Mercury Exposure in Pregnancy and Diagnosed Autism and Autistic Traits in the Offspring: Results from a Prospective Birth Cohort

Background: For several decades mercury has been suspected of causing autism, leading to alarm over thiomersal-containing vaccinations and eating fish during pregnancy. Reviews of the literature have found few robust studies. The only prospective longitudinal study, based in the Seychelle islands, found no association between prenatal mercury levels and an autistic trait measure. Numbers were too small to assess relationships with diagnosed autism. Thus there is a need to determine whether there is prospective evidence of links between fetal exposure to mercury and autism or autistic traits.

Objectives:  To determine the association between 1) prenatal exposure to mercury and 2) indirect measures of mercury levels during pregnancy, and the risk of diagnosed autism and of autistic traits in the offspring.

Methods: In the Avon Longitudinal Study of Parents and Children (ALSPAC), a large prebirth cohort in England, we used three strategies: (1) direct comparison of prenatal mercury levels in 48 cases of autism with 3837 controls; (2) comparison of prenatal mercury levels in those with high scores on each of four autistic traits with the rest of the population at risk (n~2800); (3) indirect measures of association of these outcomes with proxies for increased mercury levels such as frequency of fish consumption, and exposure to dental amalgam (n > 8000). We used logistic regression models adjusted for various potential confounders including maternal age, housing circumstances, maternal education and parity.

Results: The mean maternal whole blood mercury in the pregnancies resulting in an offspring with diagnosed autism was 2.10µg/L [SD 0.95] compared with 2.08[1.09]µg/L for the rest of the population. There was no evidence of an adverse effect of rising mercury levels on diagnosed autism [adjusted odds ratio (AOR) 0.89; 95% CI 0.65, 1.22] per SD of mercury (P = 0.485) or on any of the autistic trait measures. The proxy measures of exposure to mercury reported by the mothers also gave no indication of adverse associations with any of the autistic traits or of diagnosed autism.

Conclusions: There was no evidence that maternal prenatal blood mercury levels, or proxies for increased mercury exposure were associated with autism or autistic traits in this large population based study with prospective data.