Communication Sciences and Disorders, University of South Carolina, Columbia, South Carolina, USA.
Center for Stroke Research Berlin, Charité-Universitätsmedizin Berlin, Berlin, Germany.
Hum Brain Mapp. 2024 Oct;45(14):e70038. doi: 10.1002/hbm.70038.
The contribution of age-related structural brain changes to the well-established link between aging and cognitive decline is not fully defined. While both age-related regional brain atrophy and cognitive decline have been extensively studied, the specific mediating role of age-related regional brain atrophy on cognitive functions is unclear. This study introduces an open-source software tool with a graphical user interface that streamlines advanced whole-brain mediation analyses, enabling researchers to systematically explore how the brain acts as a mediator in relationships between various behavioral and health outcomes. The tool is showcased by investigating regional brain volume as a mediator to determine the contribution of age-related brain volume loss toward cognition in healthy aging. We analyzed regional brain volumes and cognitive testing data (Montreal Cognitive Assessment [MoCA]) from a cohort of 131 neurologically healthy adult participants (mean age 50 ± 20.8 years, range 20-79, 73% females) drawn from the Aging Brain Cohort Study at the University of South Carolina. Using our open-source tool developed for evaluating brain-behavior associations across the brain and optimized for exploring mediation effects, we conducted a series of mediation analyses using participant age as the predictor variable, total MoCA and MoCA subtest scores as the outcome variables, and regional brain volume as potential mediators. Age-related atrophy within specific anatomical networks was found to mediate the relationship between age and cognition across multiple cognitive domains. Specifically, atrophy in bilateral frontal, parietal, and occipital areas, along with widespread subcortical regions mediated the effect of age on total MoCA scores. Various MoCA subscores were influenced by age through atrophy in distinct brain regions. These involved prefrontal regions, sensorimotor cortex, and parieto-occipital areas for executive function subscores, prefrontal and temporo-occipital regions, along with the caudate nucleus for attention and concentration subscores, frontal and parieto-occipital areas, alongside connecting subcortical areas such as the optic tract for visuospatial subscores and frontoparietal areas for language subscores. Brain-based mediation analysis offers a causal framework for evaluating the mediating role of brain structure on the relationship between age and cognition and provides a more nuanced understanding of cognitive aging than previously possible. By validating the applicability and effectiveness of this approach and making it openly available to the scientific community, we facilitate the exploration of causal mechanisms between variables mediated by the brain.
年龄相关的结构性脑变化对已确立的衰老与认知能力下降之间的关系的贡献尚不完全明确。虽然年龄相关的区域性脑萎缩和认知能力下降都得到了广泛研究,但年龄相关的区域性脑萎缩对认知功能的具体中介作用尚不清楚。本研究引入了一个带有图形用户界面的开源软件工具,该工具简化了高级全脑中介分析,使研究人员能够系统地探索大脑在各种行为和健康结果之间的关系中作为中介的作用。该工具通过研究区域性脑容量作为中介来确定与健康衰老相关的脑容量损失对认知的贡献,从而得到展示。我们分析了来自南卡罗来纳大学衰老大脑队列研究的 131 名神经健康成人参与者(平均年龄 50 ± 20.8 岁,范围 20-79 岁,73%为女性)的区域性脑容量和认知测试数据(蒙特利尔认知评估[MoCA])。使用我们开发的用于评估大脑与行为之间关联的开源工具,并针对探索中介效应进行了优化,我们使用参与者年龄作为预测变量、总 MoCA 和 MoCA 子测试分数作为结果变量,以及区域性脑容量作为潜在中介变量进行了一系列中介分析。在多个认知领域中,特定解剖网络中的与年龄相关的萎缩被发现介导了年龄与认知之间的关系。具体来说,双侧额叶、顶叶和枕叶区域以及广泛的皮质下区域的萎缩介导了年龄对总 MoCA 分数的影响。各种 MoCA 子分数受到年龄的影响,是由于大脑不同区域的萎缩。这涉及到执行功能子分数的前额叶区域、感觉运动皮层和顶枕叶区域,注意力和集中力子分数的前额叶和颞顶叶区域,以及尾状核,注意力和集中力子分数的额叶和顶枕叶区域,以及连接皮质下区域,如视束,用于视空间子分数和额顶叶区域用于语言子分数。基于大脑的中介分析为评估大脑结构在年龄与认知之间关系中的中介作用提供了一个因果框架,并提供了比以前更细致的认知老化理解。通过验证这种方法的适用性和有效性,并向科学界公开提供,我们促进了对大脑介导的变量之间因果机制的探索。