Serum Bisphenol A, glucose homeostasis, and gestational diabetes mellitus in Chinese pregnant women: a prospective study.

Lab studies have suggested that exposure to Bisphenol A (BPA) could disturb glucose homeostasis, but epidemiologic studies are limited and show inconsistent results for pregnant women. For this, 535 pregnant women were selected from a pregnant women cohort established in Tangshan City in North China between 2013 and 2014. Serum concentrations of BPA were measured in the early term of pregnancy, and fasting glucose and insulin levels were repeatedly measured in each of three terms of pregnancy (early, middle, and late). Gestational diabetes mellitus (GDM) were examined by Oral Glucose Tolerance Test (OGTT) in the middle and late terms of pregnancy. BPA was detected in 97.5% of pregnant women with a median of 6.50 ng/ml. Natural log-transformed BPA (Ln BPA) was positively associated with fasting glucose level (β (95% CI): 0.038 (0.015~0.061)), fasting insulin level (0.195 (0.069~0.321)), and homeostasis model insulin resistance index (HOMA-IR) (0.226 (0.087~0.364)) in the middle term of pregnancy by multiple linear regression model after adjusting for potential confounders. After serum BPA levels were divided into three groups (low, middle, and high), BPA showed a positive dose-response relationship with blood glucose, insulin, and HOMA-IR in the middle term of pregnancy. Increased BPA concentration tended to increase the RR of GDM although not statistically significant (risk ratio: 2.51 (95% CI: 0.68~9.30) for high vs low tertile of BPA concentrations). These findings suggested that exposure to BPA might affect glucose homeostasis and the middle term of pregnancy was a potentially sensitive period.