Prenatal Fluoride Exposure and Autism Spectrum Disorder-Like Behaviors in a Sample of Canadian Preschool Aged Children: The MIREC Study

Abstract

BACKGROUND AND AIM: A growing body of evidence indicates that prenatal exposure to fluoride is neurotoxic to child development. However, epidemiologic research examining fluoride’s effect on autism spectrum disorder (ASD) is lacking. Our aim was to examine the association of prenatal fluoride exposure with ASD-like behaviors in a sample of preschool aged children living in regions with fluoridated and non-fluoridated community water. METHODS: We analyzed 502 mother-child pairs from the Maternal-Infant Research on Environmental Chemicals (MIREC), a prospective multicentered Canadian birth cohort, with available data on prenatal fluoride exposure, ASD-like behaviors, and covariates. We measured fluoride exposure by averaging maternal urinary fluoride adjusted for specific gravity (MUFsg) across three trimesters. We assessed children’s ASD-like behaviors at 3-4 years of age using the parent-reported Social Responsiveness Scale-2 (SRS-2) Preschool Form, with higher scores indicating more behavioral symptomology. Covariate adjusted linear and quantile regression models were used to estimate the difference in the relationship between MUFsg and SRS-2 conditional mean and quantile subscale T-scores, respectively. We adjusted all regression models for baseline maternal age, second-hand smoke, marital status, parity, child sex, study site, and the HOME score that was assessed at child follow-up visit. RESULTS:Mean MUFsg concentration was 0.53 mg/L (SD=0.38), and mean SRS-2 Total T-score was 45.3 (SD=6.2). In multiple linear regression, a 1 mg/L increase in MUFsg concentration was not statistically associated with SRS-2 Total T-score (β=0.12; 95% CI, -1.36, 1.60; P=0.87). Likewise, we observed no statistical association for all other SRS-2 subscale T-scores. In quantile regression, there was no statistical association between MUFsg and SRS-2 Total T-score across percentiles: 0.1, 0.3, 0.5, 0.7 and 0.9. Effect modification by sex was not observed. CONCLUSIONS:In this Canadian cohort, we found no evidence of a statistical association between prenatal fluoride exposure and ASD-like behaviors among preschool aged children. KEYWORDS: biomarkers of exposure, children’s environmental health, epidemiology, fluoride, neurodevelopmental outcomes