Authors :
Xiyue Sun, M.E – University of Electronic Science and Technology of China
Honghao Han, PhD – University of Electronic Science and Technology of China
Zhi Fang, M.E – University of Electronic Science and Technology of China
Qiangqiang Liu, MD – Ruijin Hospital, Shanghai Jiao Tong University School of Medicine
Li Feng, PhD – Xiangya Hospital Central South University
Presenting Author: Rong Li, PhD – University of Electronic Science and Technology of China
Rationale:
Interictal epileptiform activity has been demonstrated to propagate from epileptogenic focus to other brain regions via white matter connectivity. However, precise mechanisms underlying the propagation of abnormal functional activity through white matter pathways remain unclear. Although multiple models have been employed for investigation, existing researches overlook a fundamental premise: epileptiform activity propagation may not conform to a singular propagation model across whole brain. Therefore, we hypothesize that propagation of epileptiform activity from epileptogenic focus to global brain networks adheres to spatially heterogeneous communication models.
Methods:
First, normative models were used to obtain individual structural connectivity (SC) and amplitude of low-frequency fluctuation (ALFF) values from healthy controls. These data were subsequently applied to (a) compute communication efficiencies for two communication models (shortest-path and diffusion models) and (b) generate actual ALFF abnormality maps in 56 patients with mesial temporal lobe epilepsy (mTLE). The hippocampus, a predominant epileptogenic focus in mTLE, was selected as the abnormality source. Predicted maps of abnormal ALFF were generated by integrating source abnormalities with communication efficiencies from source to other brain regions. Then, the regional adherence to specific model was determined through comparing correlations between actual and predicted ALFF abnormalities under dual models. Furthermore, the node-wise traversal counts were quantified to identify epileptogenic hub regions in two models. Last, the propagation path from the hippocampus to the orbitofrontal cortex was identified, which served as a prototypical target for path analysis.
Results:
The propagation of epileptogenic activity, as modelled by dual communication pathways, is dominated by the shortest path model in terms of subcortical dissemination (95.7%), while the diffusion model drives prevalent cortico-cortical synchronisation (74.2%). Pathological propagation from hippocampus demonstrated significantly higher communication efficiency under shortest path model versus diffusion model. Shortest-path-based propagation exhibited the highest level of efficiency in the hippocampo-thalamic pathways, whereas diffusion-based propagation demonstrated the highest level efficiency from the hippocampus to the dorsal attention network. The analysis of nodal traversal counts demonstrated predominant intra-hippocampal propagation across both models and thalamic hub predominance in shortest-path routing versus visual network preferential involvement in diffusion patterns. In particular, average traversal counts positively correlated with ALFF deviations in epileptic nodes when propagating under diffusion model. The results of the pathway analysis indicated two distinct transmission routes from the rostral and caudal hippocampus to the orbitofrontal cortex.
Conclusions:
Collectively, this study reveals the region-specific propagation mechanisms of epileptogenic activity, which may offer mechanistic insights for personalised epilepsy interventions.
Funding:
National Natural Science Foundation of China (Nos. 82372085 and 62333003)