Department of Clinical and Experimental Epilepsy, UCL Queen Square Institute of Neurology, London, UK.
MRI Unit, Epilepsy Society, Buckinghamshire, UK.
Epilepsia. 2024 Sep;65(9):2641-2661. doi: 10.1111/epi.18058. Epub 2024 Jul 11.
Anterior temporal lobe resection (ATLR) effectively controls seizures in medically refractory temporal lobe epilepsy but risks significant episodic memory decline. Beyond 1 year postoperatively, the influence of preoperative clinical factors on episodic memory and long-term network plasticity remain underexplored. Ten years post-ATLR, we aimed to determine biomarkers of successful memory network reorganization and establish presurgical features' lasting impact on memory function.
Twenty-five ATLR patients (12 left-sided) and 10 healthy controls underwent a memory-encoding functional magnetic resonance imaging paradigm alongside neuropsychometry 10 years postsurgery. Generalized psychophysiological interaction analyses modeled network functional connectivity of words/faces remembered, seeding from the medial temporal lobes (MTLs). Differences in successful memory connectivity were assessed between controls and left/right ATLR. Multivariate regressions and mixed-effect models probed preoperative phenotypes' effects on long-term memory outcomes.
Ten years post-ATLR, lower baseline functioning (verbal and performance intelligence quotient) and a focal memory impairment preoperatively predicted worse long-term memory outcomes. Poorer verbal memory was significantly associated with longer epilepsy duration and earlier onset age. Relative to controls, successful word and face encoding involved increased functional connectivity from both or remnant MTL seeds and contralesional parahippocampus/hippocampus after left/right ATLR. Irrespective of surgical laterality, successful memory encoding correlated with increased MTL-seeded connectivity to frontal (bilateral insula, right anterior cingulate), right parahippocampal, and bilateral fusiform gyri. Ten years postsurgery, better memory performance was correlated with contralateral frontal plasticity, which was disrupted with longer epilepsy duration.
Our findings underscore the enduring nature of functional network reorganizations to provide long-term cognitive support. Ten years post-ATLR, successful memory formation featured stronger connections near resected areas and contralateral regions. Preoperative network disruption possibly influenced effectiveness of postoperative plasticity. These findings are crucial for enhancing long-term memory prediction and strategies for lasting memory rehabilitation.
前颞叶切除术（ATLR）可有效控制药物难治性颞叶癫痫发作，但存在显著的情景记忆下降风险。术后 1 年以上，术前临床因素对情景记忆和长期网络可塑性的影响仍未得到充分探索。在 ATLR 术后 10 年，我们旨在确定成功记忆网络重组的生物标志物，并确定术前特征对记忆功能的持久影响。
25 名 ATLR 患者（12 名左侧）和 10 名健康对照者在术后 10 年接受了记忆编码功能磁共振成像范式以及神经心理学测试。广义心理生理交互分析模型从内侧颞叶（MTLs）种子模拟了单词/面孔记忆的网络功能连接。评估了对照组和左/右侧 ATLR 之间成功记忆连接的差异。多元回归和混合效应模型探测了术前表型对长期记忆结果的影响。
ATLR 术后 10 年，基线功能（言语和操作智商）较低和术前局灶性记忆障碍预测长期记忆结果较差。言语记忆较差与癫痫持续时间较长和发病年龄较早显著相关。与对照组相比，成功的单词和面孔编码涉及到左/右侧 ATLR 后双侧或残余 MTL 种子和对侧旁海马/海马的功能连接增加。无论手术侧如何，成功的记忆编码与增加的 MTL 种子到额（双侧岛叶、右侧前扣带回）、右侧旁海马和双侧梭状回的连接相关。术后 10 年，更好的记忆表现与对侧额叶可塑性相关，而较长的癫痫持续时间则破坏了这种可塑性。
我们的研究结果强调了功能网络重组的持久性质，为长期认知支持提供了依据。ATLR 术后 10 年，成功的记忆形成表现为在切除区域和对侧区域附近具有更强的连接。术前网络中断可能影响术后可塑性的有效性。这些发现对于提高长期记忆预测和持久记忆康复策略至关重要。
