The associations of phthalate biomarkers during pregnancy with later glycemia and lipid profiles

• Published in: Environment international 2021-10, Vol.155, p.106612-106612, Article 106612
• Main Authors: Wu, Haotian, Just, Allan C., Colicino, Elena, Calafat, Antonia M., Oken, Emily, Braun, Joseph M., McRae, Nia, Cantoral, Alejandra, Pantic, Ivan, Pizano-Zárate, María Luisa, Tolentino, Mary Cruz, Wright, Robert O., Téllez-Rojo, Martha M., Baccarelli, Andrea A., Deierlein, Andrea L.
• Format: Article
• Language: English
• Subjects: Diabetes; Gestation; Metabolic; Phthalates; Postpartum; Pregnancy
• ISSN: 0160-4120; 1873-6750
• DOI: 10.1016/j.envint.2021.106612

•Phthalate mixture at pregnancy was associated with long-term insulin resistance.•Phthalate mixture at pregnancy was associated with long term adverse lipid profiles.•Associations with glycemia biomarkers were primarily driven by MECPTP and ∑DBP.•Associations with lipid biomarkers were primarily driven by MECPTP, ∑DBP, and MEP.•Pregnancy phthalates exposure may be associated with long-term metabolic health. Pregnancy induces numerous cardiovascular and metabolic changes. Alterations in these sensitive processes may precipitate long-term post-delivery health consequences. Studies have reported associations between phthalates and metabolic complications of pregnancy, but no study has investigated metabolic outcomes beyond pregnancy. To examine associations of exposure to phthalates during pregnancy with post-delivery metabolic health. We quantified 15 urinary phthalate biomarker concentrations during the second and third trimesters among 618 pregnant women from Mexico City. Maternal metabolic health biomarkers included fasting blood measures of glycemia [glucose, insulin, Homeostatic Model Assessment of Insulin Resistance [HOMA-IR], % hemoglobin A1c (HbA1c%)] and lipids (total, high-density lipoprotein (HDL), low-density lipoprotein (LDL) cholesterol, triglycerides), at 4–5 and 6–8 years post-delivery. To estimate the influence of the phthalates mixture, we used Bayesian weighted quantile sum regression and Bayesian kernel machine regression; for individual biomarkers, we used linear mixed models. As a mixture, higher urinary phthalate biomarker concentrations during pregnancy were associated with post-delivery concentrations of plasma glucose (interquartile range [IQR] difference: 0.13 SD, 95%CrI: 0.05, 0.20), plasma insulin (IQR difference: 0.06 SD, 95%CrI: −0.02, 0.14), HOMA-IR (IQR difference: 0.08 SD, 95% CrI: 0.01, 0.16), and HbA1c% (IQR difference: 0.15 SD, 95%CrI: 0.05, 0.24). Associations were primarily driven by mono-2-ethyl-5-carboxypentyl terephthalate (MECPTP) and the sum of dibutyl phthalate biomarkers (∑DBP). The phthalates mixture was associated with lower HDL (IQR difference: −0.08 SD, 95%CrI: −0.16, −0.01), driven by ∑DBP and monoethyl phthalate (MEP), and higher triglyceride levels (IQR difference: 0.15 SD, 95%CrI: 0.08, 0.22), driven by MECPTP and MEP. The overall mixture was not associated with total cholesterol and LDL. However, ∑DBP and MEP were associated with lower and higher total cholesterol, respectively, and MECPTP and ∑DBP were