An Examination of Head Circumference In Autism Spectrum Disorders

Background: Increased head circumference (HC) is often noted in children and adults with autism spectrum disorder (ASD). Occipitofrontal HC is a proxy for brain volume and brain overgrowth appears to represent both greater brain tissue volume and greater lateral ventricle volume (Piven, et al., 1995). Despite this well-replicated finding of increased head size and brain volume in ASD, it is unclear whether macrocephaly defines a specific subgroup of ASD clinical features. Limited research available suggests that head size is also increased in family members of individuals with ASD. Only one study has examined the relation between HC and broader autism phenotype (BAP) traits in relatives of individuals with ASD (Elder, et al, 2007). 

Objectives: The first objective of this study is to examine the relationship between HC and behavioral symptoms in children with ASD. Our second objective is to explore HC and BAP traits in family members within simplex families.

Methods: The sample consisted of mothers, fathers, and probands (1,814 males, 281 females) from 2,095 families participating in the Simons Simplex Collection (SSC) project (data release 10; SFARI.org). Data from 1,979 siblings were included as well (935 males, 1044 females). Families were carefully screened for a family history of ASD. Probands met clinical cutoffs on the ADI and ADOS and received a clinical diagnosis of ASD. HC and height were measured during the family’s research visit. Proband behavioral measures included the ADOS severity score, Social Responsiveness Scale: Parent and Teacher Forms, and an assessment of IQ. To assess the BAP, the Broader Autism Phenotype Questionnaire (BAPQ) was used in parents and the SRS was used in parents and siblings.

Results: Multiple linear regressions were conducted to analyze the relationship between HC and BAP/ASD symptoms. Bushby and colleagues (1992) reported significant effects of height on HC. We found a moderate relationship between height and HC in our sample as well (r=.31). Therefore, height z score was entered into the regression equation to control for its effects on HC. Z scores for HC were computed using the normative sample reported in Roche, et al, 1987 while z scores for height were computed using the CDC norms published in 2000. HC significantly predicted multiple outcomes in probands across modalities and informants. Specifically, HC predicted SRS: Parent (ß=.06, p<.05) and Teacher (ß=.08, p<.05), ADOS severity score (ß=.07, p<.01), VIQ (ß=-.07, p<.01), NVIQ (ß=-.05, p< .05), and FSIQ (ß=-.05, p<.05). HC was not predictive of BAPQ or SRS scores in parents and siblings.

Conclusions: Larger HC was generally maladaptive for probands in our sample across parent, teacher, and clinician ratings and assessments. Higher SRS scores, greater symptom severity on ADOS, and lower IQ scores were associated with larger HC. This finding is consistent with previous research in younger children showing that increased rate of growth in the first year of life (resulting in larger HC) was associated with greater symptom severity (Courschesne, et al, 2003). HC was not found to be related to the BAP in parents and siblings.