Dynamic functional connectivity and graph theory metrics in a rat model of temporal lobe epilepsy reveal a preference for brain states with a lower functional connectivity, segregation and integration

Emma Christiaen,Marie-Gabrielle Goossens,Benedicte Descamps,Lars E Larsen,Paul Boon,Robrecht Raedt,Christian Vanhove

DOI: https://doi.org/10.1016/j.nbd.2020.104808

Abstract:Epilepsy is a neurological disorder characterized by recurrent epileptic seizures. The involvement of abnormal functional brain networks in the development of epilepsy and its comorbidities has been demonstrated by electrophysiological and neuroimaging studies in patients with epilepsy. This longitudinal study investigated changes in dynamic functional connectivity (dFC) and network topology during the development of epilepsy using the intraperitoneal kainic acid (IPKA) rat model of temporal lobe epilepsy (TLE). Resting state functional magnetic resonance images (rsfMRI) of 20 IPKA animals and 7 healthy control animals were acquired before and 1, 3, 6, 10 and 16 weeks after status epilepticus (SE) under medetomidine anaesthesia using a 7 T MRI system. Starting from 17 weeks post-SE, hippocampal EEG was recorded to determine the mean daily seizure frequency of each animal. Dynamic FC was assessed by calculating the correlation matrices between fMRI time series of predefined regions of interest within a sliding window of 50 s using a step length of 2 s. The matrices were classified into 6 FC states, each characterized by a correlation matrix, using k-means clustering. In addition, several time-variable graph theoretical network metrics were calculated from the time-varying correlation matrices and classified into 6 states of functional network topology, each characterized by a combination of network metrics. Our results showed that FC states with a lower mean functional connectivity, lower segregation and integration occurred more often in IPKA animals compared to control animals. Functional connectivity also became less variable during epileptogenesis. In addition, average daily seizure frequency was positively correlated with percentage dwell time (i.e. how often a state occurs) in states with high mean functional connectivity, high segregation and integration, and with the number of transitions between states, while negatively correlated with percentage dwell time in states with a low mean functional connectivity, low segregation and low integration. This indicates that animals that dwell in states of higher functional connectivity, higher segregation and higher integration, and that switch more often between states, have more seizures.