The associations between prenatal phthalate exposure measured in child meconium and cognitive functioning of 12-month-old children in two cohorts at elevated risk for adverse neurodevelopment

Background: Phthalate metabolites in gestational-maternal urine represents short-term maternal exposure, but meconium, the newborn's first stool may better capture cumulative fetal exposure. We quantified phthalate metabolites in meconium from two cohorts of children at higher risk of adverse neurodevelopment and evaluated associations with their cognitive function at 12 months. Methods: Meconium phthalate metabolites were quantified in the Safe Passage Study (SPS), N = 720, a pregnancy cohort with high community-levels of prenatal alcohol use, and the Early Autism Risk Longitudinal Investigation (EARLI), N = 236, a high familial autism risk pregnancy cohort. EARLI also had second and third trimester (T2/T3) maternal urine for exposure assessment. Molar sum of di (2-ethylhexyl) (∑DEHP) metabolites and an anti-androgenic score (AAS) using mono-isobutyl, mono-n-butyl, monobenzyl (MBZP), and DEHP metabolites were computed. Cognitive function was assessed at 12 months using the Mullen Scales of Early Learning-Composite (ELC). Multivariable linear regression assessed associations between log_e-transformed metabolites and ELC. Quadratic terms explored nonlinearity and interaction terms of metabolite by child's sex examined effect modification. Results: In SPS, MBzP (β_Linear = −6.73; 95% CI: 12.04, −1.42; β_quadratic = 1.95; 0.27, 3.62) and mono (2-ethyl-5-carboxypentyl), (β_Linear = −3.81; −7.53, −0.27; β_quadratic = 0.93; 0.09, 1.77) had U-shaped associations with ELC. In EARLI, T2 urine mono-carboxyisononyl was associated with linear decrease in ELC, indicating lower cognitive function. Interaction with sex was suggested (P < 0.2) for several urine metabolites, mostly indicating negative association between phthalates and ELC among girls but reversed among boys. Only mono-isononyl phthalate and ∑DEHP had consistent main effect associations across matrixes and cohorts, but similar interaction with sex was observed for meconium-measured ∑DEHP, AAS, MBzP, and mono (2-ethylhexyl) in both cohorts. Conclusions: Few phthalate metabolites were consistently associated with children's cognitive function, but a similar set of meconium metabolites from both cohorts displayed sex-specific associations. Gestational phthalate exposure may have sexually-dimorphic associations with early cognitive function in children at higher risk for adverse neurodevelopment.