Current clinical, cognitive and neuroimaging observations converge to propose two neurodevelopmental hypotheses in autism. A first hypothesis posits impairments in the maturation of the cerebral regions concerned with social competences, but detailed examinations of the nature of these changes (i.e. cortical thickness or gyrification) are yet missing. A second hypothesis, based on post-mortem and functional connectivity studies, postulates altered cerebral connectivity in autism. The structural counterpart of this functional dysconnectivity is however largely unknown.
Objectives:
Provide whole-brain measurements of cortical thickness (an index of brain maturation), cortical gyrification (a marker of early brain development), and structural connectivity using tractography in children with autism.
Methods:
We analyzed cerebral morphometry in a sample of 11 children with autism compared to their individually age- and gender-matched controls. This group consists of 8 male and 3 female children with autism, aged 9 to 15 years old, with an average IQ of 79.4 ± 18.1. Cerebral MRI acquisitions included both anatomical T1-weighted images and Diffusion Tensor Images (DTI).
The T1-weighted images were processed using validated procedure embedded in FreeSurfer (http://surfer.nmr.mgh.harvard.edu/fswiki), yielding accurate measurements of regional cortical and white matter volumes. Cortical thickness and local Gyrification Index (lGI, Schaer et al., 2008) were measured with an exquisite resolution across the cortex. All analyses covaried out the effect of age and gender and were corrected for multiple comparisons. To relate the cortical anatomy with the underlying architecture of white matter bundles, we used tools embedded in the Human Connectome (http://www.connectomics.org/connectomemapper/).
Results:
No between-group difference in cortical thickness was found. We observed 3 clusters with significant lGI reduction in patients with autism as compared to controls, in the right inferior parietal region, the lower part of the precentral gyrus and the inferior frontal gyrus. Concomitantly with altered frontal gyrification, decreased volume was found in the anterior part of the corpus callosum. Tracking the white matter fibers from cortical areas of significant gyrification difference, we observed significant positive correlations between the length of short-range connections and lGI in patients with autism (p<0.001), but not in controls.
Conclusions:
Reduced gyrification in the inferior fronto-parietal regions lends support for early-disrupted cortical growth in the fronto-parietal mirror neuron system, implicated in action imitation (Rizzolatti and Craighero, 2004). Early impaired neurodevelopment in the mirror-neurons system may alter embodied aspects of the self-others relationship in children with autism, providing initial substrate for altered maturation cascade of the cerebral networks responsible for more sophisticated social skills, such as empathy and theory of mind. We show that these gyrification abnormalities are further related to a an imbalance of the short-to-long connectivity. Decreased volume of the anterior corpus callosum, connecting frontal hemispheres, points to reduced long-range connectivity in areas with reduced gyrification. Further, increased short-range connectivity was correlated with decreased cortical gyrification. Our results support the hypothesis that an imbalance between the short- and long-range connectivity, thought to impair high integration of information and coordination of multiple neural systems, also alters the shape of the brain of patients with autism during early stages of brain development.
See more of: Brain Imaging
See more of: Brain Structure & Function