Exposure to mercury among Spanish adolescents: Eleven years of follow-up

The toxic effects of mercury exposure on human health are a public health concern. The most important source of this exposure is the consumption of fish and marine mammals. This study aims to describe hair mercury con-centrations and their evolution from birth until eleven years of age in adolescents from the INMA (Environment and Childhood) birth cohort study, and to assess the association of hair mercury concentrations at eleven years of age with sociodemographic and dietary factors. The sample comprised 338 adolescents from the sub-cohort of Valencia (in eastern Spain). Total mercury (THg) was measured in hair samples collected at 4, 9 and 11 years old and in cord blood at birth. The equivalent of hair for cord-blood THg concentrations was calculated. Fish consumption and other characteristics at 11 years old were collected through questionnaires. Multivariate linear regression models were conducted to explore the association between THg concentrations, fish consumption and covariates. The geometric mean of hair THg concentrations at 11 years of age was 0.86 & mu;g/g (95%CI: 0.78-0.94) and 45.2% of the participants presented concentrations above the equivalent RfD proposed by the US EPA (1 & mu;g/g). Consumption of fish such as swordfish, canned tuna and other large oily fish was associated with higher levels of hair mercury at 11 years of age. Swordfish had the highest effect with an increase of 125% in hair mercury (95% CI: 61.2-214.9%) given a 100 g/week increase in its consumption, and, taking into account the frequency of consumption, canned tuna was the main contributor to Hg exposure among our population. The hair THg concentrations at 11 years of age represented a reduction of around 69% with respect to that estimated at childbirth. Even though THg exposure shows a sustained decreasing trend, it can still be considered elevated. INMA birth cohort studies provide a longitudinal assessment of mercury exposure in a vulnerable population, its associated factors and temporal trends, and this information could be used to adjust recommendations about this issue.