Prenatal and Neonatal Thyroid Stimulating Hormone Levels and Autism Spectrum Disorder

Background: Thyroid hormones are critical for normal brain development.  Intrauterine thyroid dysfunction has been linked to neurologic deficits and has been hypothesized to contribute to autism spectrum disorders (ASD).

Objectives: We investigated the association between ASD and levels of thyroid stimulating hormone (TSH) measured in maternal serum from mid-pregnancy and in infant dried bloodspots from the newborn period.

Methods: The study sample was drawn from the Early Markers for Autism (EMA) Study, a population-based case-control study designed to evaluate early biomarkers for ASD.  Cases were children with ASD (n=84), identified from the California Department of Developmental Services.  Controls (n=160) were randomly sampled from birth certificate files and frequency matched to cases by gender, birth month, and birth year.  Maternal serum specimens from 15-19 weeks gestation were retrieved from the California Department of Public Health prenatal screening specimen archives. TSH was measured in maternal serum using immunoradiometric assay (IRMA). TSH levels were obtained from routine newborn screening by state regional labs, using solid-phase, time-resolved fluoroimmunoassay (AutoDELFIA). Multivariable logistic regression was used to estimate adjusted odds ratios (AORs) of ASD associated with a one log_10- unit increase (1 μIU/ml) in maternal and infant TSH levels. 

Results: Maternal and neonatal levels of TSH were marginally lower for children with ASD compared to controls [maternal AOR = 0.65, 95% CI: 0.42-1.00; neonatal AOR = 0.72, 95% CI: 0.42-1.25], after adjustment for maternal age, race/ethnicity, place of birth, maternal weight at blood draw, and gestational age at blood draw in the maternal model, and for maternal age, race/ethnicity, place of birth, infant age at blood draw, gestational age at birth in the newborn model. Looking at subgroups among cases, no differences were observed in maternal and neonatal levels of TSH between ASD cases with or without mental retardation, or ASD with regression, compared to controls.  However, maternal TSH levels were significantly lower for children with early onset ASD compared to controls [AOR = 0.59, 95% CI (0.37, 0.96)].

Conclusions: Future studies with larger sample sizes are needed to confirm these findings and to identify factors that might contribute to altered hormonal status during pregnancy.