Department of Radiology, The First Affiliated Hospital, Nanchang University, Nanchang, Jiangxi, People's Republic of China.
Jiangxi Province Medical Imaging Research Institute, Nanchang, Jiangxi, People's Republic of China.
BMC Neurosci. 2020 Sep 15;21(1):37. doi: 10.1186/s12868-020-00590-4.
Although previous studies have shown that intra-network abnormalities in brain functional networks are correlated with clinical/cognitive impairment in multiple sclerosis (MS), there is little information regarding the pattern of causal interactions among cognition-related resting-state networks (RSNs) in different disease stages of relapsing-remitting MS (RRMS) patients. We hypothesized that abnormalities of causal interactions among RSNs occurred in RRMS patients in the acute and remitting phases.
Seventeen patients in the acute phases of RRMS, 24 patients in the remitting phases of RRMS, and 23 appropriately matched healthy controls participated in this study. First, we used group independent component analysis to extract the time courses of the spatially independent components from all the subjects. Then, the Granger causality analysis was used to investigate the causal relationships among RSNs in the spectral domain and to identify correlations with clinical indices.
Compared with the patients in the acute phase of RRMS, patients in the remitting phase of RRMS showed a significantly lower expanded disability status scale, modified fatigue impact scale scores, and significantly higher paced auditory serial addition test (PASAT) scores. Compared with healthy subjects, during the acute phase, RRMS patients had significantly increased driving connectivity from the right executive control network (rECN) to the anterior salience network (aSN), and the causal coefficient was negatively correlated with the PASAT score. During the remitting phase, RRMS patients had significantly increased driving connectivity from the rECN to the aSN and from the rECN to the visuospatial network.
Together with the disease duration (mean disease duration < 5 years) and relatively better clinical scores than those in the acute phase, abnormal connections, such as the information flow from the rECN to the aSN and the rECN to the visuospatial network, might provide adaptive compensation in the remitting phase of RRMS.
虽然先前的研究表明，脑功能网络的内网络异常与多发性硬化症（MS）的临床/认知障碍相关，但关于复发缓解型 MS（RRMS）患者不同疾病阶段与认知相关静息态网络（RSN）之间因果关系的模式信息较少。我们假设 RRMS 患者在急性和缓解期的 RSN 之间的因果关系存在异常。
本研究纳入 17 例 RRMS 急性期患者、24 例 RRMS 缓解期患者和 23 例匹配良好的健康对照者。首先，我们使用组独立成分分析从所有受试者中提取空间独立成分的时间序列。然后，使用格兰杰因果分析来研究 RSN 之间的因果关系，并确定与临床指标的相关性。
与 RRMS 急性期患者相比，RRMS 缓解期患者的扩展残疾状况量表评分、修正疲劳影响量表评分降低， paced auditory serial addition test（PASAT）评分升高。与健康受试者相比，在急性期，RRMS 患者右侧执行控制网络（rECN）到前突显网络（aSN）的驱动连通性增加，因果系数与 PASAT 评分呈负相关。在缓解期，RRMS 患者的 rECN 到 aSN 和 rECN 到视觉空间网络的驱动连通性增加。
与急性期相比，RRMS 缓解期的异常连接（如 rECN 到 aSN 和 rECN 到视觉空间网络的信息流）可能与疾病持续时间（平均疾病持续时间<5 年）和相对较好的临床评分有关，可能提供适应性代偿。
