Prenatal phenol and phthalate exposures and birth outcomes

Abstract

Background: Many phthalates and phenols are hormonally active and are suspected to alter the course of development.

Objective: We investigated prenatal exposures to phthalate and phenol metabolites and their associations with body size measures of the infants at birth.

Methods: We measured 5 phenol and 10 phthalate urinary metabolites in a multiethnic cohort of 404 women in New York City during their third trimester of pregnancy and recorded size of infants at birth.

Results: Median urinary concentrations were > 10 microg/L for 2 of 5 phenols and 6 of 10 phthalate monoester metabolites. Concentrations of low-molecular-weight phthalate monoesters (low-MWP) were approximately 5-fold greater than those of high-molecular-weight metabolites. Low-MWP metabolites had a positive association with gestational age [0.97 day gestational age per ln-biomarker; 95% confidence interval (CI), 0.07-1.9 days, multivariate adjusted] and with head circumference. Higher prenatal exposures to 2,5-dichlorophenol (2,5-DCP) predicted lower birth weight in boys (-210 g average birth weight difference between the third tertile and first tertile of 2,5-DCP; 95% CI, 71-348 g). Higher maternal benzophenone-3 (BP3) concentrations were associated with a similar decrease in birth weight among girls but with greater birth weight in boys.

Conclusions: We observed a range of phthalate and phenol exposures during pregnancy in our population, but few were associated with birth size. The association of 2,5-DCP and BP3 with reduced or increased birth weight could be important in very early or small-size births. In addition, positive associations of urinary metabolites with some outcomes may be attributable partly to unresolved confounding with maternal anthropometric factors.

Keywords: 2,5-DCP; BMI; birth length; birth weight; creatinine; phenols; phthalates; pregnancy; urinary biomarker.

Figure 1

Adjusted mean birth weight ± SE predicted by prenatal maternal 2,5-DCP (A) and BP3 (B) tertiles of creatinine-corrected urine concentrations. The male sex*biomarker interaction terms in models with tertiles of biomarkers had p < 0.01. Adjusted means differed significantly between the first and third tertiles of 2,5-DCP for boys (p = 0.0016), and of BP3 for boys (p = 0.026) and girls (p = 0.021). Birth-weight predicted means are adjusted for race/ethnicity, gestational age, ln-creatinine, smoking during pregnancy, maternal education, marital status, and prepregnancy BMI and are limited to samples with ≥ 20 mg/dL creatinine. The ranges of values in the third tertiles were 114–9,950 μg/gC, 27–13,300 μg/L for 2,5-DCP, and 26–104,000 μg/gC, 7–92,700 μg/L for BP3.