BACKGROUND Cognitive decline remains difficult to predict as structural brain damage cannot fully explain the extensive heterogeneity found between MS patients. OBJECTIVE To investigate whether functional brain network organization measured with magnetoencephalography (MEG) predicts cognitive decline in MS patients after 5 years and to explore its value beyond structural pathology. METHODS Resting-state MEG recordings, structural MRI, and neuropsychological assessments were analyzed of 146 MS patients, and 100 patients had a 5-year follow-up neuropsychological assessment. Network properties of the minimum spanning tree (i.e. backbone of the functional brain network) indicating network integration and overload were related to baseline and longitudinal cognition, correcting for structural damage. RESULTS A more integrated beta band network (i.e. smaller diameter) and a less integrated delta band network (i.e. lower leaf fraction) predicted cognitive decline after 5 years (Radj2=15%), independent of structural damage. Cross-sectional analyses showed that a less integrated network (e.g. lower tree hierarchy) related to worse cognition, independent of frequency band. CONCLUSIONS The level of functional brain network integration was an independent predictive marker of cognitive decline, in addition to the severity of structural damage. This work thereby indicates the promise of MEG-derived network measures in predicting disease progression in MS.

中文翻译：

---

用脑磁图测量的功能性大脑网络组织预测多发性硬化症的认知能力下降

背景认知下降仍然难以预测，因为结构性脑损伤不能完全解释 MS 患者之间发现的广泛异质性。目的 研究脑磁图 (MEG) 测量的功能性脑网络组织是否能预测 MS 患者 5 年后的认知能力下降，并探索其超越结构病理学的价值。方法 对 146 名 MS 患者进行静息态 MEG 记录、结构 MRI 和神经心理学评估，其中 100 名患者进行了 5 年的随访神经心理学评估。表明网络整合和过载的最小生成树（即功能性大脑网络的主干）的网络特性与基线和纵向认知相关，纠正了结构损伤。结果 一个更加集成的 beta 波段网络（即 较小的直径）和整合度较低的 delta 带网络（即较低的叶分数）预测 5 年后认知能力下降（Radj2=15%），与结构损伤无关。横断面分析表明，集成度较低的网络（例如较低的树层次结构）与较差的认知相关，与频带无关。结论 除了结构损伤的严重程度外，功能性脑网络整合水平是认知能力下降的独立预测标志。因此，这项工作表明 MEG 衍生的网络测量在预测 MS 疾病进展方面的前景。较低的树层次）与较差的认知相关，与频带无关。结论 除了结构损伤的严重程度外，功能性脑网络整合水平是认知能力下降的独立预测标志。因此，这项工作表明 MEG 衍生的网络测量在预测 MS 疾病进展方面的前景。较低的树层次）与较差的认知相关，与频带无关。结论 除了结构损伤的严重程度外，功能性脑网络整合水平是认知能力下降的独立预测标志。因此，这项工作表明 MEG 衍生的网络测量在预测 MS 疾病进展方面的前景。