: Drug resistant epilepsy, Stereotaxic techniques, Electroencephalography, Graph theory (not in MeSH)

摘要：

目的: 分析耐药性癫痫患者立体定向脑电图（SEEG）数据，探讨致痫区与非受累区脑网络图论指标差异。方法: 回顾分析2022年8月至2023年12月在首都医科大学附属北京天坛医院行立体定向电极植入术的11例耐药性颞叶内侧型癫痫患者的SEEG和结构影像数据，计算致痫指数并绘制致痫图，以此划分致痫区和非受累区；采用Granger因果分析分别计算两区域发作间期和发作期功能连接矩阵，结合全局效率（Eg_lob）、局部效率（E_loc）、特征路径长度（Lp）、聚类系数（Cp）、标准化特征路径长度（λ）、标准化聚类系数（γ）和小世界标量（σ）图论指标，对比分析发作间期与发作期图论指标变化。结果: 与发作间期相比，非受累区发作期γ（t=-3.730，P=0.005）和λ（t=-6.436，P=0.001）降低；而致痫区发作间期与发作期γ和λ差异无统计学意义（均P>0.05）。网络效率方面，与发作间期相比，致痫区和非受累区发作期Eg_lob（t=5.360，P=0.002；t=5.577，P=0.001）和E_loc（t=4.450，P=0.003；t=4.038，P=0.005）均增高，Lp均降低（t=-5.127，P=0.002；t=-3.912，P=0.005）。结论: 致痫区和非受累区在癫痫发作期存在全脑网络同步性增强，而图论指标的变化，特别是γ和λ可以作为区分癫痫患者致痫区与非受累区的潜在生物学标志物。

关键词: 耐药性癫痫, 立体定位技术, 脑电描记术, 图论(非MeSH词)