Functional network topology in drug resistant and well-controlled idiopathic generalized epilepsy: a resting state functional MRI study



Pegg, Emily J, McKavanagh, Andrea, Bracewell, R Martyn, Chen, Yachin ORCID: 0000-0002-6623-8658, Das, Kumar, Denby, Christine, Kreilkamp, Barbara AK ORCID: 0000-0001-6881-5191, Laiou, Petroula, Marson, Anthony ORCID: 0000-0002-6861-8806, Mohanraj, Rajiv
et al (show 2 more authors) (2021) Functional network topology in drug resistant and well-controlled idiopathic generalized epilepsy: a resting state functional MRI study. BRAIN COMMUNICATIONS, 3 (3). fcab196-.

[img] Text
Pegg-2021-Brain-communications.pdf - Published version

Download (750kB) | Preview

Abstract

Despite an increasing number of drug treatment options for people with idiopathic generalized epilepsy (IGE), drug resistance remains a significant issue and the mechanisms underlying it remain poorly understood. Previous studies have largely focused on potential cellular or genetic explanations for drug resistance. However, epilepsy is understood to be a network disorder and there is a growing body of literature suggesting altered topology of large-scale resting networks in people with epilepsy compared with controls. We hypothesize that network alterations may also play a role in seizure control. The aim of this study was to compare resting state functional network structure between well-controlled IGE (WC-IGE), drug resistant IGE (DR-IGE) and healthy controls. Thirty-three participants with IGE (10 with WC-IGE and 23 with DR-IGE) and 34 controls were included. Resting state functional MRI networks were constructed using the Functional Connectivity Toolbox (CONN). Global graph theoretic network measures of average node strength (an equivalent measure to mean degree in a network that is fully connected), node strength distribution variance, characteristic path length, average clustering coefficient, small-world index and average betweenness centrality were computed. Graphs were constructed separately for positively weighted connections and for absolute values. Individual nodal values of strength and betweenness centrality were also measured and 'hub nodes' were compared between groups. Outcome measures were assessed across the three groups and between both groups with IGE and controls. The IGE group as a whole had a higher average node strength, characteristic path length and average betweenness centrality. There were no clear differences between groups according to seizure control. Outcome metrics were sensitive to whether negatively correlated connections were included in network construction. There were no clear differences in the location of 'hub nodes' between groups. The results suggest that, irrespective of seizure control, IGE interictal network topology is more regular and has a higher global connectivity compared to controls, with no alteration in hub node locations. These alterations may produce a resting state network that is more vulnerable to transitioning to the seizure state. It is possible that the lack of apparent influence of seizure control on network topology is limited by challenges in classifying drug response. It is also demonstrated that network topological features are influenced by the sign of connectivity weights and therefore future methodological work is warranted to account for anticorrelations in graph theoretic studies.

Item Type: Article
Uncontrolled Keywords: connectivity, epilepsy, networks, drug-resistance, fMRI
Divisions: Faculty of Health and Life Sciences
Faculty of Health and Life Sciences > Institute of Systems, Molecular and Integrative Biology
Depositing User: Symplectic Admin
Date Deposited: 06 Sep 2021 10:48
Last Modified: 18 Jan 2023 21:29
DOI: 10.1093/braincomms/fcab196
Related URLs:
URI: https://livrepository.liverpool.ac.uk/id/eprint/3136083