Longitudinal Head Circumference and Accelerated Brain Growth in Infants at Risk for Autism

Background: There has been a range of findings on head circumference in populations with autism spectrum disorders (ASD).  Independent studies have identified increased head circumference, no difference in head circumference, and even smaller head circumference in ASD samples.  Difficulties in understanding the discrepancy of these results come from the heterogeneity of samples (e.g., severely impaired vs. higher functioning), the nature of the comparison group (e.g., normative data vs. local controls), and the paucity of true longitudinal studies.

Objectives: To characterize head circumference trajectories from 6 to 24 months in a large longitudinal sample of infant siblings at high familial risk for ASD.

Methods: Head circumference data was obtained from The Infant Brain Imaging Study (IBIS).  The IBIS Network is an NIH-funded Autism Center of Excellence following a cohort of high risk (HR) and low risk infants from 6 months of age.  Participants received developmental and behavioral assessments and MRI scans at 6, 12, and 24 months.  Classification for ASD occurred at the 24 month visit, using DSM-IV and ADOS diagnostic criteria to determine which HR infants met criteria for ASD.  Low risk infants were negative for ASD and had no first degree relatives with an ASD.  Three groups were examined for our analyses:  36 HR-ASD, 150 HR-non ASD, and 87 low risk infants.  Automated measures of head circumference were obtained directly from the MRI scans.  Data were analyzed using longitudinal random coefficient growth curve models controlling for sex, length, and cognitive developmental level (from the Mullen).   

Results: We observed a trend for group differences in the trajectory of head circumference from 6 to 24 months, although this was only marginally significant (p = .05).   Cross-sectional comparisons found that the HR-ASD group had significantly increased head circumference at 24 months (p=.04), but did not look different from the other two groups at 6 or 12 months.  However, brain overgrowth was evident in this group, with the HR-ASD group showing a significantly accelerated rate of growth (e.g., intracranial volume, p = .02). 

Conclusions:  Increased brain volume and head circumference was present by 24 months in HR-ASD infants, but not at 6 or 12 months of age, suggesting an underlying postnatal mechanism involving brain overgrowth.  These findings indicate that in a high risk population, increased head circumference may not be observed before age 2, but that accelerated rates of brain volume growth can be detected in infants later identified with ASD.  Interestingly, these trajectories appear to be aligned with the emergence of clinical symptoms.  Since head circumference is only a proxy for brain volume, neuroimaging studies focused on understanding the underlying neurobiology of autism may provide the best information about possible mechanisms of early brain overgrowth.