Overview
Hoen and colleagues examined whether urinary arsenic at 6 weeks of age related to infant gut-microbiome composition in the New Hampshire Birth Cohort Study. This is a4 biomonitoring and microbiome evidence: it does not report infant-formula product concentrations, but it does provide human infant exposure-biomarker data and sex-specific microbiome associations in a formula-relevant early-life population.
Key numbers
The study analyzed 204 infants with both 6-week urine and stool samples. Table 1 reports urinary arsenic concentrations in infants of median 0.4 micrograms/L, mean 0.63 micrograms/L, and range <0.05-4.8 micrograms/L. Home tap-water arsenic had mean 1.5 micrograms/L and range <0.05-57.0 micrograms/L.
Feeding-stratified urinary arsenic means reported in the Results were 0.5 micrograms/L for exclusively breastfed infants, 0.9 micrograms/L for mixed-fed infants, and 1.2 micrograms/L for formula-only infants.
Overall microbiome composition at 6 weeks was marginally associated with natural-log urinary arsenic after adjustment for feeding and urine specific gravity (pseudo-F 1.85, p = 0.05). The authors identified 8 OTUs positively associated with urinary arsenic and 14 OTUs negatively associated with urinary arsenic. Positively associated taxa included a strong positive association with Ruminococcus; negatively associated taxa included OTUs assigned to Bacteroides, Bifidobacterium, and Clostridiaceae.
Sex- and feeding-stratified analysis found the strongest association in formula-fed males. Among exclusively or partially formula-fed males (n = 33), overall microbiome composition was associated with urinary arsenic after adjustment for formula exclusivity and urine specific gravity (pseudo-F 2.30, p = 0.008). No such association was observed in formula-fed females (n = 23, pseudo-F 0.54, p = 0.93), exclusively breastfed males (n = 85, pseudo-F 0.58, p = 0.83), or exclusively breastfed females (n = 58, pseudo-F 1.14, p = 0.39).
Among formula-fed males, urinary arsenic was associated with 9 OTUs after adjustment, with 2 positively associated and 7 negatively associated. Negatively associated taxa included Bifidobacterium, Bacteroides, and Lactobacillus; positively associated taxa included a Lachnospiraceae OTU and a Streptococcus OTU.
The cohort discussion notes that popular infant-formula brands in prior work by the same group contained 2.2-12.6 ng/g arsenic, most of it inorganic, and that drinking water used to mix formula contributed additional infant exposure.
Methods (brief)
Infant urine at 6 weeks was analyzed for total arsenic concentration, and stool microbiome composition was measured by sequencing of bacterial 16S rRNA gene regions from DNA extracted from stool. The current-study inclusion set came from the New Hampshire Birth Cohort Study and required urine, stool, and feeding data at 6 weeks. Statistical models used natural-log urinary arsenic, generalized UniFrac distance, feeding adjustment, and urine specific gravity adjustment; subgroup analyses were stratified by sex and feeding mode.
Implications
Certification: Do not use this source as product-occurrence evidence for infant formula. It provides infant urinary biomonitoring and microbiome-association evidence rather than formula-powder or ready-to-feed concentrations.
Courses: Useful for showing that early-life arsenic exposure can be detected at low urinary concentrations in a US cohort and that formula-fed male infants showed the clearest microbiome signal.
App: Context only. This paper supports infant exposure and microbiome-risk framing, not contamination profiling of a specific market product.
Microbiome: Primary human infant source for arsenic-gut-microbiome association, with strongest signal in formula-fed males.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.