Prenatal Phthalate Exposure and Child Cognitive Performance in a Large, Well-Characterized Pregnancy Cohort Study

Abstract

PDS 63: Chemicals and metals: health effects, Exhibition Hall (PDS), Ground floor, August 27, 2019, 10:30 AM - 12:00 PM Background: Studies examining associations between prenatal phthalate exposure and neurodevelopment, and whether associations are sex-specific have yielded inconsistent findings. We investigated associations between prenatal phthalate exposure and child Full Scale IQ (FSIQ) in the Conditions Affecting Neurocognitive Development and Learning in Early childhood (CANDLE) study of the ECHO PATHWAYS Consortium. Methods: We measured 13 phthalate metabolites in third trimester urine in mother-child dyads who completed a preschool visit. Child FSIQ (ages 4-6) was assessed using the Stanford Binet-5. We used multivariable linear regression to estimate associations with individual phthalates, and Weighted Quantile Sum (WQS) regression to identify independent phthalate mixtures that were negatively and positively associated with FSIQ. Final models were adjusted for: maternal IQ, race, marital status, smoking, BMI, socioeconomic status; child age, sex, and breastfeeding. We assessed effect modification by sex in multivariable and WQS regressions, and estimated WQS results in sex-stratified samples. Results: Our sample (N=1022) was predominantly African American (62%) with a high school education or less (57%). No significant associations were observed in individual phthalate or full-sample WQS regressions. In girls, sex-stratified WQS analyses identified a phthalate mixture dominated by mono-carboxy-isononyl phthalate (mCINP; weight=73%) that was positively associated with IQ [B=0.99; 95%CI: 0.02, 1.96], as well as a mixture of mono-benzyl phthalate (mBzBP; weight=34%), mono-2-isobutyl phthalate (mIBP; weight=16%), and mono-[(2-carboxymethyl)hexyl] phthalate (mCMHP; weight=14%) that was negatively associated with IQ [B=-1.61; 95%CI: -2.93, -0.28]. In boys, a phthalate mixture weighted on mono-methyl phthalate (mMP; weight=35%) and mono-carboxy-isooctyl phthalate (mCIOP; weight=19%) was positively associated with IQ (B=1.96; 95%CI: 0.38, 3.54). Conclusions: In one of the largest studies to date, we observed no association between individual phthalate metabolites and child FSIQ, which may be explained by rigorous adjustment for potential confounders. Our WQS analyses suggest that there may be small, sex-specific associations between prenatal phthalate exposure and child cognitive performance.