Phenols such as bisphenol A (BPA), triclosan, and benzophenone-3 were shown in rodent models to exert adverse effects, including alterations in uterine arterial remodeling, and on decidualization, implantation, placental development, and intrauterine growth restriction (33Yuan M. Hu M. Lou Y. Wang Q. Mao L. Zhan Q. et al.Environmentally relevant levels of bisphenol A affect uterine decidualization and embryo implantation through the estrogen receptor/serum and glucocorticoid-regulated kinase 1/epithelial sodium ion channel α-subunit pathway in a mouse model., 34Santamaria C.G. Meyer N. Schumacher A. Zenclussen M.L. Teglia C.M. Culzoni M.J. et al.Dermal exposure to the UV filter benzophenone-3 during early pregnancy affects fetal growth and sex ratio of the progeny in mice., 35Cao X. Hua X. Wang X. Chen L. Exposure of pregnant mice to triclosan impairs placental development and nutrient transport., 36Müller J.E. Meyer N. Santamaria C.G. Schumacher A. Luque E.H. Zenclussen M.L. et al.Bisphenol A exposure during early pregnancy impairs uterine spiral artery remodeling and provokes intrauterine growth restriction in mice.). Although these disruptions may increase the preterm birth risk, rodent studies constitute an inadequate experimental model (2Environmental chemicals and preterm birth: biological mechanisms and the state of the science.). Human studies are thus needed to investigate the environmental influences on preterm birth. A limited number of studies have examined the associations between prenatal phenol exposure and preterm birth (24Aker A.M. Ferguson K.K. Rosario Z.Y. Mukherjee B. Alshawabkeh A.N. Cordero J.F. et al.The associations between prenatal exposure to triclocarban, phenols and parabens with gestational age and birth weight in northern Puerto Rico., 37Huang S. Li J. Xu S. Zhao H. Li Y. Zhou Y. et al.Bisphenol A and bisphenol S exposures during pregnancy and gestational age—a longitudinal study in China., 38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth., 39Cantonwine D. Meeker J.D. Hu H. Sanchez B.N. Lamadrid-Figueroa H. Mercado-Garcia A. et al.Bisphenol A exposure in Mexico City and risk of prematurity: a pilot nested case control study., 40Tang R. Chen M.J. Ding G.D. Chen X.J. Han X.M. Zhou K. et al.Associations of prenatal exposure to phenols with birth outcomes., 41Weinberger B. Vetrano A.M. Archer F.E. Marcella S.W. Buckley B. Wartenberg D. et al.Effects of maternal exposure to phthalates and bisphenol A during pregnancy on gestational age.). The findings were inconsistent, possibly because of the heterogeneity in study design and the timing of exposure assessment (24Aker A.M. Ferguson K.K. Rosario Z.Y. Mukherjee B. Alshawabkeh A.N. Cordero J.F. et al.The associations between prenatal exposure to triclocarban, phenols and parabens with gestational age and birth weight in northern Puerto Rico., 37Huang S. Li J. Xu S. Zhao H. Li Y. Zhou Y. et al.Bisphenol A and bisphenol S exposures during pregnancy and gestational age—a longitudinal study in China., 38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth., 39Cantonwine D. Meeker J.D. Hu H. Sanchez B.N. Lamadrid-Figueroa H. Mercado-Garcia A. et al.Bisphenol A exposure in Mexico City and risk of prematurity: a pilot nested case control study., 40Tang R. Chen M.J. Ding G.D. Chen X.J. Han X.M. Zhou K. et al.Associations of prenatal exposure to phenols with birth outcomes., 41Weinberger B. Vetrano A.M. Archer F.E. Marcella S.W. Buckley B. Wartenberg D. et al.Effects of maternal exposure to phthalates and bisphenol A during pregnancy on gestational age.). Most previous studies assessed phenol exposure at a single pregnancy time-point, for example, in the second (24Aker A.M. Ferguson K.K. Rosario Z.Y. Mukherjee B. Alshawabkeh A.N. Cordero J.F. et al.The associations between prenatal exposure to triclocarban, phenols and parabens with gestational age and birth weight in northern Puerto Rico.) or third (39Cantonwine D. Meeker J.D. Hu H. Sanchez B.N. Lamadrid-Figueroa H. Mercado-Garcia A. et al.Bisphenol A exposure in Mexico City and risk of prematurity: a pilot nested case control study.) trimester, or just before delivery (40Tang R. Chen M.J. Ding G.D. Chen X.J. Han X.M. Zhou K. et al.Associations of prenatal exposure to phenols with birth outcomes., 41Weinberger B. Vetrano A.M. Archer F.E. Marcella S.W. Buckley B. Wartenberg D. et al.Effects of maternal exposure to phthalates and bisphenol A during pregnancy on gestational age.). These studies failed to account for the high variability of phenol exposure across pregnancy and the potentially different periods of vulnerability during pregnancy (2Environmental chemicals and preterm birth: biological mechanisms and the state of the science., 42Meeker J.D. Cantonwine D.E. Rivera-González L.O. Ferguson K.K. Mukherjee