Table 1.
Major cohort studies examining early-life methylmercury (MeHg) and total mercury (Hg) exposure and neurodevelopment in children. IQR, inter-quartile range (25th to 75th percentile).
| Population | Study sample | Measure of exposure | Average exposure (ppm) | Neurological associations |
|---|---|---|---|---|
| Faroe Islands (Grandjean et al., 1997; Grandjean et al., 2014) | 1022 singleton births, 917 children at age 7 | Hg concentrations in maternal hair at delivery, cord blood, child blood and hair at age 7 years | Geometric mean and IQR at 7 years: hair Hg 3.03 (1.68–6.33), maternal hair Hg in pregnancy: 4.35 (2.63–42.2) | Neurodevelopmental deficits (i.e. visuospatial memory) at birth and early school years when comparing high and low exposure groups |
| Italy (Deroma et al., 2013) | 149 children | Total Hg and MeHg in maternal hair and breast milk and child's hair at 7–9 years | Median maternal hair Hg (total): 1.38 | Children with high prenatal Hg exposure had lower verbal, scale and performance IQ than children with low prenatal Hg exposure, but this difference was not significant. In contrast, children's fresh fish consumption was positively associated with scale and performance IQ |
| Italy (Valent et al., 2013) | 606 children at 18 months of age | Maternal and child fish intake; total Hg in maternal hair and blood during pregnancy, umbilical cord bood, and breast milk | Mean maternal hair Hg: 1.06 | No evidence of prenatal Hg exposure linked to children's neurodevelopment. Children's fish intake, but not maternal PUFAs (EPA, DHA and other fatty acids), were positively associated with neurodevelopmental test scores |
| United States – Massachusetts (Oken et al., 2005) | 135 infant-mother pairs | Self reported Fish consumption during 2nd trimester of pregnancy, maternal total Hg in hair at delivery | Mean maternal hair Hg: 0.55 (range 0.02–2.38) | Increased maternal fish intake during pregnancy associated with increased infant cognition at 6 months of age. This association was stronger after adjusting for maternal hair Hg at delivery. Higher Hg levels were associated with lower infant cognition at 6 months of age |
| Seychelles (Davidson et al., 1998; Myers et al., 2003; Myers et al., 2009) | Seychelles Child Development Study Main Cohort: 770 mother-child pairs (children through 107 months) | MeHg exposure (measured as total Hg in hair) from maternal hair, and children's hair at 66 and 107 months | Mean maternal hair Hg: 6.8 Mean child hair-Hg at 66 months: 6.5 (sd: 3.3); at 107 month: 6.1 (sd: 3.6) |
Hg not consistently associated with neurodevelopmental outcomes |
| Seychelles (Strain et al., 2015) | Seychelles Child Development Study Nutrition Cohort 2: 1265 mother-child pairs (children at age 20 months) | Total Hg in maternal hair at delivery and maternal weekly fish consumption | Mean maternal hair Hg: 3.92 (sd. 3.46) Maternal estimate of weekly fish meals: 8.52 (4.56) |
No overall association of Hg with neurodevelopment, but evidence for possible interaction of Hg with fish oils for neurodevelopment: higher levels of Hg were negatively associated with psychomotor scores for children of mothers with higher ratio of n-6 to n-3 fatty acids; whereas higher Hg was positively associated with psychomotor development among children born to mothers with higher n-3 fatty acids |
| Seychelles (Davidson et al., 2008) | 300 mothers and 229 children at ages 5, 9, 25 and 30 months | Number of fish meals per week of mother during pregnancy | Mean maternal hair MeHg: 5.9 | Neurodevelopmental performance at 30 months decreased with increased MeHg, adjusted for nutritional factors |
| Tohoku, Japan (Tatsuta et al., 2014) | 387 42-month old children | Cord blood total Hg levels | Median cord blood Hg: 0.01 | No significant correlations between neurodevelopmental score and total mercury |
| New Zealand, North Island (Crump et al., 1998 re-analysis of Kjellström et al., 1986; Kjellström et al., 1989) | 237 children ages 6–7 (paired with their mothers) | Average maternal hair Hg concentration during pregnancy | 61 children with hair Hg > 6 ppm matched to lower-Hg-exposed children. Crump et al. use continuous hair Hg measures, but do not report average Hg value | Negative association of maternal hair Hg with academic attainment, language development, fine and gross motor coordination, and intelligence – after omitting one highly influential point from the analysis |
IQR, inter-quartile range (25th and 75th percentiles of distribution).