Abstracts

Increased Small-world Network Topology After Commencing Anti-seizure Medication

Abstract number : 2.217
Submission category : 5. Neuro Imaging / 5B. Functional Imaging
Year : 2023
Submission ID : 353
Source : www.aesnet.org
Presentation date : 12/3/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: David Vaughan, PhD – The Florey Institute of Neuroscience and Mental Health

Heath Pardoe, PhD – The Florey Institute of Neuroscience and Mental Health; Remika Mito, PhD – The Florey Institute of Neuroscience and Mental Health; Moksh Sethi, MBBS, FRACP – Eastern Health, Melbourne, Australia; David Vaughan, MBBS, FRACP, PhD – The Florey Institute of Neuroscience and Mental Health; Patrick Carney, MBBS, FRACP, PhD – Eastern Health, Melbourne, Australia; Graeme Jackson, MBBS, FRACP, MD – The Florey Institute of Neuroscience and Mental Health

Rationale:
Anti-seizure medications (ASM) target cellular processes, such as ion channel functions, to reduce cortical excitability. As a consequence, introducing ASMs is likely to change large-scale networks. The direct effects of ASM treatment on large-scale networks remain poorly understood as they require drug-naive epilepsy cohorts. In this study, we carried out a prospective study using advanced functional MRI (fMRI) data gathered before and six months after the start of ASM treatment.

Methods:
We studied 28 participants (15F; mean age 30.8±12.6 S.D, at first scan) after their first seizure. Participants were prescribed a variety of ASMs: Lamotrigine (n=14); Carbamazepine (n=6); Levetiracetam (n=3); Valproic Acid (n=2) and other medications (n=3). The clinical outcome was not included in the analysis. Two 15-minute long resting-state fMRI scans were completed for each participant (TR=3s and 3x3x3mm voxel size). The data were preprocessed using fmriprep1 and filtered between 0.01-0.1 Hz. The global signal, cerebrospinal fluid, white matter, and six movement parameters were regressed from the data. Parcellation was performed using the Glasser mask with 180 symmetric regions2 and Pearson r correlation was used to estimate functional connectivity between regions. Binary and proportionally thresholded networks between 20%-50% network density were used to calculate clustering coefficient and path length. A greater clustering coefficient (more local nodal connectedness) and shorter path length (greater network efficiency) suggest that a network has a small-world topology, with efficient local and global processing.3 The average of all network densities was used for statistical analysis using paired-sample t-tests.

Results:
Whole-brain averaged results: We observed greater clustering coefficient (t[27] = 2.53, p = 0.021) and reduced path length (t[27] = -2.84, p = 0.016), after ASM administration (Figure 1A). As seen in Figure 2B, 21/28 (75%) participants displayed greater small-world network topology after commencing ASM.

Region-specific results: Brain regions with greater clustering coefficient and reduced path length after ASM administration included the precuneus, inferior parietal lobe, anterior/mid insula and lateral frontotemporal areas (False Discovery Rate corrected, p< 0.05 – Figure 2).
Neuro Imaging