Urinary concentrations of early and mid-pregnancy parabens and gestational diabetes: A nested case-control study within the PETALS cohort

Abstract

Parabens are widely used preservatives with endocrine-disrupting properties, but their role in glucose metabolism during pregnancy is unclear. This study examines prospective associations between urinary concentrations of four parabens in early and mid-pregnancy and gestational diabetes (GDM). A matched case-control study nested within a diverse longitudinal pregnancy cohort (PETALS) with universal GDM screening matched GDM cases to two controls (111 cases; 222 controls). Urine samples collected 2015-2017 in early (14 ± 2.3 weeks) and mid-pregnancy (20 ± 2.4 weeks) were analyzed for paraben concentrations with mass spectrometry. Area-under-the-time-concentration-curve (AUC) assessed cumulative exposure. Conditional logistic regression models evaluated associations between paraben concentrations and GDM, adjusting for covariates. We a priori examined effect modification by Asian/Pacific Islander (A/PI) race/ethnicity due to the case-control matching and GDM prevalence highest among A/PI. Participants were 31 ± 5 years old and 40 % A/PI, 33 % Hispanic, 14 % White and 9 % Black. Methylparaben and propylparaben had >94 % detection, while ethylparaben and butylparaben ranged from 22 %-51 %. Paraben exposure was not associated with GDM overall. Among A/PI, higher methylparaben concentrations exhibited higher odds of GDM: early-pregnancy OR 1.14 per IQR (95 % CI: 0.89,1.45) and AUC 1.07 (0.89,1.30) compared to non-A/PI (early-pregnancy 0.81 [0.62,1.06] and AUC 0.70 [0.44,1.12]; P_interaction = 0.01 and 0.03, respectively). A/PI mid-pregnancy ethylparaben exposure (detectable vs non-detectable) was linked to higher GDM odds (2.00 [0.84,4.76] vs. non-A/PI 0.47 [0.17,1.27]; P_interaction = 0.04) as was mid-pregnancy propylparaben exposure (Tertile 2 vs. 1: 3.67 [1.21,11.1] vs. non-A/PI 0.70 [0.22, 2.25]; P_interaction = 0.04). Although overall paraben exposure was not associated with GDM, interactions by A/PI race/ethnicity suggested potential increased odds of GDM related to propylparaben, methylparaben, and ethylparaben exposure. Future studies should explore paraben exposure in diverse populations.

Keywords: EDCs; Endocrine disrupting chemicals; GDM; Gestational diabetes; Parabens; Pregnancy.