White and Grey Matter Abnormalities in Autism Spectrum Disorder Associate with Verbal Performance

Background:  Recent diffusion tensor imaging (DTI) studies point to impairments in most or all major neural networks rather than isolated, discrete brain regions in ASD (McAlonan et al., 2005; Mueller, 2007; Shukla, Keehn, & Mueller, 2011). Several brain regions have been implicated consistently in ASD but results often differ on whether there are increased or decreased brain volumes and whether differences can be found at all in certain regions. The discrepancies may result from differences in imaging protocols, data analysis, subject ages, and samples sizes, but the overall inhomogeneity of ASD samples, which disregard neurological and diagnostic subtypes, may play a bigger role than generally assumed.

Objectives:  ASD subjects may vary in the extent of verbal impairment, and the potential anatomical correlates of this are a concern of this paper along with the general profile of white and gray matter anomalies associates with ASD.

Methods:  Twenty participants (ages 12 – 20 years 5 female, 15 male) with a previous diagnosis of an autism spectrum disorder and 10 typically developing (TD) participants (ages 12 – 18 years, 1 female, 9 male) were enrolled in our study. This whole-brain structural magnetic resonance imaging (MRI) investigation used voxel-based morphometry to analyze white and grey matter concentration (VBM-GM and VBM-WM) and tensor based morphometry (TBM) to analyze shape differences. Diffusion tensors were computed from the registered diffusion-weighted images using a nonlinear procedure and tensors were converted into fractional anistropy (FA), mean diffusivity (MD), and radial diffusivity (RD) values (method DTI-FA, DTI-MD, DTI-RD).

Results:  For adolescents with ASD, VBM and TBM revealed regions of lower GMC alongside higher WMC in the left and right precentral gyrus, lower GMC with decreased WMC in the left anterior cingulate, as well as lower GMC in the left operculum, the left superior frontal gyrus, the right and left superior temporal gyrus, and WM shape differences in the left frontal lobe and the right and left claustrum. TBM also indicated significant expansion in areas that are in overprojection with the uncinate fasciculus. When subjects were defined by a measure of receptive verbal ability (PPVT-III), we found increased GMC in the parahippocampal gyrus, along with an increased WMC and increased FA in the adjacent WM of the temporal lobe. Increased WM density, increased FA, and decreased MD and RD were also found in the lower scoring subjects in frontal WM of both hemispheres, along the left and right arcuate fascicle and the uncinate fascicle.

Conclusions:  Our findings of significant differences in white and grey matter concentration and shape largely support prior research. We found even more pronounced differences when we used verbal ability as an independent variable, which also revealed FA and RD differences widely described in previous DTI studies.