Authors :
Presenting Author: TIPAKORN TUMNARK, MD – University of Pittsburgh medical school
Jessica Barrios-Martinez, MD – university of pittsburgh medical school
Chandana Belly, MS – University of Pittsburgh Medical School
Lilly Tang, MD – University of Pittsburgh Medical School
Yeh Fang-Cheng, MD,PhD – University of Pittsburgh Medical School
Thandar Aung, MD,MS – University of Pittsburgh Medical School
Jorge González-Martínez, MD,PhD – University of Pittsburgh Medical School
Rationale:
Temporal lobe epilepsy (TLE) is the most common form of drug-resistant (DR) focal epilepsy, yet a substantial proportion of patients continue to experience seizures after anterior temporal lobectomy. We aimed to investigate the contribution of thalamo-cortical structural connectivity to surgical outcomes in this cohort. Methods:
With the approval of the University of Pittsburgh IRB, 35 consecutive patients with DR-TLE undergoing SEEG between 2020-2021 followed by resective surgery (≥12 months follow-up). As a normative cohort, 97 healthy controls from the Human Connectome Project were included. Pre-implantation 256-direction diffusion MRI was analyzed. Diffusion metrics were computed bilaterally, and the temporal lobe ipsilateral to the SEEG-defined seizure-onset zone (epileptic side’) was compared with the contralateral side (within-subject analysis). Fiber tracking was performed using both automated tractography pipelines and manual region-of-interest (ROI)–guided selection to ensure accurate delineation of relevant tracts. From these tracts, quantitative anisotropy (QA) and fractional anisotropy (FA) were extracted. Patients were then stratified by surgical outcome into seizure-free (Engel class IA) and non–seizure-free (all other Engel classes). Group-level comparisons of QA and FA were performed across three groups: seizure-free (SF), non–seizure-free (NSF), and healthy controls. Multiple comparisons were controlled using the Benjamini–Hochberg false discovery rate (FDR) with a significance threshold of q = 0.05. Results:
Among 35 patients, 17 became SF (mean age 40.82 y, range 23-73) and 18 were NSF (mean age 46.03 y, range 24-80). For normal controls, the mean age of 97 healthy controls were 31.80, range 13-62). The duration of epilepsy in the SF group is 14.71 years, (range 3-33 years) and in NSF group is 14.72 years, (range 3-44 years). A total of 35 DR-TLE patients and a normal controlled population of 97 patients were enrolled. No significant difference in structural connectivity between SF, NSF and normal controlled groups, using the automatic fiber tracking was performed for the uncinate fasciculus, cingulum, fornix, hippocampal alveus, and thalamic radiation. Using the manual ROI selection based on the SEEG electrode contacts between ipsilateral hippocampus and pulvinar showed significantly increased FA and QA in NSF group, when compared to the SF group (FDR= 0.001953). In addition, when compared to the normal population, the SF group has no significant difference in structural connectivity (FR =1.0). Conclusions:
Our findings suggested that thalamo-cortical structural connectivity is closely linked to surgical outcomes in DR-TLE, highlighting its potential as a structural biomarker to guide prognostication and improve patient selection.
Funding: no specific funding supported the research.