Endocrine disrupting chemicals in maternal and umbilical cord plasma and their associations with birthweight in the GUSTO cohort.

Abstract

Background: With daily exposure to multiple endocrine disrupting chemicals (EDCs), understanding individualized co-exposure patterns could better identify chemicals that threaten health. This is particularly pertinent for the vulnerable fetus during in-utero development, where exposure can have long lasting health consequences. As there is limited information of EDC exposure in Asian maternal-offspring populations, this study aimed to (1) determine levels of a selected range of EDCs (focusing on Substances of Very High Concern by the European Chemical Agency) in maternal and corresponding cord blood plasma, (2) investigate the sociodemographic factors associated with plasma EDC concentrations, and (3) associate EDC-mixtures with birthweight, in a Singapore cohort.

Methods: Targeted liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used to determine the concentration of 30 chemicals of interest in 780 maternal and 782 cord plasma samples collected at delivery in the multi-ethnic Asian (Chinese, Malay, Indian) mother-offspring GUSTO study. Quantile-based g-computation was used to estimate the combined effect of chemical mixtures and its association with birthweight.

Results: Twenty-seven out of the thirty selected chemicals were reliably detected in both maternal and cord plasma. Perfluorooctanesulfonic, perfluorooctanoic, perfluorobutanesulfonic and perfluorobutanoic acids (PFOS, PFOA, PFBS, PFBA, respectively) were the predominant perfluoroalkyl acids (detected in > 90% of samples), while mono (2-ethylhexyl) phthalate (MEHP) and monobutyl phthalate were the main phthalate metabolites (detected in > 99% of samples). Concentrations of fourteen chemicals, including PFBA, PFBS and bisphenol S (BPS) were higher in cord plasma than in corresponding maternal plasma; eight being > 1.5 times higher (ranging from 1.75 to 2.93). A mixture of chemicals in cord plasma associated with higher birthweight [116.5 g (95%CI 3.1, 229.9) per quantile increase], but no association was observed for the maternal mixture. Further, different chemicals from the same EDC group in either cord or maternal plasma showed associations in opposite directions with birthweight.

Discussion: Our results suggest substantial transplacental transfer and fetal accumulation of many chemicals, particularly the newer replacement compounds. Stronger associations with birthweight were found for the cord chemical mixture than for the maternal mixture, supporting the idea that these chemicals may have direct effects in the fetus to influence growth. Moreover, individual chemicals within each EDC group appear to have different mechanisms of effect resulting in divergent associations with birthweight.

Conclusion: This study adds to the growing concern about the impact of EDCs, especially the newer chemicals on vulnerable groups such as the developing fetus, warranting further research on the potential effects of in-utero EDC exposure on child health.