Amygdala Growth Trajectories, Fear Potentiated Startle Response, and Anxiety in Children with Autism Spectrum Disorder

Background: Symptoms of anxiety are reported commonly in ASD, with approximately 40-50% of individuals with ASD meeting criteria for an anxiety disorder (Kerns & Kendall, 2012). Alterations in amygdala morphology and activity are also commonly reported in ASD, yet the relationship between alterations in the amygdala and the development of co-morbid anxiety disorders is unclear. Given that the amygdala is functionally related to fear and anxiety (Amaral et al., 2003), investigating amygdala growth trajectories and probing amygdala function using a fear potentiated startle paradigm may provide insight into autism-related anxiety disorders that emerge during middle childhood.

Objectives: To examine amygdala growth trajectories, fear potentiated startle response and clinically significant anxiety in children with ASD and age-matched typically developing (TD) children.

Methods:  Participants to date include 22 children with ASD (15 male/7 female; IQ range 32-138; ADOS-2 composite score range 3-10) and 14 typically developing children (10 male/4 female; IQ range 98-140). The children are currently between 9 and 13 years of age and enrolled in the ongoing Autism Phenome Project. T1-weighted structural MRIs were collected for each child at around 2-3 years-of-age and again at present. Amygdala volumes were measured bilaterally at each age based on these scans. The rate of amygdala growth was calculated as the change in volume over time, normalized to total brain volume growth. Fear potentiated startle response was measured as a probe of amygdala function. The fear potentiated startle response was evaluated using a paradigm previously tested in adolescents with ASD (Sterling et al., 2013) that was adapted for testing with children with low IQ and limited verbal abilities. Clinically significant anxiety was measured using the Anxiety Disorders Interview Schedule (ADIS-IV) with the Autism Spectrum Addendum.

Results:  In the fear potentiated startle task, children with ASD had significantly reduced mean percent potentiation compared to TD children. Interestingly, the ASD group also demonstrated significantly increased baseline startle magnitudes. Within the ASD group, the rate of amygdala growth was negatively correlated with percent fear potentiation such that children with the fastest amygdala growth rates demonstrated the most reduction in percent fear potentiation. In addition, percent fear potentiation was also significantly correlated with the clinical severity rating scores for each child’s principal anxiety diagnosis. Children with ASD and the highest ratings of anxiety demonstrated the greatest level of fear potentiation.

Conclusions: A significant association between amygdala growth and fear potentiated startle response suggest that alterations in the developmental trajectory of the amygdala may be driving the reduced fear response in the ASD group. At the same time, greater fear potentiated startle response was related to higher clinical ratings of anxiety, indicating that this probe of amygdala function may be a useful tool for evaluating anxiety level particularly in non-verbal lower functioning individuals.