Callosal degeneration topographically correlated with cognitive function in amnestic mild cognitive impairment and Alzheimer's disease dementia

Abstract

Degeneration of the corpus callosum (CC) is evident in the pathogenesis of Alzheimer's disease (AD). However, the correlation of microstructural damage in the CC on the cognitive performance of patients with amnestic mild cognitive impairment (aMCI) and AD dementia is undetermined. We enrolled 26 normal controls, 24 patients with AD dementia, and 40 single-domain aMCI patients with at least grade 1 hippocampal atrophy and isolated memory impairment. Diffusion tensor imaging (DTI) with fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (DA), and radial diffusivity (DR) were measured. The entire CC was parcellated based on fiber trajectories to specific cortical Brodmann areas using a probabilistic tractography method. The relationship between the DTI measures in the subregions of the CC and cognitive performance was examined. Although the callosal degeneration in the patients with aMCI was less extended than in the patients with AD dementia, degeneration was already exhibited in several subregions of the CC at the aMCI stage. Scores of various neuropsychological tests were correlated to the severity of microstructural changes in the subregional CC connecting to functionally corresponding cortical regions. Our results confirm that CC degeneration is noticeable as early as the aMCI stage of AD and the disconnection of the CC subregional fibers to the corresponding Brodmann areas has an apparent impact on the related cognitive performance.

Keywords: Brodmann area; corpus callosum; diffusion tensor imaging; neuroimaging; probabilistic tractography.

Figure 1

Three dimensional surface reconstruction of the entire corpus callosum , cortical masks and global topography of corpus callosum. (a) The three‐dimensional surface reconstruction of the entire corpus callsoum which the primary volume of interest in current study. (b) BAs were reassigned to seven cortical target masks including frontal cortex (red, including BA8, BA9, BA10, BA11, BA44, BA45, BA46, and BA47), premotor and supplementary motor area [SMA] cortex (orange, including BA6), primary motor cortex(yellow, including BA4), primary sensory cortex (green, including BA1, BA2, BA3, and BA5), parietal cortex (dark blue, including BA7, BA39, and BA40), occipital cortex (violet, including BA17, BA18, and BA19,), and temporal cortex (light blue, including BA20, BA21, BA22,BA37, BA38, BA41, and BA42). (c) The global topography of the single subject's CC was constructed using connectivity‐based parcellation approach with hard segmentation method in sagittal view (along with x‐axis). The color of each CC subdivision corresponded to its connection profile of different cortical areas. BAs, Brodmann areas; CC, corpus callosum. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 2

Bar charts of different diffusivity indices of frontal, premotor/SMA, primary motor, primary sensory, temporal, parietal, occipital subregions of corpus callosum, and entire corpus callosum in the normal controls, amnestic mild cognitive impairment and Alzheimer's disease groups. Significant differences after Tukey's correction of post‐hoc ANCOVA test are marked with asterisks (P < 0.05). The unit of the diffusivity values (MD, DA, and DR) is 10⁻⁶ mm²/s. CC, corpus callosum; FA, fractional anisotropy; MD, mean diffusivity; DR, radial diffusivity; DA, axial diffusivity; NC, normal control; aMCI, amnesic mild cognition impairment; AD, Alzheimer disease. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 3

Partial correlation coefficient between mean diffusivity (MD) values of corpus callosum and neuropsychological tests. The Pearson partial correlation was estimated with age, sex, and former education years as the confounding covariates. The grids in black meant there were no correlation between the subregions and the neuropsychological tests.(i.e., P value more than 0.05); the white grid indicated P value less than 0.05, the light gray grid indicated P value less than 0.01, and the dark gray indicated the corrected‐P value less than 0.05 after Bonferroni correction. Abbreviation of neuropsychological test; MMSE, Mini‐Mental Screening Examination; CVVLT, Chinese version of the Verbal Learning Test; CFT, Complex Figure Test, BNT, Boston Naming Test; TMT‐B, Trail‐making Test Part B. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]