Microstructural Alterations In Alzheimer's Disease And Mild Cognitive Impairment

Purpose To investigate microstructural white matter and gray matter changes in patients with mild cognitive impairment(MCI)and Alzheimer's disease(AD)by using neurite orientation dispersion and density imaging(NODDI)and to explore the relationship between microstructural changes and cognitive.Materials and Methods Fourteen AD patients,fourteen MCI patients,and fourteen normal controls(NC)were collected for head MRI scanning using diffusion and T1-MPRAGE sequence.DTI-derived parameter FA and NODDI-derived parameters neurite density index(NDI),orientation dispersion index(ODI),and volume fraction of isotropic water molecules(Viso),were calculated.1.The study on the microstructural white matter alterations by NODDI The Tract-Based Spatial Statistics(TBSS)method was used to analyse the differencs among the three groups.The threshold-free cluster enhancement(TFCE)method was used for multiple comparison correction.The parameter values were extracted and the correlations between the parameter values in the fibers and Minimental state examination(MMSE)and Montreal cognitive assessment(Mo CA)scores were examined.A receiver operating characteristic(ROC)curve was conducted to assess the diagnostic performance of different parameters of fibers.2.The study on the microstructural gray matter alterations by NODDI The bilateral anterior cingulate gyrus,bilateral posterior cingulate gyrus,bilateral hippocampus,bilateral parahippocampal gyrus,bilateral amygdala,bilateral caudate nucleus,bilateral putamen,bilateral globus pallidus,bilateral thalamus,bilateral frontal cortex,bilateral occipital cortex,bilateral temporal cortex,and bilateral parietal cortex were selected as region of interests(ROIs).The ROIs were made into templates.The NDI,ODI,and Viso maps were co-registered onto Montreal Neurological Institute(MNI)space and the parameter values were extracted.One-way ANOVA was performed for statistical analysis among the three groups.The Bonferroni test was used as a post hoc analysis.p?0.016 was set as the significance threshold.The correlations between ROIs parameter values and MMSE and Mo CA scores were performed.Results 1.The results of microstructural white matter changes were as follows:(1)The results of TBSS: For the FA values,there was no significant difference between the NC and MCI groups at the significant threshold of p<0.01.However,NDI value and ODI value in the MCI group significantly decreased in multiple fibers(p<0.01,TFCE-corrected).The Viso values increased significantly in multiple fibers in the MCI group(p<0.01,TFCE-corrected).Compared with the MCI group,the FA value,NDI value and ODI value in the AD group decreased significantly and the Viso value increased significantly in multiple white matter areas(p<0.01,TFCE correction).The differences in FA,NDI,ODI,and Viso were greater between the NC and AD groups than between the NC and MCI groups and between the MCI and AD groups.NDI showed more significant differences in the MCI and AD groups than FA in the same slices.(2)The results of correlation: The NDI value of the splenium of the corpus callosum(r=0.633,p<0.001),the ODI value of the left superior longitudinal fasciculus(r=0.649,p<0.001),and the Viso value of the right cerebral peduncle(r=-0.508,p=0.001)showed the strongest correlations with MMSE scores.The NDI value of the splenium of the corpus callosum(r=0.647,p<0.001),the ODI value of the left superior longitudinal fasciculus(r=0.653,p<0.001),and the Viso value of the left cingulum(r=-0.505,p=0.001)showed the strongest correlations with the Mo CA scores.(3)The results of ROC analysis: For the diagnosis of MCI,there were no significant differences in the areas under the ROC curves(AUC)between FA and NODDI parameters that had been most strongly correlated with MMSE and Mo CA scores.For the diagnosis of AD,the AUC for the NDI value of the splenium of the corpus callosum was larger than the corresponding FA value(AUC=0.885,0.714,p=0.042).The AUC for the Viso value of the right cerebral peduncle was larger than the corresponding FA value(AUC=0.934,0.531,p=0.004).The AUC for the left superior longitudinal fasciculus(AUC=0.918,0.816,p=0.282)and left cingulum(AUC=0.837,0.740,p=0.482)showed no significant difference between FA and NODDI parameters.2.The results of microstructural gray matter changes were as follows:(1)The results of One-way ANOVA: Compared with the NC group,the NDI values of the left temporal cortex and the ODI values of multiple ROIs decreased significantly in the MCI group.In the comparison between NC and AD groups,the NDI and ODI values decreased significantly and Viso values increased significantly in multiple ROIs in the AD group.The ROIs with significant differences in ODI values were more than that in NDI values.Compared with the MCI group,the Viso values of multiple ROIs increased significantly in the AD group.(2)The results of correlation: The NDI value of the right posterior cingulate gyrus(r=0.497,p=0.001),the ODI value of the left parietal cortex(r=0.542,p<0.001),and the Viso value of the left temporal cortex(r=-0.712,p<0.001)showed the strongest correlations with MMSE scores.The NDI value of the right posterior cingulate(r=0.520,p<0.001),the ODI value of the left parietal cortex(r=0.569,p<0.001),and the Viso value of the right hippocampus(r=-0.661,p<0.001)showed the strongest correlations with the Mo CA scores.Conclusion (1)It was found that the microstructural white matter changes in AD and MCI patients were the decrease of neurites density and fibers orientation dispersion.NODDI could more accurately evaluate the microstructural white matter changes.(2)The results indicated that the microstructural gray matter changes in the AD and MCI stage were the loss of neurites and the reduction of dendritic complexity.Additionally,the atrophy of gray matter was shown in the AD stage,and may be the main changes in the progress from MCI conversion to AD.The Viso of gray matter may be a reference indicator for predicting the disease progression.(3)NODDI could effectively predict the clinical cognitive status of AD and MCI patients.NDI was much more sensitive to white matter changes than FA.ODI was much more sensitive to gray matter changes.NODDI may be superior to DTI for the diagnosis of AD.