Authors :
Presenting Author: Fenglai Xiao, MD PhD – Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology,University College London
Davide Giampiccolo, MD – Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology,University College London; Lawrence Binding, PhD – Department of Clinical and Experimental Epilepsy, Institute of Neurology, Faculty of Brain Sciences, University College London; Marine Fleury, PhD – Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology,University College London; Bernardo Pimentel, MD – Department of Neurology - Christian-Doppler University Hospital Paracelsus Medical University; Luisa Delazer, MD – Ludwig Maximilian University of Munich; Marian Galovic, MD – University of Zurich; Lorenzo Caciagli, MD PhD – Bern University Hospital; John, S Duncan, MD PhD FRCP – Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology,University College London; Britta Wandschneider, MD PhD – Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology,University College London; Matthias J Koepp, MD PhD – Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology,University College London
Rationale:
Adults with focal epilepsy show progressive atrophy on longitudinal magnetic resonance imaging (MRI), but it is not known whether such progression is also seen in adults with idiopathic generalised epilepsy (IGE).
Methods:
Retrospectively, we analysed longitudinal MRI scans from 67 individuals diagnosed with IGE between 2004 and 2019 at a tertiary epilepsy referral center. For clinical reasons, ongoing generalised tonic-clonic seizures (GTCS), participants underwent two or more high-resolution T1-weighted MRI scans spaced at least six months apart. Computational Anatomy Toolbox (CAT12) running in Statistical Parametric Mapping 12 (SPM12) and Matlab 2021a (Mathworks) was employed to estimate cortical thickness. To measure hippocampal volumes, we employed Hipposeg (http://niftyweb.cs.ucl.ac.uk/program.php?p=HIPPOSEG), an open-source, multi-atlas-based, previously-validated hippocampal segmentation algorithm. Volumes of other subcortical structures relevant in epilepsy, including the thalamus, amygdala, caudate, putamen, and globus pallidus, and total intracranial volume (TIV), were extracted using a parcellation algorithm based on Geodesic Information Flows (GIF), freely available within NiftyWeb (http://cmictig.cs.ucl.ac.uk/niftyweb, UCL Centre for Medical Image Computing, UK). Vertexwise cortical thickness measurements were analysed with CAT12 fitted in a flexible factorial analysis. We fitted linear mixed-effects models in SPSS for hippocampal and subcortical volumes. All models were corrected for a random effect of participant and fixed effects of age at scan, sex and group.
Results:
Significant increases in cortical thickness were observed in bilateral middle occipital lobes and lingual gyri (both left and right: p< 0.0001, FWE-corrected), the bilateral superior parietal lobes (left: p=0.002, FWE-corrected; right: p=0.018, FWE-corrected), and the bilateral middle temporal gyri (left: p=0.036, FWE-corrected; right: p=0.032, FWE-corrected). Cortical thinning was noted in the left superior temporal gyrus (p=0.027, FWE-corrected), the anterior cingulate cortex (p=0.006, FWE-corrected), the right rolandic opercula (p=0.015, FWE-corrected) and bilaterally in the thalamus (left: p=0.039, Bonferroni-corrected; right: p=0.003, Bonferroni-corrected), putamen (left: p=0.012, Bonferroni-corrected; right: p=0.001, Bonferroni-corrected), and pallidum (left: p=0.003, Bonferroni-corrected; right: p=0.001, Bonferroni-corrected).