Maternal polycystic ovarian syndrome in autism spectrum disorder: a systematic review and meta-analysis

Mol Psychiatry. 2019 Dec;24(12):1787-1797. doi: 10.1038/s41380-019-0398-0. Epub 2019 Mar 13.

Abstract

There is evidence showing a positive correlation between prenatal androgens and their effect on the development of central nervous system and the autistic spectrum disorder (ASD) phenotype in offspring of mothers with polycystic ovary syndrome (PCOS). We applied a systematic review to investigate whether women with PCOS have increased odds of having a child with ASD, while, secondarily, if these women themselves are at high risk of having the disease. Major databases from inception until 14th October 2018 were searched. The primary outcome measure was the odds of an ASD diagnosis in children of mothers with diagnosed PCOS, while the secondary outcome was the odds of ASD diagnosis in women with PCOS. Scheduled subgroup analyses were according to the time of birth and maternal age. We assessed the odds ratio (OR), using a random-effects model; heterogeneity was assessed by I2 and τ2 statistics. The quality of the evidence was evaluated using the Newcastle-Ottawa Scale. Ten studies were eligible for inclusion, including a total of 33,887 ASD children and 321,661 non-ASD children. Diagnosed PCOS was associated with a 1.66 times increase in the odds of ASD in the offspring [95% CI: 1.51, 1.83, p = 1.99 × 10-25, 7 studies, I2 = 0%, τ2 = 0]. Women with PCOS were 1.78 times more likely to be diagnosed with ASD (95% CI: 1.10, 2.87, p = 0.0179, 5 studies, I2 = 85.4%, τ2 = 0.2432). Additional analyses did not change the initial result. The overall quality of the evidence was high. The pooled effects size displayed low heterogeneity (I2 = 0%) for the primary outcome. While the heterogeneity in the secondary outcome appears to attenuate when only high quality studies are synthesized, still the result exhibits significant heterogeneity. Τhe available data allowed a subgroup analysis only for classification system for PCOS diagnosis and showed a significant increase of ASD diagnosis in the offspring of women with Read Code and ICD diagnosed PCOS. In conclusion, the available evidence suggests that women with PCOS have increased odds of having a child with ASD, an effect size estimate based on a large number of patients from studies of good quality. Regarding the evidence on the prevalence of ASD in PCOS women, results suggest that women with PCOS are more likely to be diagnosed with ASD.

Publication types

  • Meta-Analysis
  • Systematic Review

MeSH terms

  • Autism Spectrum Disorder / etiology*
  • Autism Spectrum Disorder / metabolism
  • Child
  • Female
  • Humans
  • Maternal Inheritance / physiology
  • Mothers
  • Odds Ratio
  • Polycystic Ovary Syndrome / complications*
  • Polycystic Ovary Syndrome / metabolism*
  • Polycystic Ovary Syndrome / physiopathology
  • Pregnancy
  • Prenatal Exposure Delayed Effects
  • Prevalence