Childhood absence epilepsy and distinct dynamic functional network connectivity patterns in self-limited epilepsy with centrotemporal spikes: a resting-state fMRI study.

Background: Alterations in dynamic brain functional connectivity (dFC) have been observed in epilepsy, few studies have directly compared the dynamic functional network connectivity (dFNC) patterns between patients with self-limited epilepsy with centrotemporal spikes (SeLECTS) and those with childhood absence epilepsy (CAE). This study aimed to explore differences in dFNC between these two epilepsy types and investigate how these patterns relate to clinical features.

Methods: Resting-state functional MRI data were collected from 34 SeLECTS patients, 22 CAE patients, and 32 healthy controls. Independent component analysis (ICA) was combined with a sliding-window technique to examine characteristics of dynamic FNC, including state transitions, connectivity strength, and temporal properties.

Results: Three recurring dFNC states were identified. SeLECTS patients spent significantly more time in a highly flexible state characterized by strong network integration, whereas CAE patients more frequently occupied a state marked by weak inter-network connectivity. Furthermore, SeLECTS patients showed greater variability in dFNC states over time. Certain clinical factors-particularly seizure frequency-were found to correlate with specific dFNC states, most notably in the SeLECTS group.

Conclusions: The study highlights distinct dynamic connectivity patterns between SeLECTS and CAE patients, suggesting that these two epilepsy types involve different network-level mechanisms. These findings contribute to a deeper understanding of epilepsy subtypes and may inform future diagnostic and treatment strategies.

Impact: This study identifies distinct dFNC patterns in two common childhood epilepsies: SeLECTS and CAE. It demonstrates the value of dynamic resting-state brain network analysis in pediatric epilepsy. These findings provide new neuroimaging biomarkers for early classification of epilepsy subtypes. Results may contribute to the development of personalized diagnosis and treatment strategies in children with epilepsy.