B. Calafat A.M. et al.Distribution, variability, and predictors of urinary concentrations of phenols and parabens among pregnant women in Puerto Rico., 43Braun J.M. Kalkbrenner A.E. Calafat A.M. Bernert J.T. Ye X. Silva M.J. et al.Variability and predictors of urinary bisphenol A concentrations during pregnancy.). Exploring windows of exposure across pregnancy can suggest pathways of adverse effects of phenols on preterm birth (2Environmental chemicals and preterm birth: biological mechanisms and the state of the science.). First trimester exposure may interfere with implantation or placentation, which may result in subsequent preterm delivery (33Yuan M. Hu M. Lou Y. Wang Q. Mao L. Zhan Q. et al.Environmentally relevant levels of bisphenol A affect uterine decidualization and embryo implantation through the estrogen receptor/serum and glucocorticoid-regulated kinase 1/epithelial sodium ion channel α-subunit pathway in a mouse model., 34Santamaria C.G. Meyer N. Schumacher A. Zenclussen M.L. Teglia C.M. Culzoni M.J. et al.Dermal exposure to the UV filter benzophenone-3 during early pregnancy affects fetal growth and sex ratio of the progeny in mice., 36Müller J.E. Meyer N. Santamaria C.G. Schumacher A. Luque E.H. Zenclussen M.L. et al.Bisphenol A exposure during early pregnancy impairs uterine spiral artery remodeling and provokes intrauterine growth restriction in mice.). Second trimester exposure might disrupt the metabolic adaptation of mothers and nutrition transfer to fetuses, resulting in fetal growth restriction, preeclampsia, glucose intolerance, or other metabolic diseases (35Cao X. Hua X. Wang X. Chen L. Exposure of pregnant mice to triclosan impairs placental development and nutrient transport.), which are direct risk factors for preterm birth. Third trimester exposure may be related to increasing triggers for preterm delivery such as inflammation and the cytokine cascade needed to trigger labor (44Cho Y.J. Park S.B. Park J.W. Oh S.R. Han M. Bisphenol A modulates inflammation and proliferation pathway in human endometrial stromal cells by inducing oxidative stress., 45Ferguson K.K. Chen Y.-H. VanderWeele T.J. McElrath T.F. Meeker J.D. Mukherjee B. Mediation of the relationship between maternal phthalate exposure and preterm birth by oxidative stress with repeated measurements across pregnancy.).To date, only two studies have examined repeated measurements across trimesters: a nested case-control study from the LIFECODES cohort in Boston, Massachusetts (38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth., 46Aung M.T. Ferguson K.K. Cantonwine D.E. McElrath T.F. Meeker J.D. Preterm birth in relation to the bisphenol A replacement, bisphenol S, and other phenols and parabens.) and a prospective cohort in China (38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth.). The case-control study examined phenol concentrations from four prenatal visits in relation to preterm birth. The investigators found that urinary BPA and BPS concentrations at the fourth visit were associated with higher odds of preterm birth (38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth., 46Aung M.T. Ferguson K.K. Cantonwine D.E. McElrath T.F. Meeker J.D. Preterm birth in relation to the bisphenol A replacement, bisphenol S, and other phenols and parabens.). As with all preterm birth studies, the findings in the third trimester may be influenced by selection bias as women delivering early would fail to attend the fourth visit and thus not be included in the case ascertainment. The Chinese cohort observed that prenatal urinary BPA concentrations, especially in the second and third trimesters, were associated with higher odds of preterm birth. However, pregnancies in this Chinese cohort were possibly at lower risk overall given only 2.5% of the births ended preterm. Thus, the findings may not be informative to the general population (37Huang S. Li J. Xu S. Zhao H. Li Y. Zhou Y. et al.Bisphenol A and bisphenol S exposures during pregnancy and gestational age—a longitudinal study in China.). In addition, several studies suggested that infant sex may modify the association between prenatal phenol exposure and preterm birth, but the evidence is limited and inconclusive (24Aker A.M. Ferguson K.K. Rosario Z.Y. Mukherjee B. Alshawabkeh A.N. Cordero J.F. et al.The associations between prenatal exposure to triclocarban, phenols and parabens with gestational age and birth weight in northern Puerto Rico., 38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth.). In this prospective preconception cohort of subfertile couples, we aimed to investigate the association of prenatal urinary phenol concentrations and the risk of preterm birth, exploring the timing of vulnerability across gestational windows and the potential modification by infant sex.

