Abstracts

Rationale: Patients with temporal lobe epilepsy (TLE) have been shown to exhibit abnormalities in the volume of specific brain regions, as well as in the connectivity strength between regions. It is currently unclear if, and how, abnormalities in these distinct modalities are related.

Methods: Data from 144 patients with TLE and 96 healthy controls. From T1-weighted MRI, we calculated the volume of each brain region in the D-K atlas (82 ROIs), for each subject. Accounting for age, sex and intra-cranial volume (ICV), the volume of each region of interest (ROI) was z-scored against healthy controls. ROIs were considered abnormal (in terms of volume) if |z_vol| > 1.96. From diffusion MRI, connectivity matrices were inferred for each subject using fractional anisotropy (FA). Accounting for age and sex, each connection was z-scored against healthy controls. Connections were considered abnormal if |z_conn| > 1.96. ROIs were considered abnormal (in terms of connectivity) if at least 5% of connections to/from were abnormal. Odds ratios were used to determine which ROIs are abnormal at a rate above and beyond that expected in healthy controls.

Results: Volumetric and structural connectivity abnormalities occurred in the same ROIs in some, but not all patients. At a group-level, only in the ipsilateral hippocampus did abnormalities occur in both modalities at a significantly greater rate than in controls. Volume-only abnormalities were primarily restricted to ipsilateral temporal and subcortical regions. Connectivity-only abnormalities were more numerous and widespread, but still primarily constrained to the ipsilateral hemisphere. Interestingly, we only rarely observed connectivity abnormalities, without corresponding volumetric abnormalities in the ROIs typically associated with TLE, i.e. ipsilateral temporal and subcortical ROIs.

Conclusions: Abnormalities in different modalities may represent different aspects or stages of TLE. This is the first study to explicitly investigate the relationship between volumetric and structural connectivity abnormalities within patients with TLE. The identification of the ipsilateral hippocampus as abnormal in both modalities simultaneously underpins its key role in TLE, where it is often the source of epileptogenic activity. The lack of accompanying connectivity abnormalities in other ROIs typically associated with epileptogenicity may hint at the mechanisms of TLE progression. Specifically, we suggest that the initial epileptogenic ROI(s) may exhibit abnormal volumes, and this epileptogenicity may, over time, spread to other ROIs, resulting in connectivity abnormalities. Improving our understanding of where TLE originates and how the disorder progresses is key to improving treatment for patients.

Funding: Please list any funding that was received in support of this abstract.: J.S.D. and S.B.V. were funded by the UCLH NIHR BRC. Scan acquisition and G.P.W. were supported by the MRC (G0802012, MR/M00841X/1). We are grateful to the Epilepsy Society for supporting the Epilepsy Society MRI scanner. Y.W. gratefully acknowledges funding from Wellcome Trust (208940/Z/17/Z). P.N.T. is supported by a UKRI Future Leaders Fellowship (MR/T04294X/1). J.H. is supported by the Centre for Doctoral Training in Cloud Computing for Big Data (EP/L015358/1).