Supplementary Material for: Multimorbidity and Regional Volumes of the Default Mode Network in Brain Aging

Introduction: The default mode network (DMN) is selectively vulnerable in brain aging. Little is known about the effect of multimorbidity as a whole onto the brain structural integrity. Objective: We aimed to investigate the association between multimorbidity and the structural integrity of DMN. Methods: We enrolled senior volunteers aged between 60 and 80 years in Hualien County during 2014–2018 and conducted in-person interview to collect information on chronic diseases. Fasting blood glucose and glycated hemoglobin (HbA1c) were tested. We assessed multimorbidity burden by the cumulative illness rating scale-geriatric (CIRS-G). MRI brain scans were standardized to measure the regional volume within the DMN. In a cross-sectional design, we employed stepwise regression models to evaluate the effects of age, sex, hyperglycemia, and multimorbidity on the DMN. Results: A total of 170 volunteers were enrolled with a mean age of 66.9 years, female preponderance (71%), an average mini-mental state examination score of 27.6, a mean HbA1c of 6.0, and a mean CIRS-G total score (TS) of 7.2. We found that older age was associated with reduced volumes in the hippocampus, left rostral anterior cingulate cortex, right posterior cingulate, right isthmus, precuneus, and right supramarginal. Higher levels of HbA1c and fasting glucose were associated with a reduced volume in the hippocampus only. A higher CIRS-G-TS was associated with reduced volumes in the left posterior cingulate cortex and right supramarginal gyrus; while a higher CIRS-G severity index was associated with a smaller right precuneus and right supramarginal. Conclusions: In the DMN, hippocampal volume shows vulnerability to aging and hyperglycemia, whereas the posterior cingulate, supramarginal, and precuneus cortices may be the key sites to reflect the total effects of multimorbidity.

引言：默认模式网络（default mode network, DMN）在脑衰老过程中呈现选择性易损性。目前学界对共病整体对脑结构完整性的影响尚缺乏深入认知。研究目的：本研究旨在探讨共病与默认模式网络结构完整性之间的关联。方法：本研究于2014至2018年间在花莲县招募60至80岁的老年志愿者，通过面对面访谈收集慢性病相关信息，并检测空腹血糖与糖化血红蛋白（glycated hemoglobin, HbA1c）水平。采用老年累积疾病评定量表（cumulative illness rating scale-geriatric, CIRS-G）评估共病负荷。对脑部MRI扫描进行标准化处理，以测量默认模式网络内各脑区的体积。本研究采用横断面设计，通过逐步回归模型分析年龄、性别、高血糖状态及共病对默认模式网络体积的影响。结果：本研究共纳入170名志愿者，平均年龄为66.9岁，女性占比71%，迷你精神状态检查（mini-mental state examination）平均得分为27.6，平均HbA1c水平为6.0，平均CIRS-G总分（TS）为7.2。研究发现，年龄增长与海马体、左侧喙前扣带回皮层、右侧后扣带回、右侧峡部、楔前叶及右侧缘上回的体积减小显著相关。较高的HbA1c水平与空腹血糖仅与海马体体积减小相关。更高的CIRS-G总分与左侧后扣带回皮层及右侧缘上回体积减小相关；而更高的CIRS-G严重程度指数则与右侧楔前叶及右侧缘上回体积更小相关。结论：在默认模式网络中，海马体积对衰老与高血糖呈现易损性，而后扣带回、缘上回及楔前叶皮层或为反映共病整体影响的关键脑区。