Variability of Urinary Phthalate Metabolite, Bisphenol, and Ultraviolet Filter Concentrations across Sensitive Windows of Human Reproduction

摘要

Background: Pre-pregnancy and early pregnancy levels of phthalate, ultraviolet filter, and bisphenol exposure may be associated with adverse reproductive health endpoints. Characterizing variability across sensitive windows of human fecundity can assist in guiding epidemiologic study design of these exposures. In particular, no prior studies have characterized non-persistent endocrine disrupting chemicals across sensitive windows of human reproduction. Methods: We characterized the variability of 12 urinary phthalate metabolites, 6 bisphenol, and 5 ultraviolet filter concentrations prior to and during early pregnancy from 501 women with 1-4 samples per person collected from 2005-2009. Couples were recruited after discontinuing contraception prior to attempting a pregnancy and were followed daily until pregnant. The 343 (68%) women who became pregnant were followed daily through 7 weeks post-conception. We characterized chemical biomarker variability using intraclass correlation coefficients (ICCs) for women from the cohort overall, and by pregnancy status and race/ethnicity. Results: All ICCs (95% CIs) were <0.50 and varied by chemical class (in descending order): 0.46 (0.39, 0.52) for mEP, 0.36 (0.28, 0.43) for BP-$10.32 (0.24, 0.39) for BP-$3and 0.26 (0.18, 0.34) for mBP. Several phthalates and one benzophenone had lower reliability (ICC 0.20 to 0.29): mCPP, mBP, mibP, mCHP, mCMHP, and 4-OH-BP. All bisphenols measured were quite variable (ICC -0.11 to 0.01). We did not find substantial differences across ICCs when data were stratified by pregnancy status and race/ethnicity. Conclusion: The observed reliability was fair for monoethyl phthalate and poor for all others (ICC<0.4). A single measurement of non-persistent EDCs does not appear to reliably capture exposures over sensitive windows. To this end, longitudinal measurements are needed to ensure the validity of findings relative to these exposures and human fecundity and fertility.