What's the (white) Matter in Autism Spectrum Disorder

Background:   Autism Spectrum Disorder (ASD) is characterized by complex underlying neuropathology that is not fully understood. There is evidence of early dysregulated brain growth in children with ASD, which may have profound effects on neural connectivity. A large body of research has pointed to atypical functional and structural networks in ASD, with evidence for both functional underconnectivity and overconnectivity. Diffusion tensor imaging (DTI) studies have shown widespread white matter abnormalities in children with ASD, which may disrupt neural circuitry, impacting crucial cognitive processes, such as working memory. However, previous studies have been limited by sample size and comprehensive diffusion analyses. 

Objectives:   To explore white matter development in a large sample of children with and without ASD using DTI, and the relations between various DTI metrics and cognitive processes, such as working memory.

Methods:   60-direction diffusion data was acquired on a 3T Siemens Trio MRI scanner with a 12-channel head coil. Subjects included 61 children with ASD (51 male) and 69 age- and sex-matched healthy control (51 male) children (ages 7-15). Values for fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD) were calculated. Tract based spatial statistics (TBSS) was used to define subject-specific FA skeletons, which were used as masks to examine differences in FA, MD, AD and RD. FSL Randomise was used to statistically evaluate these differences. Associations (Pearson’s) between FA in select tracts and working memory, as measured by the Working Memory Test Battery, were examined in a subgroup of children with (n= 39) and without ASD (n = 34).

Results:   Compared to controls, children and adolescents with ASD showed widely distributed reduced FA and AD, with no group differences in MD or RD values.  Both groups showed age-related changes, but there was no interaction between age and group.  Regions of reduced FA included the corpus collosum, cerebral peduncle and projection fibres (corona radiata, internal capsule, posterior thalamic radiation). Similarly, regions of reduced AD included the corpus collosum, cingulum, cerebral peduncle, corona radiata, internal capsule and external capsule. In control children only, FA of the bilateral superior fronto-occipital fasciculus showed a positive association with working memory (p<0.05, uncorrected).

Conclusions:  In one of the largest samples studied with DTI and ASD, our findings highlight widespread atypical white matter in children with ASD that persists well into adolescence. Altered white matter structure was not limited to specific networks, and was observed primarily in the corpus callosum and thalamocortical fibres—tracts crucial for interhemispheric exchange and higher order information processing. Widespread white matter impairment in ASD is consistent with the view that ASD is a disorder of generalized complex information processing. The relation between patterns of white matter integrity and specific aspects of cognition in ASD remains elusive.