Estimated Exposure to Arsenic in Breastfed and Formula-Fed Infants in a United States Cohort

Background: Previous studies indicate that concentrations of arsenic in breast milk are relatively low even in areas with high drinking-water arsenic. However, it is uncertain whether breastfeeding leads to reduced infant exposure to arsenic in regions with lower arsenic concentrations. Objective: We estimated the relative contributions of breast milk and formula to arsenic exposure during early infancy in a U.S. population. Methods: We measured arsenic in home tap water (n = 874), urine from 6-week-old infants (n = 72), and breast milk from mothers (n = 9) enrolled in the New Hampshire Birth Cohort Study (NHBCS) using inductively coupled plasma mass spectrometry. Using data from a 3-day food diary, we compared urinary arsenic across infant feeding types and developed predictive exposure models to estimate daily arsenic intake from breast milk and formula. Results: Urinary arsenic concentrations were generally low (median, 0.17 μg/L; maximum, 2.9 μg/L) but 7.5 times higher for infants fed exclusively with formula than for infants fed exclusively with breast milk (β = 2.02; 95% CI: 1.21, 2.83; p < 0.0001, adjusted for specific gravity). Similarly, the median estimated daily arsenic intake by NHBCS infants was 5.5 times higher for formula-fed infants (0.22 μg/kg/day) than for breastfed infants (0.04 μg/kg/day). Given median arsenic concentrations measured in NHBCS tap water and previously published for formula powder, formula powder was estimated to account for ~ 70% of median exposure among formula-fed NHBCS infants. Conclusions: Our findings suggest that breastfed infants have lower arsenic exposure than formula-fed infants, and that both formula powder and drinking water can be sources of exposure for U.S. infants. Citation: Carignan CC, Cottingham KL, Jackson BP, Farzan SF, Gandolfi AJ, Punshon T, Folt CL, Karagas MR. 2015. Estimated exposure to arsenic in breastfed and formula-fed infants in a United States cohort. Environ Health Perspect 123:500–506; http://dx.doi.org/10.1289/ehp.1408789