Age-Related Differences in White Matter in Children and Adolescents with Autism

Background:

Abnormalities in white matter circuits are implicated in autism.  White matter maldevelopment may drive the dramatic changes in brain volume in young children and in older children and adults, and structural or functional MRI datasets suggest widespread disconnectivity.  The period of development from childhood into adolescence is marked by heterochronous brain changes in typically developing individuals which parallel maturation of cognitive and emotional abilities.  However, the trajectory of age-related changes in white matter from childhood into adolescence is not well known.

Objectives: To map age-related differences in white matter in children and adolescents with autism aged between 6 and 16 years old.

Methods:

T2/PD scans were acquired from children with ASD (n = 36) and typically developing controls (n = 55) group matched for age, gender, ethnicity and handedness.  Diffusion Tensor Imaging (DTI) scans were acquired from a subgroup of participants.  Voxel based morphometry (VBM) was used to map age-related changes in white matter volume in both groups.  The age-volume maps from each group were overlaid and used to define ‘regions-of-interest’ (ROI) for DTI analysis in a subgroup of children. Fractional anisotropy (FA) values and correlations between age and FA in these ROI were examined.

Results:

There was a positive correlation between age and white matter volumes in ventral frontal lobe in controls only.  There was a negative correlation between age and cerebellar white matter volume in the autism group only.  There were positive correlations between age and white matter volumes throughout parietal, occipital and dorsal frontal lobe in both groups, as between age and FA in these regions.    There was a significant positive correlation between age and FA in the ventral prefrontal lobe and left internal capsule of the control group, not the autism group.  There were no age-related changes in FA in the cerebellum of either group.  

Conclusions:

The antero-posterior gradient of brain maturation is disrupted in autism.  This study indicates dysmaturation of frontal and cerebellar pathways in autism with relative sparing of posterior hemisphere systems in autism. Given the joint roles of cerebellum and frontal lobe in cognitive and social functioning, this disordered development likely makes a major contribution to persistent symptoms in autism.