Molecular Autism (Jul 2021)

Maternal steroid levels and the autistic traits of the mother and infant

  • A. Tsompanidis,
  • E. Aydin,
  • E. Padaigaitė,
  • G. Richards,
  • C. Allison,
  • G. Hackett,
  • T. Austin,
  • R. Holt,
  • S. Baron-Cohen

DOI
https://doi.org/10.1186/s13229-021-00453-7
Journal volume & issue
Vol. 12, no. 1
pp. 1 – 14

Abstract

Read online

Abstract Background Prenatal sex steroids have been associated with autism in several clinical and epidemiological studies. It is unclear how this relates to the autistic traits of the mother and how early this can be detected during pregnancy and postnatal development. Methods Maternal serum was collected from pregnant women (n = 122) before or during their first ultrasound appointment [mean = 12.7 (SD = 0.7) weeks]. Concentrations of the following were measured via immunoassays: testosterone, estradiol, dehydroepiandrosterone sulphate, progesterone; and sex hormone-binding globulin which was used to compute the free fractions of estradiol (FEI) and testosterone (FTI). Standardised human choriogonadotropin (hCG) and pregnancy-associated plasma protein A (PAPP-A) values were obtained from clinical records corresponding to the same serum samples. Mothers completed the Autism Spectrum Quotient (AQ) and for their infants, the Quantitative Checklist for Autism in Toddlers (Q-CHAT) when the infants were between 18 and 20 months old. Results FEI was positively associated with maternal autistic traits in univariate (n = 108, Pearson’s r = 0.22, p = 0.019) and multiple regression models (semipartial r = 0.19, p = 0.048) controlling for maternal age and a diagnosis of PCOS. Maternal estradiol levels significantly interacted with fetal sex in predicting infant Q-CHAT scores, with a positive relationship in males but not females (n = 100, interaction term: semipartial r = 0.23, p = 0.036) after controlling for maternal AQ and other covariates. The opposite was found for standardised hCG values and Q-CHAT scores, with a positive association in females but not in males (n = 151, interaction term: r = −0.25, p = 0.005). Limitations Sample size of this cohort was small, with potential ascertainment bias given elective recruitment. Clinical covariates were controlled in multiple regression models, but additional research is needed to confirm the statistically significant findings in larger cohorts. Conclusion Maternal steroid factors during pregnancy are associated with autistic traits in mothers and their infants.

Keywords