Note: Most Internet Explorer 8 users encounter issues playing the presentation videos. Please update your browser or use a different one if available.

Association of Early Generalized Overgrowth to Clinical Outcome in ASD

Saturday, 4 May 2013: 09:00-13:00
Banquet Hall (Kursaal Centre)
D. J. Campbell1, J. Chang2 and K. Chawarska1, (1)Child Study Center, Yale University School of Medicine, New Haven, CT, (2)Statistics, Yale University, New Haven, CT
Background: Accelerated rate of head circumference (HC) growth in infancy is a well-replicated but poorly understood phenomenon in autism spectrum disorders (ASD) (Courchesne et al, 2003; Fukumoto et al, 2008). A recent comprehensive study has shown that accelerated HC growth in boys with autism was accompanied by rapid overgrowth with regard to height and weight, representing a generalized overgrowth (Chawarska et al, 2011).

Objectives:  In this study we replicate and extend our work on generalized overgrowth by examining the phenomenon of overgrowth in a larger sample of both boys and girls with ASD, as well as by examining the predictive associations between clinical features at the age of three and body size at birth as well as the rate of growth from birth to 24 months.

Methods:  HC, height, and weight measurements were collected retrospectively from 347 children with autism (AUT, N=139), PDD-NOS (PDD, N=61), or typical development (TD, N=147). Diagnostic groups and individual growth curves for HC, height, and weight were modeled using spline curves, with gender and gestational age as covariates.  Principal components analysis (PCA) was applied to fitted HC, height, and weight curves, yielding PC curves as functions of age (see Chawarska et al, 2011).  Features of PC curves were included in multivariate linear regression models to predict ADOS severity score and Verbal and Nonverbal DQ based on Mullen Scales at 36-48 months within subjects with ASD.

Results:  87% of the variance in the three morphological measures was explained by two principal components: generalized body overgrowth (PC1), and large head relative to body size (PC2).  Increased PC1 at birth was associated with increased autism severity score (regression coefficient β=0.35, p=0.027), lower Verbal DQ (β =-5.15, p=0.028) and lower Nonverbal DQ (β =-4.14, p=0.008) at three years of age.  Increased rate of growth in PC1 between birth and 24 months was associated with lower Verbal DQ (β =-4.24, p=0.009) and lower Nonverbal DQ (β =-2.69, p=0.012) at three years.  PC2 at birth was negatively associated with autism severity score (β =-0.78, p=0.025), but neither PC2 at birth or rate of growth in PC2 had any significant effects on Verbal or Nonverbal DQ.  Interaction effects between overgrowth features and gender were not significant, indicating that the relationships between overgrowth and phenotypic outcome were similar for boys and girls.

Conclusions:   Larger overall body size at birth predicted increased autism severity symptoms and decreased verbal and nonverbal skills at age three.  In addition, presence of an accelerated rate of growth of overall body size between birth and 24 months led to an additional decrease in both Verbal and Nonverbal DQ by the age of three years, above and beyond the effects seen at birth. These findings suggest that features of morphological growth may constitute useful biological predictors of developmental outcome in young children with ASD.  However, the mechanisms underlying the relationship between overgrowth and clinical presentation remain to be investigated.

| More