阻塞性睡眠呼吸暂停患者默认网络小世界属性的fMRI研究

Small-World Property of the Default Mode Network in Obstructive Sleep Apnea:A Resting-State Functional Magnetic Resonance Imaging Study

目的采用基于图论的分析方法研究阻塞性睡眠呼吸暂停(OSA)患者默认网络(DMN)的"小世界"网络拓扑属性改变。方法搜集首次确诊的重度OSA患者及年龄、受教育程度相似的正常睡眠组(GS组)各46例进行静息态功能磁共振成像(rs-fMRI)数据采集,选取20个特定的DMN感兴趣区(ROI)构建二值化网络,获取并比较DMN整体及局部属性参数,分析网络参数与临床变量的相关性。结果 OSA及GS组DMN均存在"小世界"属性。与GS组相比,OSA患者DMN在0.05≤稀疏度(Sp)≤0.12和0.15≤Sp≤0.50范围内聚类系数(Cp),Eloc显著减低,OSA组DMN的特征性最短路径长度(Lp),Eglob值无明显变化。与GS组相比,OSA患者左侧后扣带回的节点度、节点介数、节点效率均显著减低,而右侧海马旁回节点度、节点介数、节点效率均显著增高;背内侧前额叶皮层的节点介数显著减低,而腹内侧前额叶皮层节点介数显著增高。结论 OSA患者DMN具有"小世界"属性,但其整体网络属性和局部属性均存在改变,且血氧饱和度减低可能导致OSA患者DMN拓扑属性重塑,而OSA患者DMN的网络属性异常可进一步解释OSA患者认知功能障碍。

Objective To study the changes of the "small world" network topology properties of the default network(DMN) of patients with obstructive sleep apnea(OSA) using graph theory-based analysis.Methods A total of 46 patients with severe OSA diagnosed at the first time and normal sleep group(GS group) with similar education and age were selected for resting f MRI data collection.20 specific DMN regions of interest were selected to construct a binary network.The overall and local attribute parameters of DMN were compared and the correlation between network parameters and clinical variables was analyzed.Results There are " small world" attributes in both OSA and GS groups.Compared with the GS group,in the OSA patients,the Cp and Eloc were significantly decreased in the range of 0.05 ≤ Sp ≤ 0.12 and 0.15≤ Sp ≤ 0.50.There was no significant change in the Lp and Eglob values of the DMN in the OSA group.Compared with the GS group,the node degree,node number,and node efficiency of the left posterior cingulate gyrus of OSA patients were significantly reduced,while the right hippocampal bypass node degree,node number,and node efficiency were significantly increased;The node media of the cortex was significantly reduced,while the median prefrontal cortex nodes were significantly increased.Conclusion DMN in OSA patients has a " small world" attribute,but its overall network attributes and local attributes are changed.The decrease of blood oxygen saturation may lead to the remodeling of DMN topological attributes in OSA patients,and the abnormal network attributes of OSN patients' DMN can be further explained.Cognitive dysfunction in OSA patients.