Exposure to environmental chemicals in early life is associated with ADHD and autism spectrum disorders in Norwegian children

摘要

Background: Environmental pollutants are ubiquitous in daily life, and exposure begins in utero. Multiple environmental chemicals are established or suspected neurotoxicants, yet few studies have assessed associations with the risk of attention-deficit hyperactivity disorder (ADHD) or autism spectrum disorders (ASD), two common neurodevelopmental disorders. We investigated measured perinatal and estimated postnatal chemical exposure levels in relation to ADHD and ASD. Methods: A birth cohort of 2606 Norwegian mother-child pairs was followed until children were a median age of 11 years (HUMIS-NoMIC cohort). ADHD (n = 40) and ASD (n = 15) cases were ascertained using the national patient register (ICD-10 classifications). In a sample of 1199 oversampled by case status, concentrations of 45 environmental chemicals [18 polychlorinated biphenyls (PCBs), 14 organochlorine pesticides, 6 brominated flame retardants, 2 novel flame retardants, 2 poly- and perfluoroalkyl substances (PFASs), 2 pyrethroid pesticides, and methylmercury] were quantified in ≥50% of pooled maternal breast milk samples, reflecting both pre- and postnatal exposures. Postnatal concentrations in the first two years of the child's life were modelled for the lipophilic chemicals using a pharmacokinetic model. To identify associations between exposures and neurodevelopmental outcomes, we used confounder-adjusted elastic net penalized logistic regression models to account for confounding by correlated co-exposures; unpenalized effect estimates for selected exposures were obtained from multivariable regression models. Results: Eight chemicals were associated with ADHD. Perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA) and β-hexachlorocyclohexane (β-HCH) were associated with an increased risk (range, OR= 1.64-2.49); the largest effect estimate was observed for β-HCH (OR = 2.49; 95% CI: 1.22, 5.09 per 2-SD increase in ln-transformed maternal breast milk levels). Several chemicals showed inverse associations with ADHD. β-HCH was associated with an increased risk of ASD (OR= 4.28, 95% CI 1.82, 10.08); associations were null for other chemicals. Associations with postnatal exposures estimates were generally attenuated. Conclusions: In a multi-pollutant analysis of 7 classes of chemicals, early-life exposure to several ubiquitous persistent organic pollutants was associated with ADHD, and one with ASD. Further chemical exposome-wide association studies are warranted to replicate these findings.