Association between exposure to persistent organic pollutants and mercury, and glucose metabolism in two Canadian Indigenous populations

International audienceBackground: The body burden of metals and persistent organic pollutants (POPs) is particularly high in populations that rely on fish and other marine species for sustenance. This exposure has been associated with an increased risk of type 2 diabetes, but results remain contrasted.Objective: We studied this association in two Indigenous populations of northern Québec (Canada) with markedly different prevalences of diabetes and levels of exposure to POPs and mercury.Methods: As part of health surveys conducted in 2004-2009, diabetes prevalence and glucose metabolism (glucose, insulin, HOMA-IR, HOMA-B) in non-diabetic fasting adults were assessed using similar protocols in two populations: Inuit from Nunavik (n = 877) and Cree from Eeyou Istchee territory (n = 780). Blood mercury, plasma polychlorinated biphenyls (PCBs), organochlorine (OC) pesticides/metabolites and polybrominated diphenylethers (PBDEs) levels were measured in samples collected at the time of examination. Logistic and linear regressions and restricted cubic splines analyses were conducted adjusting for sex, age, waist circumference, smoking and omega-3 fatty acid content in plasma phospholipids.Results: Diabetes prevalence was higher in Cree (20%) than in Inuit (7%), whereas environmental exposure was 2 to 3-fold greater in Inuit than in Cree participants. In the range of exposure common to the two populations, we observed similar linear increases in the risk of diabetes with increasing contaminant exposure. Among Cree participants, fasting glucose was positively associated with plasma PBDE level, and HOMA-B negatively associated with concentrations of ∑PCBs, dichlorodiphenyldichloroethylene, PBDEs and ∑OC pesticides. Among Inuit participants, a trend towards reduced insulin secretion was observed in association with most contaminants, but the relation was nonlinear (greater reduction at intermediate levels of exposure). A significant increase in fasting glucose levels was observed at elevated blood mercury levels (>16 μg/L).Conclusion: The observed association between POPs exposure and diabetes risk in the two populations studied should be confirmed using prospective design. Our results suggest the need for additional research on the physiopathological process through which POPs exposure may induce type 2 diabetes in these Indigenous populations