Overview
This study used the 24-hour duplicate diet method to measure methylmercury (MeHg) and total mercury (THg) intake in 260 Japanese children aged 0–5 years and examined correlations with hair THg levels. The median estimated MeHg intake was 18.3 ng/kgbw/day (range 0.65–209.2), with 13.8% of children exceeding the US EPA reference dose and 4.3% exceeding the WHO provisional tolerable weekly intake. MeHg comprised 90% of total mercury in dietary samples with THg ≥1 ng/g, reflecting Japan’s high fish consumption even in infants. These findings highlight the importance of monitoring MeHg exposure during complementary feeding, as infants and toddlers are highly susceptible due to low body weight and high food intake per kilogram.
Key numbers
| Dietary Stage | Sample N | THg Median (ng/g-wet) | THg Range (ng/g-wet) | THg Median Intake (ng/kgbw/day) | THg Range (ng/kgbw/day) |
|---|---|---|---|---|---|
| Formula milk (0–4 m) | 32 | 0.020 | 0.015–0.099 | 2.4 | 0.5–6.9 |
| Baby food Stage 1 (5–6 m) | 30 | 0.085 | 0.017–0.974 | 4.2 | 0.5–14.0 |
| Baby food Stage 2 (7–8 m) | 38 | 0.150 | 0.020–2.905 | 7.0 | 0.6–103.3 |
| Baby food Stage 3 (9–11 m) | 38 | 0.445 | 0.181–5.500 | 25.0 | 8.5–492.8 |
| Baby food Stage 4 (12–17 m) | 35 | 0.500 | 0.134–3.620 | 41.0 | 7.9–329.6 |
| Toddler meal Stage 1 (18–35 m) | 33 | 0.510 | 0.191–3.630 | 32.4 | 11.0–259.9 |
| Toddler meal Stage 2 (36–47 m) | 33 | 0.490 | 0.204–4.260 | 36.0 | 9.5–252.9 |
| Toddler meal Stage 3 (48–71 m) | 37 | 0.450 | 0.192–1.900 | 25.2 | 8.4–125.6 |
| All stages combined | 276 | 0.34 | 0.02–3.53 | 20.3 | 0.72–232.5 |
Estimated MeHg intake (derived from THg × 90%):
| Dietary Stage | Estimated MeHg Median (ng/kgbw/day) | Range (ng/kgbw/day) | % Exceeding RfD (0.1 µg/kgbw/day) | % Exceeding PTWI (1.6 µg/kgbw/week) |
|---|---|---|---|---|
| Formula milk | 2.2 | 0.5–6.2 | — | — |
| Baby food Stage 1 | 3.8 | 0.4–12.6 | — | — |
| Baby food Stage 2 | 6.3 | 0.5–93.0 | — | — |
| Baby food Stage 3 | 22.5 | 7.6–443.5 | 26.3% | 7.9% |
| Baby food Stage 4 | 36.9 | 7.1–296.6 | 34.3% | 11.4% |
| Toddler meal Stage 1 | 29.2 | 9.9–233.9 | 30.3% | 18.2% |
| Toddler meal Stage 2 | 32.4 | 8.5–227.6 | 30.3% | 18.2% |
| Toddler meal Stage 3 | 22.7 | 7.5–113.0 | 18.9% | 8.1% |
| All participants (n=260) | 18.3 | 0.65–209.2 | 13.8% (38 children) | 4.3% (12 children) |
Hair THg concentrations:
| Dietary Stage | Median (ppm) | Range (ppm) | % with Hair THg >1.0 ppm (RfD equivalent) | % with Hair THg >2.2 ppm (PTWI equivalent) |
|---|---|---|---|---|
| Formula milk | 1.2 | 0.4–3.9 | 59.4% | 15.6% |
| Baby food Stage 1 | 1.0 | 0.3–2.5 | 46.7% | 6.7% |
| Baby food Stage 2 | 0.6 | 0.2–1.4 | 23.7% | 2.6% |
| Baby food Stage 3 | 0.7 | 0.3–1.9 | 34.2% | 0.0% |
| Baby food Stage 4 | 1.0 | 0.3–4.4 | 51.4% | 20.0% |
| Toddler meal Stage 1 | 1.4 | 0.6–2.8 | 78.8% | 21.2% |
| Toddler meal Stage 2 | 1.3 | 0.3–6.0 | 63.6% | 30.3% |
| Toddler meal Stage 3 | 1.3 | 0.4–4.5 | 62.2% | 29.7% |
| Overall (n=260) | 1.05 | 0.31–3.96 | 51.8% (143 children) | 15.6% (43 children) |
MeHg speciation in samples with THg ≥1 ng/g (n=63):
- MeHg median concentration: 1.70 ng/g (range 0.87–6.21 ng/g)
