White Matter Abnormalities in Autism Spectrum Disorders: Evidence of Abnormal Neural Connectivity

Background: Advances in anatomical and functional imaging techniques studying brain-behavior relationships, and the application of these technologies to the study of autism spectrum disorders (ASDs), has resulted in substantial evidence attributing both core and secondary symptoms to abnormalities in brain connectivity.  There is preliminary evidence from structural MRI, functional MRI, and diffusion tensor imaging suggesting that the brain phenotype in ASD includes a generalized overabundance of short-range connections (i.e. U-fibers connecting adjacent gyri) with deficiencies of long-range connections (i.e. association fibers connecting different lobes).

Objectives: This study examined structural abnormalities in white matter using structural MRI and diffusion tensor imaging to test the hypothesis that ASDs are characterized by generalized overabundance of short-range connections with deficiencies of long-range connections.
Methods: Subjects included 15 boys with ASD (mean age = 10.5 ±3.7 years) and eight gender- and age-matched controls (mean age = 11.5 ±2.8 years).  High-resolution structural MRI and diffusion imaging (directions = 30 and b0 = 5) were obtained using a Siemens 3-Tesla Trio scanner.  Structural MRI data was processed and analyzed using the FreeSurfer image analysis suite.  FreeSurfer consists of automated tools for reconstruction of the brain from structural MRI data, facilitating the quantification of regional white matter volumes.  Cortical and central white matter volumes were taken as representing short-range and long-range connections, respectively.  Volumes were entered into a statistical program and comparisons of volumes between groups were be conducted using Student’s t test with significance level will be set at p < 0.05 (two-tailed).  FMRIB Software Library was used to process and analyze diffusion data.  Fractional anisotropy was chosen as the primary measure of the structural integrity of axonal fiber tracts.  Voxel-wise analysis of multi-subject diffusion data was conducted using Tract-Based Spatial Statistics.  Areas of significant difference were computed using Threshold-Free Cluster Enhancement and displayed as p-value images, where p < 0.05 corrected for multiple comparisons across space.
Results: There were no significant group differences in age and intracranial volume.  A significant decrease in central white matter volume was observed in the ASD group (p = 0.02).  There was neither a significant difference in cortical white matter volumes nor total cerebral white matter volumes.  The ASD group had significant bilateral reductions in fractional anisotropy involving numerous association, commissural, and projection tracts.   Affected association tracts included the inferior longitudinal fasciculus, superior longitudinal fasciculus, inferior fronto-occipital fasciculus, uncinate fasciculus, cingulum, and fornix.  Commissural fibers included the corpus callosum (genu, body, and splenium), and both forceps major and minor.  Projection tracts included the anterior thalamic radiation and corticospinal tract.  There were no areas of increased fractional anisotropy in the ASD group.
Conclusions: This study provides preliminary evidence of both reduced central white matter volume and reduced fractional anisotropy along numerous long-range fiber tracts in ASDs, suggesting the existence of reduced long-range connectivity.  The diffuse distribution of these findings suggests a global abnormality with long-range connections which may not only contribute to both core and associated symptoms, but also the well-known heterogeneity of this spectrum of disorders.