Hormone Disruption, Bone Mineral Density, and Persistent Organic Pollutants in Postmenopausal Women

Background: Persistent organic pollutants (POPs), such as polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins (PCDDs), and polychlorinated dibenzofurans (PCDFs), are potential endocrine disruptors. Objectives: The aims of this research were to evaluate the associations of classification schemes of PCB, PCDD, and PCDF congeners with bone mineral density (BMD), hormones, and binding proteins in postmenopausal women who were not taking exogenous hormones. We hypothesized that associations of exposures with endpoints would be modified by factors affecting endogenous hormones. Methods: Cross-sectional analyses were conducted using data from two sources: the National Health and Nutrition Examination Survey (NHANES) and the Great Lakes Fish Consumption Study (GLFCS). Associations of POPs and endpoints were assessed using regression models or correlation coefficients. Effect modification and confounding were assessed using regression models. When evidence of interaction was found, conditional effects were estimated and reported. Results: Findings were not always consistent across studies, which may be attributed to differences between the NHANES and GLFCS participants. The GLFCS participants were primarily Caucasian, older, and leaner than NHANES participants. Fewer GLFCS participants were smokers and were taking diabetes medications, while a higher proportion consumed alcohol on a regular basis and took thyroid hormones C-reactive protein (CRP) and gamma-glutamyl transferase (GGT) levels of GLFCS participants were relatively low, while PCB levels were higher than in NHANES. There were inconsistent associations of POPs with BMD among 603 participants in the 1999–2004 NHANES. We found inverse associations of luteinizing hormone with exposure to anti-estrogenic and/or dioxin-like POPs among 89 participants in the 1999–2002 NHANES. The GGT, a potential confounder, attenuated these associations. We found inverse associations of follicle-stimulating hormone and sex hormone-binding globulin with exposure to estrogenic and/or non-dioxin-like PCBs among 77 women participating in the 2001–2005 GLFCS. Confounders varied across studies, but all models were minimally adjusted for age, body mass index, and lipids. Effects were generally observed in obese participants, and other effect modifiers varied across studies. Conclusions: Some findings of associations of POPs with BMD and gonadotropin hormone levels in postmenopausal women are consistent with estrogenic effects found in previous investigations. Other effects are novel and require further investigation.