- MeHg as percentage of THg: 90.0% (range 67.4%–99.1% across stages)
- MeHg was not detected in 213 samples where dietary THg <1 ng/g
Regulatory limits cited:
- US EPA Reference Dose (RfD) for MeHg: 0.1 µg/kgbw/day
- WHO Provisional Tolerable Weekly Intake (PTWI): 1.6 µg/kgbw/week (revised 2003 from 3.3 µg/kgbw/week in 1972)
- Hair THg equivalent to RfD: 1.0 µg/g
- Hair THg equivalent to PTWI: 2.2 µg/g
Methods
Dietary samples were collected using the WHO-recommended 24-hour duplicate diet method over 3 consecutive days from 260 children (subdivided into 276 diet collections when accounting for age transitions). All foods and beverages consumed were collected in pre-cleaned polypropylene containers and stored under refrigeration. Samples were homogenized, lyophilized, and analyzed for THg by cold vapor atomic absorption spectrometry (CVAAS, Hg-201, Sanso Inc.) with MDL/MQL of 0.03/0.08 ng, and for MeHg by gas chromatography-electron capture detector (GC-ECD, G2700, Yanaco Analytical Systems) with MDL/MQL of 0.18/0.45 ng. Hair samples (3 cm from root, occipital region) were analyzed for THg by direct thermal decomposition-gold amalgamation-CVAAS (MA-3000). Quality control used certified reference materials with recovery rates of 88–92% (THg) and 98–102% (MeHg). Values below MDL were substituted at half the MDL.
Implications
This study presents the first comprehensive assessment of postnatal MeHg exposure in Japanese infants and toddlers using the gold-standard duplicate diet method. The finding that 90% of dietary mercury in fish-containing meals is methylated underscores the importance of fish consumption guidance in complementary feeding. The convergence of WHO and US EPA thresholds in exposure percentages—with 13.8% exceeding the RfD—indicates that current Japanese infant feeding guidelines, which recommend iron-rich fish from 9 months onward, warrant reconsideration in light of neurotoxicological risk, particularly given the heightened vulnerability of developing brains to MeHg. Hair THg measurements show only weak correlation with dietary MeHg intake at early feeding stages, likely due to placental transfer during fetal life, highlighting the necessity of careful interpretation of hair biomarkers in infants. Further research using both dietary assessment and blood biomarkers is essential to characterize the true burden of postnatal MeHg exposure and its relationship to infant neurodevelopment.
Wiki pages updated on ingest
- MeHg
- tHg
- Fish
- tuna
- dairy
- milk
- breast-milk
- Baby Food
- epa-reference-dose-methylmercury
- who-ptwi-methylmercury
Update history
The five most recent substantive edits to this page, classified major (evidence or structure moved), correction (a published value or statement was wrong and has been fixed), or minor (narrative rewritten without changing the underlying evidence). Each description is derived from what the edit did to this page; the linked commit is the authoritative record, routine regeneration passes are excluded, and the full version history lives in git. When DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.