Preoperative white matter network organization and memory decline after epilepsy surgery

Area of Science:

• Neuroscience
• Epilepsy Research
• Brain Imaging

Background:

• Temporal lobe epilepsy (TLE) frequently leads to memory decline, a significant concern for patients considering surgery.
• While global and local network abnormalities in TLE are documented, their predictive value for postsurgical memory decline remains unclear.
• Understanding these network disruptions is crucial for predicting and potentially mitigating cognitive deficits.

Purpose of the Study:

• To investigate the role of preoperative global and local white matter network organization in predicting postsurgical memory decline in TLE patients.
• To determine if network abnormalities, specifically medial temporal lobe (MTL) network asymmetry, are associated with verbal memory outcomes after surgery.
• To compare the predictive power of MTL network asymmetry against hippocampal volume and global network metrics.

Main Methods:

• Prospective longitudinal study of 101 individuals with TLE and 56 controls, including T1-weighted MRI, diffusion MRI, and neuropsychological testing.
• Generation of preoperative structural connectomes using diffusion tractography to analyze global network integration/specialization and local MTL network asymmetry.
• Postoperative memory testing for 44 TLE patients who underwent temporal lobe surgery.

Main Results:

• Higher preoperative global network integration and specialization correlated with better preoperative verbal memory in left TLE.
• Greater leftward MTL network asymmetry and higher global network metrics predicted significant verbal memory decline post-surgery in left TLE patients.
• MTL network asymmetry uniquely explained 25%-33% of verbal memory decline variance in left TLE, outperforming other metrics and showing good diagnostic classification.

Conclusions:

• Preoperative global white matter network disruption contributes to verbal memory impairment and predicts postsurgical outcomes in left TLE.
• Leftward asymmetry in MTL white matter networks appears to be a key predictor of verbal memory decline risk.
• Characterizing preoperative local white matter network properties is vital for presurgical planning and understanding contralateral network reserve capacity.