Linear associations between clinically assessed upper motor neuron disease and diffusion tensor imaging metrics in amyotrophic lateral sclerosis

• Published in: PloS one 2014-08, Vol.9 (8), p.e105753-e105753
• Main Authors: Woo, John H, Wang, Sumei, Melhem, Elias R, Gee, James C, Cucchiara, Andrew, McCluskey, Leo, Elman, Lauren
• Format: Article
• Language: English
• Subjects: Adult; Age Factors; Aged; Aged, 80 and over; Amyotrophic lateral sclerosis; Amyotrophic Lateral Sclerosis - pathology; Amyotrophic Lateral Sclerosis - physiopathology; Anisotropy; Biology and Life Sciences; Brain; Computer and Information Sciences; Diagnostic imaging; Diffusion; Diffusion Tensor Imaging; Female; Handedness; Humans; Magnetic resonance imaging; Male; Medical research; Medicine; Medicine and Health Sciences; Middle Aged; Models, Biological; Motor neuron diseases; Neuroimaging; Neurons; NMR; Normalizing; Nuclear magnetic resonance; Physical Sciences; Pyramidal tracts; Pyramidal Tracts - pathology; Pyramidal Tracts - physiopathology; Registration; Regression analysis; Regression models; Research and Analysis Methods; Studies
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0105753

To assess the relationship between clinically assessed Upper Motor Neuron (UMN) disease in Amyotrophic Lateral Sclerosis (ALS) and local diffusion alterations measured in the brain corticospinal tract (CST) by a tractography-driven template-space region-of-interest (ROI) analysis of Diffusion Tensor Imaging (DTI). This cross-sectional study included 34 patients with ALS, on whom DTI was performed. Clinical measures were separately obtained including the Penn UMN Score, a summary metric based upon standard clinical methods. After normalizing all DTI data to a population-specific template, tractography was performed to determine a region-of-interest (ROI) outlining the CST, in which average Mean Diffusivity (MD) and Fractional Anisotropy (FA) were estimated. Linear regression analyses were used to investigate associations of DTI metrics (MD, FA) with clinical measures (Penn UMN Score, ALSFRS-R, duration-of-disease), along with age, sex, handedness, and El Escorial category as covariates. For MD, the regression model was significant (p = 0.02), and the only significant predictors were the Penn UMN Score (p = 0.005) and age (p = 0.03). The FA regression model was also significant (p = 0.02); the only significant predictor was the Penn UMN Score (p = 0.003). Measured by the template-space ROI method, both MD and FA were linearly associated with the Penn UMN Score, supporting the hypothesis that DTI alterations reflect UMN pathology as assessed by the clinical examination.