In this prospective cohort of women attending a fertility clinic, we found suggestive positive associations between middle and late pregnancy BPA concentrations and preterm birth, especially among women carrying female fetuses. Mid-to-late pregnancy may be a potentially vulnerable period of risk for preterm birth in relation to BPA exposure, although imprecision may have precluded a firmer determination of this window. In support of this finding, we also reported that BPA concentrations were higher across all gestational weeks comparing mothers with and without infants born preterm, although CIs were again imprecise. Consistent with the multiple informant model analysis, sex-specific associations were found among female infants but not males, with a similar pattern of higher exposure across mid-pregnancy among women with infants born preterm. Restricting our sample to women providing all three urine samples and applying inverse probability weights strengthened our findings, suggesting possible differences among women providing complete exposure data. Triangulation of our results using two separate statistical methods showed consistency, counteracting the imprecision and potential selection bias issue of our study.

First trimester urinary ∑Paraben concentrations were associated with a higher risk of preterm birth, especially among female infants. Protective associations with first trimester bezonphenone-3 and first and third trimester triclosan concentrations and preterm birth were also detected. We did not observe any meaningful differences in biomarkers concentration distributions across gestational weeks comparing preterm and non-preterm births using quadratic mixed models for parabens, benzophenone-3, or triclosan.

In this study, the intraclass correlation coefficients were low for BPA and BPS and moderate for parabens and benzophenone-3 across gestation, which is consistent with previous studies (43Braun J.M. Kalkbrenner A.E. Calafat A.M. Bernert J.T. Ye X. Silva M.J. et al.Variability and predictors of urinary bisphenol A concentrations during pregnancy., 64Assens M. Frederiksen H. Petersen J. Larsen T. Skakkebæk N. Juul A. et al.Variations in repeated serum concentrations of UV filters, phthalates, phenols and parabens during pregnancy., 57Liu B. Xu G. Sun Y. Qiu X. Ryckman K.K. Yu Y. et al.Maternal cigarette smoking before and during pregnancy and the risk of preterm birth: A dose-response analysis of 25 million mother-infant pairs.). The high variability of bisphenols and the moderate variability of the remaining phenols may be attributed to differences in lifestyles linked to personal care product use, diet, and other behaviors; however, it may also represent changes in metabolism during pregnancy.Our finding that mid-to-late pregnancy BPA exposure was positively associated with the risk of preterm birth among female infants is compatible with several previous studies. Importantly, a nested case-control study in Boston, Massachusetts, reported a consistent fetus sex modification as in our study. Cantonwine et al. (38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth.) found mean prenatal urinary BPA concentration was associated with the risk of preterm birth only among female infants. Furthermore, they observed urinary BPA concentrations at the early-to-middle second trimester and late third trimesters were positively associated with placental preterm birth and spontaneous preterm birth risk, respectively (38Cantonwine D.E. Ferguson K.K. Mukherjee B. McElrath T.F. Meeker J.D. Urinary bisphenol A levels during pregnancy and risk of preterm birth.). In our study, one third of the preterm birth cases were spontaneous, which might explain why we observed less obvious associations in the third trimester. But we did consistently observe a positive association between third trimester BPA concentrations and preterm birth risk for female infants. One cohort study in China reported that both second and third trimester urinary BPA concentrations were associated with higher preterm birth risk (37Huang S. Li J. Xu S. Zhao H. Li Y. Zhou Y. et al.Bisphenol A and bisphenol S exposures during pregnancy and gestational age—a longitudinal study in China.), which is also consistent with our study findings. Similarly, another nested case-control study with 30 preterm birth cases in Mexico City found that third trimester urinary BPA concentrations were higher among mothers who delivered preterm compared with concentrations in those who delivered full term; however, the results were only significant in unadjusted models and the study did not quantify BPA at first or second trimester (39Cantonwine D. Meeker J.D. Hu H. Sanchez B.N. Lamadrid-Figueroa H. Mercado-Garcia A. et al.Bisphenol A exposure in Mexico City and risk of prematurity: a pilot nested case control study.). These two studies did not examine infant sex as a modifier. Additionally, other studies examined gestational age as a continuous outcome in relation to prenatal BPA exposure. Two cross-sectional studies reported lower gestational age (days) with higher urinary BPA concentrations at delivery (40Tang R. Chen M.J. Ding G.D. Chen X.J. Han X.M. Zhou K. et al.Associations of prenatal exposure to phenols with birth outcomes., 41Weinberger B. Vetrano A.M. Archer F.E. Marcella S.W. Buckley B. Wartenberg D. et al.Effects of maternal exposure to phthalates and bisphenol A during pregnancy on gestational age.). In