Age Related Differences in White Matter Diffusion Measures in Autism Spectrum Disorder

Background: It is well established that autism spectrum disorder (ASD) is accompanied by developmental differences in brain anatomy and connectivity.  However, the neural systems underlying ASD are complex and remain poorly understood. White matter differences in ASD have been widely studied with Diffusion Tensor Imaging (DTI) but results are heterogeneous and vary across the age range of study participants. In order to characterize the neurodevelopmental trajectory of white-matter maturation, it is thus necessary to examine a broader age range of individuals with ASD and typically developing controls, and to dissect potential age x group interactions. 

Objectives: The objectives of the present study were (1) to investigate the neurodevelopmental trajectory of white-matter maturation in a broad age range of individuals with ASD and typically developing controls, and (2) to examine age x group interactions. 

Methods: We employed a spatially unbiased probabilistic Tract-Based Spatial Statistics (TBSS) approach and a region-of-interest analysis to examine age-related differences in white-matter connectivity in a sample of 41 individuals with ASD, and 41 matched controls between 7-17 years of age. DTI data was acquired for all participants, and pre-processed using FSL software.

Results: We found significant age-related difference between the ASD and control group in a spatially distributed network of brain regions using the voxel-wise approach, and the region-of-interest analysis. In these regions, measures of fractional anisotropy (FA) significantly increased with age in both groups.  However, the age-related increase in FA was significantly larger within the ASD group relative to controls. Furthermore, we found that measures of radial diffusivity (RD) significantly decreased with age in both groups, and that the decrease in RD was stronger in the ASD group relative to controls. We also investigated between-group differences in lateralization of FA, and report FA values in the post-central gyrus to be significantly more left lateralized in ASD. Last, we examined the relationship between DTI measures and symptom severities in the ASD group. Here we found significant correlations between FA values and restricted and repetitive behaviour as measured by the ADOS in several large-scale white-matter fiber tracts in the brain. 

Conclusions: Taken together, our findings suggest that individuals with ASD have an atypical trajectory of white matter maturation relative to controls. Future histological validation and longitudinal analyses are required to further characterize the extent, time course and aetiology of these differences.