Electroencephalography based functional networks in newly diagnosed childhood epilepsies

Objective: It remains unclear to what extent brain networks are altered at an early stage of epilepsy, which may be important to improve our understanding on the course of network alterations and their association with recurrent seizures and cognitive deficits. Methods: 89 Drug-naïve children with newly diagnosed focal or generalized epilepsies and 179 controls were included. Brain networks were based on interictal electroencephalography recordings obtained at first consultation. Conventional network metrics and minimum spanning tree (MST) metrics were computed to characterize topological network differences, such integration and segregation and a hub measures (betweenness centrality). Results: Network alterations between groups were only identified by MST metrics and most pronounced in the delta band, in which a loss of network integration and a significant lower betweenness centrality was found in children with focal epilepsies compared to healthy controls (p <0.01). A reversed group difference was found in the upper alpha band. The network topology in generalized epilepsies was relatively spared. Conclusions: Interictal network alterations - only identifiable with the MST method - are already present at an early stage of focal epilepsy. Significance: We argue that these alterations are subtle at the early stage and aggravate later as a result of persisting seizures.