33-24 – The impact of white matter hyperintensities on bundle tract disconnection and surgical outcomes in temporal lobe epilepsy

Epilepsy is one of the most common chronic neurological disorders with 30-40% of patients having temporal lobe epilepsy (TLE). Many patients with TLE are pharmacoresistant and may be offered surgical resection of the temporal lobe as an alternative treatment, but 30-50% fail to achieve seizure freedom after surgery. Increasing evidence suggests that epilepsy is not caused by a single focal region and therefore surgical resection fails to remove the epilepsy network. White matter (WM) abnormalities have been identified in patients with TLE and may be a potential biomarker to show the extent of the epileptogenic network for a better prediction of surgical outcome.

For this study, it is hypothesised that patients with TLE and a greater burden of WM disconnection will have worse postoperative seizure outcomes compared to patients with a reduced burden. This will be assessed using data collected from the Medical University Bonn, Germany from 122 patients with TLE with both T1 and T2-FLAIR MRI preoperative scans, subdivided into 5 groups by Engel’s score – a measure of postoperative seizure outcome. DeepLesionBrain (DLB) – an online open access resource – was selected for use to auto-segment the brain providing detailed volumetric data and identification of WM lesions. DLB quantifies the degree to which these lesions impact 64 normative WM tract bundles resulting in a disconnectome map. A Kruskal-Wallis test will be used to compare the 5 Engel’s groups against the WM lesion burden and disconnection probabilities of each tract, with a follow-up Bonferroni test for multiple comparisons.

The aim of this research is to better understand the relationship between preoperative imaging, postoperative outcomes and changes to WM through detection of hyperintensities on preoperative scans. This will be key to understanding the impact of TLE on brain connectivity and could lead to the development of a novel biomarker that predicts surgical outcome based on preoperative imaging.