Background: Increasing evidence shows that polycyclic aromatic hydrocarbons (PAHs) exposure may adversely affect human health. However, the links between combined exposure to PAHs and neuropsychiatric manifestations in perimenopausal women remain unclear.
Methods: To explore these relationships further, we used the data from the National Health and Nutrition Examination Surveys (NHANES) of the 2005-2012 cycles. After filtering, five hundred forty-seven perimenopausal women aged 45-55 years were included in our analysis. Eight PAHs metabolites were measured to represent PAHs exposure in the body. In our study, depression, sleep disorders, and frequent mental distress (FMD) were used to describe the neuropsychiatric manifestations. Because of the bivariate correlations among PAHs compounds, principal component analysis (PCA) was conducted to achieve the dimension reduction process of PAHs compounds. To figure out if there is a relationship between urinary PAH metabolites and outcomes, multiple logistic regression, restricted cubic splines (RCS), and the Bayesian kernel machine regression (BKMR) were used.
Results: The findings showed that urinary PAHs concentrations in a certain range were related to neuropsychiatric manifestations. In detail, the results of logistic regressions, RCS, and BKMR all indicated that urinary PAHs were positively correlated with depression. In addition, the results of principal components regression and RCS showed associations between urinary PAHs and the risk of FMD or sleep disorders, respectively.
Conclusions: Exposure to PAHs was linked to neuropsychiatric manifestations in perimenopausal women, but more pertinent researches are required to understand the connections fully.
Keywords: Bayesian kernel machine regression; Chemical mixture exposure; NHANES; Nonlinear.
Copyright © 2023. Published by Elsevier B.V.