The voxel-based morphometry study of brain gray matter changes in amyotrophic lateral sclerosis

Objective: To evaluate the gray matter alterations in amyotrophic lateral sclerosis (ALS) by voxel-based morphometry (VBM), and further analyze the correlation between volume changes of gray matter and clinical characteristics. Methods: Twenty-seven non-demented patients with ALS and 27 age and gender matched healthy controls were recruited. FSL-VBM was used to detect whole brain gray matter differences between the two groups. Seven prior ROIs were set to be analyzed, including bilateral precentral gyrus, postcentral gyrus, superior, medial, inferior, middle frontal gyrus, and insular cortex. The mean gray matter density of the ROIs was extracted in order to explore the correlation with several clinical measurements such as disease durations and disease severity scores, by using partial correlation analysis with age as covariates. Results: Whole brain analysis showed significant gray matter loss in the left precentral gyrus, superior frontal gyrus and postcentral gyrus (numbers of voxel in clusters were 338, 112, 127, Z = 4.83, 4.09, 6.42, P < 0.05, FWE corrected). A prior seven ROIs analysis detected gray matter loss in the left precental gyrus, right precentral gyrus, left postcentral gyrus, superior frontal gyrus and left insular cortex (numbers of voxel in clusters were 1104, 34, 114, 91, 107, Z = 5.87, 3.71, 4.26, 6.29 and 3.51, P < 0.05, FWE corrected). No statistical significant correlation between regional gray matter loss and clinical measurements were found. Conclusions: Motor and extra-motor gray matter loss are present among patients with ALS, which demonstrates ALS as a multi-system disorder. In contrast to whole brain gray matter analysis, ROI analysis is more sensitive to detect extensive cortical changes. The heterogeneity of disease and sensitivity of method may contribute to the lack of correlation between gray matter volume decrease revealed by VBM and clinical characteristics. (authors)