Prenatal exposure to phthalates and preschool children's intellectual scores: Effect modification by child sex.

BACKGROUND

Some epidemiological studies suggest that phthalates adversely affect children's neurodevelopment, but results are inconsistent and often only consider individual chemicals.

OBJECTIVES

We examined associations between gestational phthalate metabolites and intelligence at ages 3-4 and investigated effect modification by child sex.

METHODS

We analyzed data from 511 mother-child pairs enrolled in the Maternal-Infant Research on Environmental Chemicals study. We measured specific gravity-standardized urinary concentrations of 20 phthalate metabolites (from nine parent compounds) in trimesters 1 and 2 and computed the average. We used linear regression models (log2-transformed), adjusting for covariates, to examine single-chemical associations with Full-Scale (FSIQ), Performance (PIQ), and Verbal IQ (VIQ) measured by the Weschler Preschool and Primary Scale of Intelligence-III. We estimated mixture effects using weighted quantile sum (WQS) regression.

RESULTS

Detection frequencies of the metabolites ranged from 42 to 99%. In single-chemical analyses, the sum of di-isodecyl phthalate metabolites (ΣDiDP) was associated with decreased FSIQ (-0.5, 95%CI: -2.1, -0.0) and PIQ (-0.6, 95%CI: -1.2, -0.0). Monomethyl phthalate (MMP) was associated with increased FSIQ (1.7, 95%CI: 0.4, 3.1), PIQ (1.7, 95%CI: 0.2, 3.1) and VIQ; (1.6, 95%CI: 0.3, 2.8). Monobenzyl phthalate (MBzP) was positively associated with PIQ (0.9, 95%CI: 0.0, 1.9). Associations between ΣDiDP and VIQ and MMP and PIQ differed by child sex (negative for boys and positive for girls, respectively). WQS analyses were not statistically significant.

CONCLUSIONS

Findings from this pan-Canadian cohort provide some evidence that gestational ΣDiDP is adversely associated with child cognition, with sex-specific vulnerabilities. The positive associations with MMP and MBzP warrant further research.