The association between blood pressure and whole blood methylmercury in a cross-sectional study among Inuit in Greenland

BACKGROUND The Inuit in Greenland have a high average consumption of marine species and are highly exposed to methylmercury, which in other studies has been related to hypertension. Data on the relation between methylmercury and hypertension is limited, especially in populations subjected to a high exposure of methylmercury. We examined the relation between whole blood mercury and blood pressure (BP) in Inuit in Greenland. METHODS A cross-sectional population-based study among adult Inuit in Greenland was performed in 2005-2009. Information on socio-demography, lifestyle, BP, blood samples and clinical measurements was obtained - the latter after overnight fasting. BP was measured according to standardized guidelines. Whole blood mercury concentration was used as a marker of exposure. The analyses were restricted to Inuit aged 30-69 years with four Greenlandic grandparents (N = 1,861). Multivariate regression analyses with inclusion of confounders were done separately for men and women with the omission of participants receiving anti-hypertensive drugs, except for logistic regression analyses of the relation between mercury and presence of hypertension (yes/no). RESULTS The mean whole blood mercury level was 20.5 μg/L among men and 14.7 μg/L among women. In multivariate analyses adjusted for confounders, diastolic BP decreased with increasing mercury concentration. In men diastolic BP decreased significantly for each four-fold increase in mercury concentration (Beta = -0.04, standard error = 0.01, p = 0.001), while no relation between mercury and diastolic BP was found among women. For systolic BP, a similar non-statistically significant result was seen only for men (Beta = -0.02, standard error = 0.01, p = 0.06). A relation between mercury and hypertension was only found in men; the odds ratio for hypertension was 0.99 (95% CI: 0.98-0.99). No relation between quintiles of mercury and hypertension was found. The relationship between mercury and BP parameters may be non-linear: In analyses of quintiles of mercury the overall effect of mercury on BP parameters was only statistically significant for diastolic BP among men (Wald test, p = 0.01), however pairwise comparisons showed that some quintiles were not statistically different. This result is supported by LOESS modelling. CONCLUSIONS No adverse associations between whole blood mercury and blood pressure were found. With increasing whole blood mercury concentrations, diastolic BP and the risk of hypertension decreased among men in the study: this may be explained by confounding by exercise or unknown factors.