# Mixtures of Urinary Phenol and Phthalate Metabolite Concentrations in Relation to Serum Lipid Levels among Pregnant Women: Results from the EARTH Study

## TL;DR

This study explored how mixtures of chemicals in urine relate to blood lipid levels in pregnant women, finding no overall link but some hints in specific groups.

## Contribution

The study provides novel insights into chemical mixtures and lipid levels during pregnancy, particularly in high BMI subgroups.

## Key findings

- No overall association found between chemical mixtures and serum lipid levels.
- Suggestive positive relationships observed in high BMI women for propylparaben and cholesterol.
- Need for larger studies to confirm these findings.

## Abstract

We examined whether mixtures of urinary concentrations of bisphenol A (BPA), parabens and phthalate metabolites were associated with serum lipid levels among 175 pregnant women who enrolled in the Environment and Reproductive Health (EARTH) Study (2005–2017), including triglycerides, total cholesterol, high-density lipoprotein (HDL), non-HDL, and low-density lipoprotein (LDL). We applied Bayesian Kernel Machine Regression (BKMR) and quantile g-computation while adjusting for confounders. In the BKMR models, we found no associations between chemical mixture and lipid levels, e.g., total cholesterol [mean difference (95% CRI, credible interval) = 0.02 (−0.31, 0.34)] and LDL [mean difference (95% CRI) = 0.10 (−0.22, 0.43)], when comparing concentrations at the 75th to the 25th percentile. When stratified by BMI, we found suggestive positive relationships between urinary propylparaben and total cholesterol and LDL among women with high BMI [mean difference (95% CRI) = 0.25 (−0.26, 0.75) and 0.35 (−0.25, 0.95)], but not with low BMI [mean difference (95% CRI) = 0.00 (−0.06, 0.07) and 0.00 (−0.07, 0.07)]. No association was found by quantile g-computation. This exploratory study suggests mixtures of phenol and phthalate metabolites were not associated with serum lipid levels during pregnancy, while there were some suggestive associations for certain BMI subgroups. Larger longitudinal studies with multiple assessments of both exposure and outcome are needed to corroborate these novel findings.