Prenatal exposure to bisphenols and visual impairment in preschool children : a prospective birth cohort study in China

Abstract

Abstract Evidence from animal experiments suggests that exposure to bisphenols during early life may lead to impaired eye development and visual functions. However, population-based study on the association remains very limited. To investigate the relationships between prenatal bisphenols exposure and visual impairment in preschool children. A total of 744 mother-infant pairs were extracted from the Guangxi Zhuang Birth Cohort in China. Maternal serum bisphenol A (BPA) and its alternatives were measured by using ultra-high liquid performance chromatography-tandem mass spectrometer. Visual acuity in preschool children were followed up in the local maternal and child health information management system. The associations of prenatal exposure to bisphenols (BPs) with visual acuity were analyzed by multivariate linear regression models. Logistic regression model, Bayesian Kernel machine regression (BKMR), and quantile g-computation (Qgcomp) models were applied to examine the single and mixture effects of prenatal exposure to the five BPs on risk of visual impairment (VI).In logistic regression models, the moderate-level BPA-exposed group (OR = 0.474, 95%CI: 0.211, 1.065) and high-level TBBPA-exposed group (OR = 0.451, 95%CI: 0.184, 1.109) had a lower risk of VI than the low-level exposed group, both approaching a level of significance (P = 0.071 and P = 0.083, respectively). When stratified analysis by child sex, the significant associations of moderate-level BPA exposure and VI risk were only found among boys (adjusted OR = 0.230, 95%CI: 0.061, 0.873, P = 0.031). And a suggestive negative association of high-level TBBPA exposure and VI risk was only found among girls (adjusted OR = 0.330, 95%CI: 0.091, 1.193, P = 0.091). Further analysis by using BKMR and g-computation models showed that mixed effects of the five BPs were also associated with decreased risk of VI, with effects-driven primarily by BPA and TBBPA for boys and girls, respectively. Findings from this study do not support the hypothesis that prenatal BPs exposure is associated with increased risk of VI. Further epidemiological studies remain warranted when confirming their associations.