contrast, a cohort study in Puerto Rico found that urinary BPA concentrations in the early second trimester (16–20 weeks) were associated with lower odds of preterm birth and prolonged gestational age among 607 women. However, these associations were attenuated in relation to middle and late second trimester concentrations (24Aker A.M. Ferguson K.K. Rosario Z.Y. Mukherjee B. Alshawabkeh A.N. Cordero J.F. et al.The associations between prenatal exposure to triclocarban, phenols and parabens with gestational age and birth weight in northern Puerto Rico.). Differences in demographics in this Puerto Rican cohort compared with the present study included a higher proportion of multiparous women (52% vs. 17%), a slightly higher preterm birth rate (10% vs. 8%), and a doubling of second trimester BPA concentrations (SG-corrected GM [GSD]: 2.16 [2.49] vs. 1.04 [0.06] ng/mL), respectively. In addition, the investigators did not consider maternal race, a strong indicator for preterm birth especially in the context of Puerto Rico (65Goldenberg R.L. Culhane J.F. Iams J.D. Romero R. Epidemiology and causes of preterm birth.) nor patterns of BPA exposure across the whole pregnancy (66Lehmler H.-J. Liu B. Gadogbe M. Bao W. Exposure to bisphenol A, bisphenol F, and bisphenol S in U.S. adults and children: the National Health and Nutrition Examination Survey 2013–2014.), which likely contributed to these inconsistencies with our findings.In our study, we report that first trimester paraben concentrations were positively associated with a higher risk of preterm birth. Similarly, a nested case-control study in Boston found that first and early second trimester urinary ethylparaben concentrations were associated with higher odds of placental preterm birth (46Aung M.T. Ferguson K.K. Cantonwine D.E. McElrath T.F. Meeker J.D. Preterm birth in relation to the bisphenol A replacement, bisphenol S, and other phenols and parabens.). Another cohort study in New York reported null findings between third trimester paraben concentrations and preterm birth. However, in examining gestational age continuously, the investigators found third trimester butylparaben exposure was associated with shorter gestational duration (67Geer L.A. Pycke B.F.G. Waxenbaum J. Sherer D.M. Abulafia O. Halden R.U. Association of birth outcomes with fetal exposure to parabens, triclosan and triclocarban in an immigrant population in Brooklyn, New York.). In contrast, Aker et al. (24Aker A.M. Ferguson K.K. Rosario Z.Y. Mukherjee B. Alshawabkeh A.N. Cordero J.F. et al.The associations between prenatal exposure to triclocarban, phenols and parabens with gestational age and birth weight in northern Puerto Rico.) observed mean second trimester urinary methyl- and propyl-paraben concentrations were associated with longer gestational duration in a cohort study in Puerto Rico. Of note, this cohort had lower parabens concentrations across trimesters than those of our study participants, and again the investigators did not consider race in their analyses, which may be an important confounder. No prior study examined sex differences in relation to paraben exposure and preterm birth risk. Our sex-specific findings should therefore be interpreted cautiously in light of the limited power to detect differences and the possibility of false-positive results.We reported that urinary triclosan concentrations were negatively associated with preterm birth risk. Only two studies investigated the association between prenatal triclosan exposure and the risk of preterm birth. A nested case-control study found second trimester triclosan concentrations were associated with diminished risk of spontaneous preterm birth (46Aung M.T. Ferguson K.K. Cantonwine D.E. McElrath T.F. Meeker J.D. Preterm birth in relation to the bisphenol A replacement, bisphenol S, and other phenols and parabens.). Although a cohort study in Puerto Rico found no association between mean second trimester urinary triclosan concentrations and preterm birth, they did report shorter gestational duration among female infants and longer pregnancy length among male infants (24Aker A.M. Ferguson K.K. Rosario Z.Y. Mukherjee B. Alshawabkeh A.N. Cordero J.F. et al.The associations between prenatal exposure to triclocarban, phenols and parabens with gestational age and birth weight in northern Puerto Rico.), which was similar to our findings. As for gestational age as the outcome, in a Chinese cohort with relatively low urinary triclosan detection frequencies, triclosan concentrations at delivery were associated with longer gestational age, although association were null in adjusted models (68Huo W. Xia W. Wu C. Zhu Y. Zhang B. Wan Y. et al.Urinary level of triclosan in a population of Chinese pregnant women and its association with birth outcomes.). However, a prospective cohort in the United States found the mean second trimester urinary triclosan concentration was associated with shorter gestational duration, and no sex differences were observed (8Etzel T.M. Calafat A.M. Ye X. Chen A. Lanphear B.P. Savitz D.A. et al.Urinary triclosan concentrations during pregnancy and birth outcomes.). Findings on gestational age may not reflect the associations on preterm birth.In our study, first trimester urinary benzophenone-3 concentrations were associated with lower preterm birth risk with this potential protective effect being more evident among male compared with female infants. Aker et al. (