Maternal Immune-Mediated Conditions in Association with Child Immune-Related Outcomes and Autism Spectrum Disorders

Thursday, May 17, 2012
Sheraton Hall (Sheraton Centre Toronto)
2:00 PM
K. Lyall1, P. Ashwood1, J. Van de Water1,2 and I. Hertz-Picciotto1,2, (1)University of California, Davis, MIND Institute, Sacramento, CA, (2)University of California, Davis, Davis, CA
Background: Prior work has suggested that aberrations in the maternal immune system, from asthma/allergies to autoantibodies, have the ability to affect risk of autism. However, whether maternal immune responses promote immune-related phenotypes within autism has not been explored in detail.

Objectives: We sought to determine whether maternal autoimmune disease, asthma, and allergies influenced whether there were immune-related subphenotypes (specifically, gastrointestinal diagnoses, asthma, and allergies) in the child with autism, and whether these maternal immune-mediated conditions affected child scores on cognitive and behavioral tests. 

Methods: Participants were members of the CHildhood Autism Risks from Genetics and the Environment (CHARGE) study, a large population-based case-control study. The primary study group included typically developing controls and confirmed cases of autism spectrum disorder (ASD) according to ADI-R and ADOS. We compared basic frequencies of child asthma, allergy, and GI problems according to maternal immune conditions, and used logistic regression to obtain crude and multivariate adjusted odds ratios for these associations overall and by case status. We compared differences in child scores on the Mullen Scales of Early Learning (MSL) and the Aberrant Behavior Checklist (ABC) according to maternal immune-related conditions using multivariate linear regression. 

Results: 553 children with ASD, 377 typically developing children  and 157 children with developmental delay were included in these analyses. Although we did not see significant associations with maternal-immune mediated conditions and overall risk of ASD in this study, other differences according to these maternal conditions were found. Maternal autoimmune disorders significantly increased risk of child GI diagnosis in cases but not controls (adjusted OR within case children=3.21, 95% confidence interval 1.65, 6.28). This association was not seen in a secondary analysis assessing GI diagnosis according to maternal autoimmune disease in developmentally delayed children (a group with high prevalence of GI diagnosis), suggesting specificity of the association to ASD. Maternal asthma and allergies did increase risk of child asthma and allergies overall, but risk did not differ by case status. With regard to child cognitive and behavioral scores, in adjusted analyses, maternal autoimmune diseases were associated with higher scores on the inappropriate speech subscale of the ABC in analyses combining cases and controls (p=.01), and in analyses of case children only (p=.03), but not in control children only. Maternal asthma was associated with higher scores on the hyperactivity subscale (p=0.01) in analyes of all children and in children with ASD. Other comparisons between maternal conditions and child scores were non-significant after adjustment for child year of birth, regional area, child sex, and maternal age.

Conclusions: Our results suggest that maternal immune-mediated conditions may account for some of the phenotypic variability within ASD, and point to the importance of the maternal immune response in affecting neurodevelopmental outcomes. In particular, case children whose mothers have an autoimmune disease may be at greater risk for GI diagnosis relative to those children whose mothers do not have such conditions.

| More