White matter network organization predicts memory decline after epilepsy surgery

biorxiv(2023)

引用 0|浏览0
暂无评分
摘要
Risk for memory decline is a common concern for individuals with temporal lobe epilepsy (TLE) undergoing surgery. Global and local network abnormalities are well-documented in TLE. However, it is less known whether network abnormalities predict post-surgical memory decline. We examined the role of pre-operative global and local white matter network organization in risk for post-operative memory decline in TLE. One-hundred one individuals with TLE (51 left; L-TLE) and 56 controls underwent T1-weighted, diffusion-MRI and neuropsychological memory testing in a prospective longitudinal study. Forty-four patients subsequently underwent temporal lobe surgery and post-operative memory testing. Structural connectomes were generated via tractography and analyzed using measures of global (integration and specialization) and local (medial temporal lobe [MTL] specialization) network organization. Pre-operatively, higher global network integration and specialization were associated with higher verbal memory function in L-TLE. Higher pre-operative global network integration and specialization, as well as more leftward local specialization, predicted greater post-surgical verbal memory decline for L-TLE. Accounting for pre-operative memory score and hippocampal volume asymmetry, local network specialization additionally explained 25-33% of the variance in outcomes for L-TLE and outperformed hippocampal asymmetry. Local network specialization alone produced good diagnostic accuracy (AUC=.80-.84) in L-TLE. Global white matter network disruption contributes to verbal memory impairment pre-operatively and predicts post-surgical memory outcomes in L-TLE. However, leftward asymmetry of MTL network organization confers the highest risk for memory decline. This demonstrates the importance of characterizing local network properties within the to-be-operated hemisphere and the reserve capacity of the contralateral MTL network. ### Competing Interest Statement The authors have declared no competing interest.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要