Head Circumference Across the First Year of Life Correlates with a Positive Screen for Autism at Age 36 Months

Background: Abnormal head size and growth patterns in early life are associated with a later diagnosis of autism or other neurodevelopmental disorders in some, but not all, studies. Accelerated growth of head circumference (HC), beginning after birth and plateauing before three years of age, is reported to be more common in autistic disorder, particularly in males, or in genetic overgrowth syndromes; however, microcephaly is also relatively frequent in subsets with other dysmorphic features, mental retardation and less male-predominant sex ratios. Although micro- and macrocephaly at birth are well-established harbingers of poor central nervous system outcomes, the clinical significance of abnormal head growth trajectories is less clear.

Objectives: Determine whether accelerated HC growth across the first year of life predicts screen-positivity at 36 months of age on the Social Communication Questionnaire (SCQ), an autism screening tool. Methods: The Autism Birth Cohort (ABC) is a subsample of a prospective pregnancy birth cohort in Norway (MoBa). Data for 29,093 children were available for this analysis. The SCQ was given at 36 months; scores ≥ 12 were defined as positive. The HC obtained at well-baby clinics at 6 weeks and 12 months of age were used to derive the first-year HC growth rate, defined as the difference between the two HC measures. HC growth rate was categorized as: slow (≤ 20^th percentile), normal (20^th through 80^th percentile), or accelerated (≥ 80^th percentile). Micro- and macrocephaly were defined as values ≤ or ≥ 2 SD of the ABC population mean at age 6 weeks or 12 months. We examined the relationship of early and late infancy HC abnormalities as well as first-year change in HC to SCQ-positivity using chi-square analyses.

Results: 1.1% screened positive on the SCQ. Average HC growth was 8.21 ± 1.1 cm. Microcephaly was more common in girls than boys at both measurement ages (6 weeks: 3.3 vs. 1.5%; 12 months: 2.6 vs 0.3%; both p<0.0001); macrocephaly showed a male-predominant pattern at both measurement points (6 weeks: 0.3% of girls vs 3.3% of boys; 12 months: 0.4 vs 4.8%; both p<0.0001). Neonatal microcephaly was more highly associated with SCQ-positivity at 36 months than macrocephaly in all children (p<0.0001); by 12 months of age, this pattern remained significant for boys (p=0.003) with a trend in the same direction for girls (p=0.074). None of the SCQ33-positive girls were macrocephalic at either measurement age. Accelerated HC growth predicted SCQ-positive scores in all children (1.2% of accelerated growth, vs. 0.9% of normal and 0.6% of slow growth children, p=0.017).

Conclusions: Head size and HC growth may be markers for autism risk. Pediatricians should carefully record values and screen at age 3 when indicated. Delineation of the mechanisms underlying abnormal head growth trajectories may also improve understanding of the pathogenesis of autism and related neurodevelopmental disorders.

Acknowledgments: Members of the ABC Study group include W. Ian Lipkin, Camilla Stoltenberg, Ezra Susser, Per Magnus, Deborah Hirtz, Mady Hornig, Michaeline Bresnahan, Synnve Schølberg, Ted Reichborn-Kjennerud and Pål Surén.