Jackson, Taylor, Punshon, Cottingham 2012 — Arsenic concentration and speciation in infant formulas and first foods
This IUPAC Pure and Applied Chemistry paper from Dartmouth’s Trace Element Analysis Laboratory and Department of Biological Sciences (Jackson, Taylor, Punshon, Cottingham) reports total arsenic and arsenic speciation (inorganic arsenic plus methylated species) in U.S. infant formulas and first foods using HPLC-ICP-MS speciation methodology. The paper is one of the few primary sources with sample-level speciation data for infant-formula matrices and is directly applicable to the milk-based powdered infant formula iAs gap that the IandC overnight final report flagged as the highest-priority Phase 3b ingest target. The Dartmouth group is the U.S. center of expertise for infant arsenic speciation; this paper and the related Carignan 2015, Carignan 2016, and Pikounis-Cottingham companion papers form the Dartmouth infant-As evidence cluster. A NIH preprint version of this paper is also present in raw/Digest as nihms374391.pdf.
Key conclusions
The paper documents that infant formulas produced from rice-derived components carry substantially higher inorganic arsenic than formulas without rice components. Total arsenic and inorganic arsenic distributions are reported per matrix; the speciation breakdown distinguishes the inorganic species (arsenite + arsenate, the regulated and toxicologically high-concern fraction) from the methylated organic species (MMA, DMA — lower toxicological concern). The findings provide the U.S. evidence base for the FDA Closer-to-Zero infant rice cereal iAs action level development and inform the formula-specific iAs assessment that the wiki has historically captured only at category-level rather than sample-level.
Key numbers
Infant formulas (n=15 across 5 main brands)
Total As ranged from 2.2 to 12.6 ng/g; non-dairy formulas were significantly higher than dairy-based (nested ANOVA F1,13 = 13.3, P = 0.003). Speciation in formulas with total As > 6 ng/g (n=9) showed essentially 100 percent inorganic arsenic, with As(V) the dominant inorganic species.
| Brand | Formula | Total As ng/g (mean ± SE) | Dairy | Rice |
|---|---|---|---|---|
| A | non-dairy | 11.27 ± 0.35 | NO | NO |
| A | non-dairy | 9.29 ± 0.43 | NO | NO |
| A | non-dairy + rice | 11.89 ± 0.64 | NO | YES |
| A | dairy | 5.36 ± 0.21 | YES | NO |
| B | non-dairy | 11.43 ± 1.09 | NO | NO |
| B | non-dairy | 6.95 ± 0.43 | NO | NO |
| B | dairy + rice | 8.19 ± 0.63 | YES | YES |
| B | dairy + rice | 6.02 ± 0.26 | YES | YES |
| B | dairy | 5.76 ± 0.4 | YES | NO |
| C | dairy | 8.14 ± 0.77 | YES | NO |
| D | dairy | 9.38 ± 0.31 | YES | NO |
| D | dairy | 2.92 ± 0.33 | YES | NO |
| E | non-dairy | 9.62 ± 1.35 | NO | NO |
| F | dairy | 3.42 ± 0.2 | YES | NO |
| F | dairy | 2.6 ± 0.44 | YES | NO |
Exposure assessment
A 3-month-old, 6.2 kg infant consuming six 120 mL formula bottles daily is exposed to 0.036 to 0.21 µg As/kg/day from formula alone. The upper end of this range exceeds the 0.17 µg/kg/day reference derived from an adult drinking 1 L of water at the WHO/EPA 10 µg/L MCL, indicating that infants on the higher-As formulas exceed adult drinking-water exposure before any complementary food is introduced.
Fruit and vegetable purees (n=41, brands D, E, G)
Most non-pear purees ranged 0.32 to 7.81 ng/g. Brand E pear-containing products carried distinctly elevated As (15.01 to 20.20 ng/g across pears and mango, pears alone, pears and raspberries), driven by source-specific contamination rather than a general pear property. Pear-product As speciation: 76 to 83 percent inorganic As, DMA the secondary species.
Stage 2/3 mixed-meal foods (n=18)
Significant meat × rice interaction (F1,37.25 = 14.42, P = 0.0005). Least-squares means:
| Subcategory | As ng/g (LS mean) |
|---|---|
| Neither meat nor rice | 3.75 |
| Meat without rice | 2.34 |
| Rice without meat | 11.90 |
| Meat and rice (highest) | 18.45 |
Speciation in foods with As > 5 ng/g was predominantly inorganic As (>70 percent), notable because U.S. rice typically has DMA as the dominant species. The inorganic-As dominance in these mixed-meal foods suggests rice bran inclusion (rice bran carries higher inorganic As than polished grain).
Implications
- Certification: Direct primary-evidence input for HMTc milk-based and soy-based infant formula iAs row distributions. Speciation-preserving methodology means iAs values are not contaminated by total-arsenic over-inclusion. Suitable for HMTc benchmark-pool admission for the iAs cells in the IandC subcategory matrix.
- Courses: Standard reference for infant-arsenic speciation methodology.
- App: Supports per-formula-type iAs risk profiles, distinguishing rice-derived-component formula products from non-rice-component products.