Associations between repeated measurements of urinary polycyclic aromatic hydrocarbon metabolites and thyroid hormones among reproductive-aged men

Abstract

BackgroundPolycyclic aromatic hydrocarbons (PAHs) have been shown to disrupt thyroid function in toxicological studies, but epidemiological evidence is inconsistent. Furthermore, little is known on potential effects of mixtures of PAHs. ObjectiveThis study aimed to examine the associations of exposure to PAHs as individual chemicals and mixtures with thyroid hormones. MethodsWe included 378 men from a Reproductive Medicine Center in Wuhan, China. Ten monohydroxylated PAH (OH-PAH) metabolites in repeated urine specimens collected at two-time points and three thyroid hormones [thyroid-stimulating hormone (TSH), free thyroxine (FT4), and free triiodothyronine (FT3)] in one serum sample were measured. Multivariable linear regression models were applied to assess the associations between individual OH-PAH metabolites and thyroid hormones, and the associations with mixtures of OH-PAH metabolites were assessed by Bayesian kernel machine regression (BKMR) models. ResultsMultivariable linear regression models showed inverse associations between urinary 1-OHNa and TSH, between urinary 1-OHPh and 9-OHPh and FT3, as well as between urinary 2-OHPh, 3-OHPh, 9-OHPh and ∑OHPh and FT4, regardless of these individual OH-PAH metabolites modeled as continuous or tertile variables (e.g., −21.57 % in TSH; 95 % CI: −35.33 %, −4.88 % for the third vs first tertiles of 1-OHNa; p for trend = 0.014). BKMR models showed negative overall effects of all urinary OH-PAH metabolite mixtures on TSH, FT3, and FT4, and 1-OHNa, 9-OHPh, and 2-OHPh as the most important contributors, respectively, with linear inverse exposure–response associations when holding other OH-PAH metabolites at their median concentrations. ConclusionUrinary OH-PAH metabolites as individual chemicals and mixtures were adversely associated with thyroid hormones among reproductive-aged men.