Altered Cortical Microstructure and Morphology Within and Beyond Resected Epileptic Foci

438 

Abstract Submission 

Ella Sahlas¹, Alexander Ngo¹, Raúl Rodriguez-Cruces¹, Thaera Arafat¹, Jessica Royer¹, Ke Xie¹, Hans Auer¹, Judy Chen¹, Raluca Pana¹, Birgit Frauscher², Neda Bernasconi¹, Andrea Bernasconi¹, Boris Bernhardt¹

¹Montreal Neurological Institute and Hospital, Montreal, QC, ²Duke University, Durham, NC

Ella Sahlas  
Montreal Neurological Institute and Hospital
Montreal, QC

Alexander Ngo  
Montreal Neurological Institute and Hospital
Montreal, QC

Raúl Rodriguez-Cruces  
Montreal Neurological Institute and Hospital
Montreal, QC

Thaera Arafat  
Montreal Neurological Institute and Hospital
Montreal, QC

Jessica Royer  
Montreal Neurological Institute and Hospital
Montreal, QC

Ke Xie  
Montreal Neurological Institute and Hospital
Montreal, QC

Hans Auer  
Montreal Neurological Institute and Hospital
Montreal, QC

Judy Chen  
Montreal Neurological Institute and Hospital
Montreal, QC

Raluca Pana  
Montreal Neurological Institute and Hospital
Montreal, QC

Birgit Frauscher  
Duke University
Durham, NC

Neda Bernasconi  
Montreal Neurological Institute and Hospital
Montreal, QC

Andrea Bernasconi  
Montreal Neurological Institute and Hospital
Montreal, QC

Boris Bernhardt  
Montreal Neurological Institute and Hospital
Montreal, QC

The advent of high-resolution neuroimaging has revolutionized how epilepsy is understood, diagnosed, and treated[1]. In particular, magnetic resonance imaging (MRI) methods have enabled previously unparalleled precision in localizing tissue responsible for epileptic seizures and assessing seizure etiology[2]. Localizing epileptogenic tissue is especially critical to delivering care to patients with pharmaco-resistant seizures, for whom resective surgery is often the most effective treatment[3-5]. Research has also revealed global changes in the properties of the cortex in epilepsy patients relative to controls[6]. However, there is a clear need to advance precise and non-invasive methods of localizing the origin of seizure activity and uncovering alterations beyond the seizure focus. This will ultimately allow more pharmaco-resistant patients to benefit from targeted resection that maximally spares surrounding tissue, with fewer risks than invasive pre-surgical investigations[7-9].

Patients were a consecutive cohort of 22 adults (17F, mean±SD age = 34.55±11.44 years) with pharmaco-resistant focal epilepsy who underwent pre-operative MRI at the Montreal Neurological Institute and Hospital, a resective neurosurgical procedure (14 temporal, 4 parietal, 2 occipital, and 2 frontal), and post-operative MRI. Of the 21 Engel outcomes available, 12 were IA, 4 were IB, 1 was IIA, 1 was IIIa, and 3 were IVC. The healthy control cohort comprised 100 adults (46F, mean±SD age = 31.62± 9.30 years) who underwent the same MRI protocol as patients did pre-surgery. This high-resolution 3T MRI protocol included diffusion-weighted imaging (DWI, 1.6mm isovoxels), T1-weighted scans (T1w, 0.8mm isovoxels), quantitative T1 relaxometry (0.8mm isovoxels), and resting-state functional MRI (rsfMRI, 3mm isovoxels). We preprocessed this data using micapipe[10]. Maps of mean diffusivity (MD) and fractional anisotropy (FA) were derived from DWI, maps of cortical thickness (CT) were derived from T1w scans, and maps of quantitative T1 relaxation time (qT1) were derived from quantitative T1 relaxometry; we normalized these features vertex-wise in each patient relative to controls (Fig. 1A). For each patient, we generated a functional connectivity matrix from rsfMRI, a structural connectivity matrix from DWI, and a geodesic distance matrix from T1w imaging. We segmented the resection site semi-automatically in each patient's post-operative T1w scan, co-registered with pre-operative T1w scans. We mapped the resection site to the surface and determined the 0-5%, 5-10%, 10-15%, 15-20%, 20-25%, 25-30%, 30-35%, 35-40%, 40-45%, and 45-50% of vertices most functionally coupled, structurally connected, and geodesically close to this site (Fig. 1B). We quantified relative alterations in MD, FA, CT, and qT1 in the surgical target and as a function of functional, structural, and geodesic distance from the target.

Alterations were significantly increased within the surgical target relative to the rest of the brain for MD in 19/22 patients (86.36%), FA in 19/22 patients (86.36%), CT in 18/22 patients (82.82%), and qT1 in 18/22 patients (82.82%), z-tests, FDR-adjusted p < 0.01. Group mean MD, FA, CT, and qT1 alterations decreased with increasing functional connectivity distance from the target (Fig. 2A), as well as with structural connectivity distance (Fig. 2B) and geodesic distance (Fig. 2C). For each cortical feature and across all three distance metrics, alterations were significantly higher in the surgical target than in vertices at the fifth nearest level of distance to the target, paired t-tests, p < 0.01.

Mapping the epicenter of alterations in cortical microstructure and morphology in individual patients carries potential to assist in localizing the surgical target non-invasively. Functional, structural, and geodesic relationships to the epileptic focus may impact the magnitude of microstructural and morphological changes in cortical regions beyond the focus of seizures.

Neurodevelopmental/ Early Life (eg. ADHD, autism) ¹

Connectivity (eg. functional, effective, structural)

Diffusion MRI Modeling and Analysis

Image Registration and Computational Anatomy

Multi-Modal Imaging ²

Computational Neuroscience

Cortex

Data analysis

Data Registration

DISORDERS

Epilepsy

FUNCTIONAL MRI

MRI

Neurological

STRUCTURAL MRI