Liao Weiqi, Long Xiaojing, Jiang Chunxiang, Diao Yanjun, Liu Xin, Zheng Hairong, Zhang Lijuan
Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, 1068 Xueyuan Ave., Shenzhen, Guangdong Province 518055, China.
Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, 1068 Xueyuan Ave., Shenzhen, Guangdong Province 518055, China.
Acad Radiol. 2014 May;21(5):597-604. doi: 10.1016/j.acra.2013.12.001. Epub 2014 Jan 13.
Differentiating mild cognitive impairment (MCI) and Alzheimer Disease (AD) from healthy aging remains challenging. This study aimed to explore the cerebral structural alterations of subjects with MCI or AD as compared to healthy elderly based on the individual and collective effects of cerebral morphologic indices using univariate and multivariate analyses.
T1-weighted images (T1WIs) were retrieved from Alzheimer Disease Neuroimaging Initiative database for 116 subjects who were categorized into groups of healthy aging, MCI, and AD. Analysis of covariance (ANCOVA) and multivariate analysis of covariance (MANCOVA) were performed to explore the intergroup morphologic alterations indexed by surface area, curvature index, cortical thickness, and subjacent white matter volume with age and sex controlled as covariates, in 34 parcellated gyri regions of interest (ROIs) for both cerebral hemispheres based on the T1WI. Statistical parameters were mapped on the anatomic images to facilitate visual inspection.
Global rather than region-specific structural alterations were revealed in groups of MCI and AD relative to healthy elderly using MANCOVA. ANCOVA revealed that the cortical thickness decreased more prominently in entorhinal, temporal, and cingulate cortices and was positively correlated with patients' cognitive performance in AD group but not in MCI. The temporal lobe features marked atrophy of white matter during the disease dynamics. Significant intercorrelations were observed among the morphologic indices with univariate analysis for given ROIs.
Significant global structural alterations were identified in MCI and AD based on MANCOVA model with improved sensitivity. The intercorrelation among the morphologic indices may dampen the use of individual morphological parameter in featuring cerebral structural alterations. Decrease in cortical thickness is not reflective of the cognitive performance at the early stage of AD.
将轻度认知障碍（MCI）和阿尔茨海默病（AD）与健康老龄化区分开来仍然具有挑战性。本研究旨在基于脑形态学指标的个体和集体效应，使用单变量和多变量分析，探讨与健康老年人相比，MCI或AD患者的脑结构改变。
从阿尔茨海默病神经影像倡议数据库中检索了116名受试者的T1加权图像（T1WI），这些受试者被分为健康老龄化组、MCI组和AD组。进行协方差分析（ANCOVA）和多变量协方差分析（MANCOVA），以探讨在年龄和性别作为协变量控制的情况下，基于T1WI在大脑半球的34个分区感兴趣脑回区域（ROI）中，由表面积、曲率指数、皮质厚度和下方白质体积所索引的组间形态学改变。将统计参数映射到解剖图像上，以方便视觉检查。
使用MANCOVA发现，与健康老年人相比，MCI组和AD组存在整体而非区域特异性的结构改变。ANCOVA显示，内嗅皮质、颞叶和扣带回皮质的皮质厚度下降更为明显，且在AD组中与患者的认知表现呈正相关，而在MCI组中则不然。在疾病发展过程中，颞叶白质出现明显萎缩。对给定ROI进行单变量分析时，形态学指标之间存在显著的相互关联。
基于MANCOVA模型在MCI和AD中识别出了显著的整体结构改变，且敏感性有所提高。形态学指标之间的相互关联可能会影响在描述脑结构改变时对单个形态学参数的使用。皮质厚度的降低并不能反映AD早期的认知表现。
