The changes in the topological properties of brain structural network based on diffusion tensor imaging in pediatric...

This study aimed to analyze the topological characteristics of brain structural network in pediatric epilepsy patients with vagus nerve stimulation (VNS) by applying graph theoretical approaches.

Nine patients with generalized seizures and eight normal controls (NC) were enrolled. Based on diffusion tensor imaging, graph theory analysis was used to characterize the topological properties in preoperative patients (EP-pre), postoperative patients (EP-post) and NC. The global properties included clustering coefficient (Cp), shortest path length (Lp), small-worldness (γ, λ, δ), global network efficiency (Eg) and local network efficiency (Eloc). The regional properties included degree centrality (DC), nodal efficiency (NE), nodal local efficiency (NLE) and nodal shortest path length (Np). Two sample t-test and paired sample t-test were utilized to compare properties difference.

All three groups followed small-world characteristics. There was no significant difference in small-worldness, Cp, Lp, Eg or Eloc between EP-pre and EP-post. Compared with EP-pre: DC in EP-post decreased in the right cuneus and right temporal gyri, while increased in the right paracentral lobule; NE in EP-post decreased in the left dorsolateral superior frontal gyrus, right cuneus, right supramarginal gyrus, and right rolandic operculum, while increased in the right paracentral lobule; NLE in EP-post decreased in the left posterior cingulate gyrus and right supramarginal gyrus, while increased in the left parahippocampal gyrus; NP in EP-post decreased in the right paracentral lobule, while increased in the right cuneus.

VNS causes topological characteristics changes in pediatric patients with generalized seizures through regulating regional properties in some brain structures.