Department of Biostatistics, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston, MA, 02115, USA.
Department of Environmental Health, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston, MA, 02115, USA; Department of Epidemiology, Harvard T.H. Chan School of Public Health, 677 Huntington Avenue, Boston, MA, 02115, USA.
Environ Res. 2021 Mar;194:110638. doi: 10.1016/j.envres.2020.110638. Epub 2020 Dec 25.
Exposure to endocrine disrupting chemicals (EDC), such as phthalates and phenols, during pregnancy may be associated with excessive gestational weight gain (GWG), an important predictor of future health of the mother and the offspring. There is however a paucity of literature examining this association, and no study has accounted for the complex nature of EDCs exposure as a time-varying mixture of chemicals.
We examined the association between trimester-specific EDCs mixture and GWG in pregnant women attending a fertility clinic, to identify windows of susceptibility to such exposures, and assess the individual contribution of each chemical over pregnancy.
We included 243 pregnant women from the Environment and Reproductive Health (EARTH) Study, who provided up to 3 urine samples (one per trimester), and with available data on GWG. Urinary concentrations of 7 phthalate metabolites, bisphenol A, and 2 parabens, corrected for specific gravity, were included in the analysis. The association between trimester-specific EDCs mixture and GWG was evaluated using multiple regression models - categorizing exposures into concentration quartiles- and with Bayesian Kernel Machine Regression (BKMR), while adjusting for potential confounders. Hierarchical BKMR (hBKMR) was used to account for the time-varying nature of chemical concentrations over pregnancy, identifying the most important trimester and most important EDC within each trimester.
During 1st trimester, higher GWG was observed at higher sum of metabolites of di (2-ethylhexyl) phthalate (ΣDEHP) from both multiple regression (e.g. comparing the 4th quartile with the 1st: β = 2.36 kg, 95% CI: 0.47, 5.19) and BKMR. During 2nd and 3rd trimesters, positive associations with mono-n-butyl phthalate and propylparaben, and negative with ΣDEHP and methylparaben were observed. When evaluating exposures as a time-varying mixture with hBKMR, 1st trimester was the most important exposure window when evaluating prenatal urinary EDCs in relation to GWG. Within the 1st trimester, urinary ΣDEHP, mono-isobutyl phthalate and propylparaben had the highest contribution in the positive association between the mixture and GWG.
We observed positive associations between urinary EDCs during pregnancy, especially DEHP metabolites, and GWG. Our results suggest the 1st trimester of pregnancy as the time window of highest susceptibility to the effects of EDCs on GWG, with potential indication for the design of public health interventions, informing prevention strategies for reducing sources of exposure at specific time points.
孕期接触内分泌干扰化学物质（EDC），如邻苯二甲酸酯和酚类物质，可能与过度妊娠体重增加（GWG）有关，而 GWG 是母亲和后代未来健康的一个重要预测指标。然而，目前关于这种关联的文献很少，并且没有研究考虑到 EDC 暴露的复杂性，即作为时间变化的化学混合物的暴露。
我们研究了在生育诊所就诊的孕妇中特定孕期 EDC 混合物与 GWG 之间的关联，以确定对这些暴露的易感性窗口，并评估每个化学物质在整个孕期对 GWG 的单独贡献。
我们纳入了来自环境与生殖健康（EARTH）研究的 243 名孕妇，她们提供了最多 3 份尿液样本（每个孕期 1 份），并提供了 GWG 的可用数据。对经过比重校正的 7 种邻苯二甲酸酯代谢物、双酚 A 和 2 种对羟基苯甲酸酯的尿液浓度进行了分析。使用多回归模型-将暴露分为浓度四分位数-以及贝叶斯核机器回归（BKMR），同时调整潜在混杂因素，评估特定孕期 EDC 混合物与 GWG 之间的关联。分层 BKMR（hBKMR）用于解释孕期化学物质浓度的时变性质，确定每个孕期最重要的孕期和最重要的 EDC。
在孕早期，从多回归（例如，将第 4 四分位数与第 1 四分位数进行比较：β=2.36kg，95%CI：0.47，5.19）和 BKMR 中观察到较高的ΣDEHP（二（2-乙基己基）邻苯二甲酸酯的代谢物）总和与较高的 GWG 相关。在孕 2 期和 3 期，观察到与单正丁基邻苯二甲酸酯和丙基对羟基苯甲酸酯呈正相关，与 ΣDEHP 和甲基对羟基苯甲酸酯呈负相关。当使用 hBKMR 评估作为时变混合物的暴露时，1 期是评估与 GWG 相关的产前尿液 EDC 时最重要的暴露窗口。在 1 期内，尿液中 ΣDEHP、单异丁基邻苯二甲酸酯和丙基对羟基苯甲酸酯在混合物与 GWG 之间的正相关关系中具有最高的贡献。
我们观察到孕期尿液内分泌干扰化学物质（EDC），特别是邻苯二甲酸酯代谢物与 GWG 之间存在正相关关系。我们的结果表明，妊娠 1 期是 EDC 对 GWG 影响的易感性最高的窗口期，这为设计公共卫生干预措施提供了潜在依据，为在特定时间点减少暴露源提供了预防策略。
