Page 212 - MEGIN Book Of Abstracts - 2023

Psychology, School of Biological Sciences, Faculty of Biol- Keywords: Connectome, EEG, Graph theory, IGE, Network
ogy, Medicine and Health, University of Manchester, United analysis, fMRI
Kingdom; Manchester Academic Health Sciences Centre,
United Kingdom; Department of Molecular and Clinical Phar- Epilepsy & behavior: E&B (2020), Vol. 106 (32193094) (20
macology, Institute of Translational Medicine, University of citations)
Liverpool, United Kingdom; The Walton Centre NHS Founda-
tion Trust, Liverpool, United Kingdom
Magnetoencephalography resting state
ABSTRACT The evaluation of the role of anomalous connectivity patterns as indicatives of surgical
neuronal networks in epilepsy using a graph theoretical outcome in epilepsy patients (2020)
approach is of growing research interest. There is cur-
rently no consensus on optimal methods for perform- Aydin, Ümit; Pellegrino, Giovanni; Ali, Obaï Bin Ka'b;
ing network analysis, and it is possible that variations in Abdallah, Chifaou; Dubeau, François; Lina, Jean-Marc;
study methodology account for diverging findings. This Kobayashi, Eliane; Grova, Christophe
review focuses on global functional and structural in-
terictal network characteristics in people with idiopath- Multimodal Functional Imaging Lab, Department of Physics
ic generalized epilepsy (IGE) with the aim of appraising and PERFORM Centre, Concordia University, Montreal, Qué-
the methodological approaches used and assessing for bec, Canada. Authors to whom any correspondence should
meaningful consensus. Thirteen studies were included be addressed
in the review. Data were heterogenous and not suitable
for meta-analysis. Overall, there is a suggestion that the OBJECTIVE Focal epilepsy is a disorder affecting sev-
cerebral neuronal networks of people with IGE have eral brain networks; however, epilepsy surgery usually
different global structural and functional characteris- targets a restricted region, the so-called epileptic focus.
tics to people without epilepsy. However, the nature There is a growing interest in embedding resting state
of the aberrations is inconsistent with some studies (RS) connectivity analysis into pre-surgical workup.
demonstrating a more regular network configuration
in IGE, and some, a more random topology. There is APPROACH In this retrospective study, we analyzed
greater consistency when different data modalities and Magnetoencephalography (MEG) long-range RS
connectivity subtypes are compared separately, with functional connectivity patterns in patients with drug-
a tendency towards increased small-worldness of net- resistant focal epilepsy. MEG recorded prior to surgery
works in functional electroencephalography/magne- from seven seizure-free (Engel Ia) and five non seizure-
toencephalography (EEG/MEG) studies and decreased free (Engel III or IV) patients were analyzed (minimum
small-worldness of networks in structural studies. 2-years post-surgical follow-up). MEG segments
Prominent variation in study design at all stages is likely without any detectable epileptic activity were source
to have contributed to differences in study outcomes. localized using wavelet-based Maximum Entropy on
Despite increasing literature surrounding neuronal net- the Mean method. Amplitude envelope correlation in
work analysis, systematic methodological studies are the theta (4-8 Hz), alpha (8-13 Hz), and beta (13-26 Hz)
lacking. Absence of consensus in this area significantly bands were used for assessing connectivity.
limits comparison of results from different studies, and
the ability to draw firm conclusions about network MAIN RESULTS For seizure-free patients, we found an
characteristics in IGE. isolated epileptic network characterized by weaker
connections between the brain region where interictal
epileptic discharges (IED) are generated and the rest of
the cortex, when compared to connectivity between
the corresponding contralateral homologous region
and the rest of the cortex. Contrarily, non seizure-free

 ontents  Index 191
C