Seizure freedom after surgical resection of diffusion-weighted magnetic resonance imaging abnormalities

Jonathan Horsley; Gerard Hall; Callum Simpson; Csaba Kozma; Rhys Thomas; Yujiang Wang; Jane de Tisi; Anna Miserocchi; Andrew McEvoy; Sjoerd Vos; Gavin Winston; John Duncan; Peter N Taylor

doi:10.1111/epi.18490

Seizure freedom after surgical resection of diffusion-weighted magnetic resonance imaging abnormalities

Epilepsia. 2025 Jun 10. doi: 10.1111/epi.18490. Online ahead of print.

Authors

Jonathan Horsley¹, Gerard Hall¹, Callum Simpson¹, Csaba Kozma¹, Rhys Thomas², Yujiang Wang^{1

2

3}, Jane de Tisi³, Anna Miserocchi³, Andrew McEvoy³, Sjoerd Vos^{3

4

5}, Gavin Winston^{3

6}, John Duncan³, Peter N Taylor^{1

2

3}

Affiliations

¹ CNNP Lab, Interdisciplinary Computing and Complex BioSystems Group, School of Computing, Newcastle University, Newcastle Upon Tyne, UK.
² Translational and Clinical Research Institute, Faculty of Medical Sciences, Newcastle University, Newcastle Upon Tyne, UK.
³ Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology, University College London, London, UK.
⁴ Centre for Microscopy, Characterisation, and Analysis, University of Western Australia, Nedlands, Western Australia, Australia.
⁵ Centre for Medical Image Computing, Computer Science Department, University College London, London, UK.
⁶ Division of Neurology, Department of Medicine, Queen's University, Kingston, Ontario, Canada.

PMID: 40492909
DOI: 10.1111/epi.18490

Abstract

Objective: Successful epilepsy surgery requires accurate localization and removal of the epileptogenic zone. Neuroimaging helps detect structural brain abnormalities to guide surgery, but current clinical practice does not use diffusion-weighted magnetic resonance imaging (dwMRI). However, previous work has shown that diffusion abnormalities are present in epilepsy and may relate to the epileptogenic zone. Here, we investigate whether surgical resection of diffusion abnormalities relates to postoperative seizure freedom.

Methods: We investigated the association between surgical resection of diffusion abnormalities and postoperative seizure freedom in 200 individuals with drug-resistant focal epilepsy using dwMRI. A cohort of 97 healthy controls provided a normative baseline for dwMRI metrics, allowing calculation of voxelwise z-scores to identify abnormal clusters in both gray and white matter.

Results: Surgical resections overlapping with the largest abnormal cluster significantly correlated with sustained seizure freedom at 12 months (83% vs. 55%; $p < .0001$ ) and over 5 years ( $p < .0001$ ). Notably, resecting only a small proportion of the largest cluster was associated with better seizure outcomes than cases with no resection of this cluster ( $p = .008$ ). Furthermore, sparing the largest cluster but resecting other large clusters still improved seizure freedom rates compared to no overlap ( $p = .03$ ).

Significance: Our results suggest that abnormal clusters, identified using dwMRI, are integral to the epileptogenic network, and even a partial removal of such an abnormal cluster is sufficient to achieve seizure freedom. This study highlights the potential of incorporating dwMRI into presurgical planning to improve outcomes in focal epilepsy by reliably identifying and targeting diffusion abnormalities.

Keywords: computational; epilepsy; neuroimaging; surgery; treatment.

Abstract

Grants and funding