Effects of prenatal single and mixed bisphenol exposure on bone mineral density in preschool children: A population-based prospective cohort study

Exposure to bisphenols can affect bone mineral density (BMD) in animals and humans. However, the effects of maternal exposure to bisphenols during pregnancy on bone health in preschool children remain unknown. We aimed to assess the effects of prenatal exposure to single and multiple bisphenols on bone health in preschool children. A total of 230 mother-child pairs were included in this study. Generalized linear regression, restricted cubic spline (RCS), principal component analysis (PCA), and Bayesian kernel machine regression (BKMR) were utilized to assess the relationship between bisphenol levels and bone health in preschool children. Each natural log (Ln) unit increase in tetrabromobisphenol A was related to a 0.007 m/s (95 % CI: -0.015, 0.000) decrease in Ln-transformed speed of sound (SOS) among girls. Decreased BMD Z scores in preschool children were found only in the high bisphenol S exposure group (beta =-0.568; 95 % CI: -1.087, -0.050) in boys. The risk of low BMD (BMDL) was significantly higher in the middle-exposure group (OR = 4.695; 95 % CI: 1.143, 24.381) and high-exposure group of BPS (OR = 6.165, 95 % CI: 1.445, 33.789) compared with the low-exposure group in boys. In girls, the risk of BMDL decreased with increasing bisphenol A concentration (OR = 0.413, 95 % CI: 0.215, 0.721). RCS analysis revealed a U-shaped nonlinear correlation between BPB concentration and BMDL in girls (P-overall = 0.011, P-nonlinear = 0.009). In PCA, a U-shaped dose-response relationship was found between PC2 and the risk of BMDL (P-overall = 0.048, P-nonlinear = 0.032), and a significant association was only noted in girls when stratified by sex. The BKMR model revealed a horizontal S-shaped curve relationship between bisphenol mixtures and BMDL in girls. The results indicated that prenatal exposure to single and mixed bisphenols can affect BMD in preschool children, exerting nonmonotonic and child sex-specific effects.