Abnormal Brain Surface Morphology and Gyrification Pattern in Children with Autism Spectrum Provides Clues to Prenatal Onset

Background:  
Cortical complexity can be described by the pial surface and gyrification, a process which begins prior to birth. In normal development, the period critical for the emergence of cortical convolutions is between 26 to 36 weeks of gestational age, ie the third trimester of pregnancy (Chi, Dooling, & Gilles, 1977); during this time the cortical surface area increases considerably, resulting in an increase of grey matter volume. Earlier investigations indicated greater folding in frontal regions in autism (Harden et al., 2004), whereas others have reported sulcal abnormalities in the parietal regions (Nordahl et al., 2007). However using one dimension to understand cortical folding provides limited explanation of cortical complexity. The present study examines the developmental trajectory of cortical gyrification using four cortical dimensions in a group of children with autism spectrum disorders (ASD) and typically developing controls over a 6–17 year range.

Objectives:
The present study examined: 1) gyrification differences in ASD compared to controls across four cortical dimensions: surface area, cortical thickness, sulcal depth and curvature; and 2) age-related differences in gyrification in each group.

Methods:  
Thirty five T1-weighted MPRAGE images were acquired from 6-17 year old children with ASD (n=17) and typically developing controls (n=18). Parental consent was obtained and the study was approved by the Hong Kong West Cluster Institutional Review Board. Both groups were matched in age and all children attended mainstream schools. We employed a surface-based approach using the Freesurfer software and a general linear model was computed to derive regional differences in each of the four dimensions. A correlation analysis of age-gyrification pattern was conducted in each group separately.

Results:  
There were significant group differences in the right parietal regions. Compared to controls, the right parietal region in the autism group had smaller surface area, but greater cortical thickness, larger sulcal depth and gyral curvature. The right posterior cingulate in the autism group had smaller sulcal depth and gyral curvature despite no differences in the surface area and cortical thickness. Correlation analyses indicated right parietal surface area increased with age in ASD but decreased with age in controls whereas cortical thickness measures did not change with age in ASD, but increased in the control group.

Conclusions:
Our study examines the surface brain morphology across development in children with autism. Preliminary findings confirm cortical abnormalities occur primarily in the parietal regions and posterior cingulate in the right hemisphere. Since sucli and gyri in these regions first emerge at around 24 to 27 gestation week and the gyrification pattern remains relatively constant from postnatal into childhood (White et al., 2010), the differences we observe may have origins during this period of prenatal life. Age-related differences in our cohort suggest dysmaturation of gyrification in the parietal and cingulate regions in the autism group is dynamic and persistent.