Associations of polychlorinated biphenyl exposure and endogenous hormones with diabetes in post-menopausal women previously employed at a capacitor manufacturing plant

There is an increasing body of literature showing associations of organochlorine exposure with risk of diabetes and insulin resistance. Some studies suggest that associations differ by gender and that diabetes risk, in turn, may be affected by endogenous steroid hormones. This report examines the relationships of serum PCBs and endogenous hormones with history of diabetes in a cohort of persons previously employed at a capacitor manufacturing plant. A total of 118 women were post-menopausal with complete data, of whom 93 were not using steroid hormones in 1996, at the time of examination, which included a survey of exposure and medical history, height, weight and collection of blood and urine for measurements of lipids, liver function, hematologic markers and endogenous hormones. This analysis examines relationships of serum polychlorinated biphenyls (PCBs), work exposure and endogenous hormones with self-reported history of diabetes after control for potential confounders. All PCB exposure groups were significantly related to history of diabetes, but not to insulin resistance as measured by the homeostatic model assessment of insulin resistance (HOMA-IR) in non-diabetics. Diabetes was also independently and inversely associated with follicle stimulating hormone (FSH), dehydroepiandrosterone sulfate (DHEAS) and triiodothyronine (T3) uptake. HOMA-IR was positively associated with body mass index (BMI) and C-reactive protein (CRP) and inversely associated with sex hormone binding globulin (SHBG) and T3 uptake after control for PCB exposure. Possible biologic mechanisms are discussed. This study confirms previous reports relating PCB exposure to diabetes and suggests possible hormonal pathways deserving further exploration.

► Occupational exposure to PCBs was associated with diabetes in post-menopausal women.
► Occupational exposure to PCBs was not associated with insulin resistance.
► Diabetes was inversely associated with selected endogenous hormones.
► Associations with diabetes were independent of potential confounders.