Relationships Between Diffusion Tensor Imaging and the Social Responsiveness Scale

Background: The social responsiveness scale (SRS) is a quantitative summary measure of key autism trait features with an emphasis on social impairments.  Recent neuroimaging studies using diffusion tensor imaging (DTI) have found microstructural differences between autistic and typically developing children and young adults. 
Objectives: The main objectives were to investigate whether there were relationships between the SRS score and DTI measures and to determine the brain locations that appear to be related to the SRS.
Methods: High resolution DTI data were collected using at 3T scanner in 37 high functioning individuals with autism and 27 control subjects.  The groups were matched for age (range: 7-33 years), handedness, total IQ, and head circumference.   SRS values were obtained for all subjects.  For most subjects, the SRS was obtained using a parental report, although in seven of the control adult subjects, the SRS was derived using a self-report.  Maps of DTI measures were computed including the fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD). State-of-the-art voxel-based analysis was performed to assess the relationships between the DTI measures and the SRS scale across all subjects.  Nonlinear warping was used for between-subject image registration to achieve optimal spatial correspondence.  A general linear model with both age and SRS as independent variables was used for statistical testing.  A corrected p < 0.05 threshold was used to report regions of significant relationships.
Results: Significant relationships were observed between the SRS and the FA, MD and RD diffusion measures.  FA appeared to negatively correlated to SRS in the genu and body of the corpus callosum, left superior temporal gyrus, bilateral superior longitudinal fasciculus, posterior cingulum bundles, and right prefrontal white matter.  Both MD and RD appeared to be positively correlated to SRS in similar regions plus bilateral thalamus and temporal stem regions.   The RD measure showed the most significant correlations.
Conclusions: This study suggests a relationship between a quantitative measure of autism traits (the SRS) and tissue microstructure as assessed with DTI.  These effects were observed in diffuse white matter regions, suggesting that there may be a relationship between white matter abnormalities and autism traits.  The specific mechanisms of the DTI changes in white matter associated with autism remain unclear.
Sponsors: The project described was supported by Grant RO1s MH080826 and MH62015 from the National Institute Of Mental Health.  The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute Of Mental Health or the National Institutes of Health. Past data collection was supported in part by NICHD/NIDCD U19 HD035476, part of The Collaborative Program of Excellence in Autism.