Flyvholm, Nielsen, Andersen 1984 — Nickel content of food and estimation of dietary intake
This Zeitschrift für Lebensmittel-Untersuchung und -Forschung review from the Danish National Food Institute and the Danish Toxicology Centre is the foundational primary-data source for nickel concentrations in foods and dietary nickel intake estimation. The authors compiled 2,221 food samples from the post-1969 literature (when AAS flame/flameless or PIXE methods began producing reliable Ni values) plus their own National Food Institute analyses, then modeled the Danish average diet (2,099 g/person/day exclusive of drinking water, 3,250 kcal). The headline number: 150 µg Ni/person/day from the Danish average diet, with drinking water adding another 16 µg/day. By replacing average-diet items with high-Ni foods, total intake can climb to 900 µg/day or more, well within the 600 to 5,600 µg single-dose range that provokes hand eczema flare in nickel-sensitive patients per oral challenge. The paper introduces the load factor F (ratio between percent of total Ni intake and percent of total consumption) as a way to identify which foods drive intake disproportionately to their dietary weight.
Key numbers
High-nickel foods not in the Danish average diet (µg/g, ordered by Ni content)
| Food | Range (µg/g) | n | Mean (µg/g) |
|---|---|---|---|
| Cocoa | 8.2-12 | 7 | 9.8 |
| Soy beans | 4.7-5.9 | 3 | 5.2 |
| Soy products | 1.08-7.8 | 7 | 5.1 |
| Walnuts | (single sample) | 1 | 3.6 |
| Peanuts | 1.6-4.9 | 2 | 2.8 |
| Oats | 0.33-4.8 | 37 | 2.3 |
| Buckwheat | 1.3-2.8 | 3 | 2.0 |
| Bitter chocolate | 1.3-2.7 | 7 | 1.9 |
| Hazelnuts | 0.66-3.3 | 12 | 1.8 |
| Dried legumes | 0.52-3.3 | 17 | 1.7 |
| Almonds | 1.0-1.5 | 5 | 1.3 |
| Pistachios | (single sample) | 1 | 0.8 |
| Milk chocolate | 0.4-1.2 | 11 | 0.7 |
| Prunes | 0.5-0.8 | 3 | 0.6 |
| Whole meal, white bread | 0.20-0.51 | 41 | 0.49 |
| Canned beans | 0.33-0.61 | 8 | 0.45 |
| Beans (frozen and fresh) | 0.15-1.0 | 23 | 0.40 |
| Whole wheat | 0.1-0.8 | 85 | 0.33 |
| Endive | 0.07-0.66 | 10 | 0.31 |
| Broccoli | 0.03-0.8 | 9 | 0.28 |
| Canned peas | 0.1-0.53 | 11 | 0.28 |
| White bread | 0.02-0.31 | 65 | 0.27 |
| Maize | 0.1-0.3 | 3 | 0.2 |
| Mushrooms | 0.02-0.80 | 19 | 0.16 |
| Rye bread | 0-0.61 | 30 | 0.13 |
Selected average-diet items with notable Ni content (µg/g, weighted means)
| Food | Mean (µg/g) | n | Daily intake contribution (µg/day) | Load factor F |
|---|---|---|---|---|
| Oatmeal | 1.76 | 15 | 14.1 | 24 |
| Cocoa-containing fats (margarine etc.) | 0.34 | 13 | 4.20 | 5.8 |
| Spinach | 0.36 | 2 | 1.44 | 5.0 |
| Strawberries | 0.13 | 9 | 0.25 | 1.0 |
| Wheat flour | 0.13 | 81 | 14.7 | 1.9 |
| Potatoes | 0.14 | 228 | 24.4 | 2.0 |
| Lettuce/salad | 0.36 | 2 | 0.40 | 3.0 |
| Leeks | 0.52 | 21 | 1.44 | 5.0 |
| Onions | 0.42 | 7 | 0.42 | 6.0 |
| Rice | 0.21 | 16 | 1.05 | 3.5 |
The load factor F is the ratio between a food’s percent of total Ni intake and its percent of total consumption; F > 1 indicates a food contributes Ni disproportionately to its dietary weight. Oatmeal at F = 24 is the standout outlier in the Danish average diet.
Dietary intake summary
| Compartment | Ni intake (µg/person/day) |
|---|---|
| Danish average diet (food only, 2,099 g/day exclusive of water) | 150 |
| Drinking water (1.8 L/day cold tap, 9 µg/L) | 16 |
| Drinking water (1.8 L/day after 8h tap stagnation, 490 µg/L peak) | up to 882 |
| Faecal Ni excretion (Danish vacuum-toilet study) | ~200 |
| Faecal + urinary Ni excretion (USA study) | 260 |
| Hand eczema oral provocation single dose range | 600-5,600 |
Sample-distribution caveats
Some foods are extensively measured (fish n=658, milk n=63, liver and kidney n=108) while others have fewer than 10 samples across the literature (cocoa, nuts, soya products). Mean values for low-n foods carry correspondingly wider uncertainty.
Methods (brief)
Literature search across Food Science and Technology Abstracts, Biological Abstracts, Chemical Abstracts, CAB, and Bulletin Signalétique, restricted to 1969-1982 (when AAS flame/flameless or PIXE methods produced reliable Ni values). Methods restricted to wet or dry digestion followed by AAS or PIXE; older methods excluded for specificity and interference concerns. The Danish average diet model was computerized from University of Aarhus statistical reports, supplemented with coffee/tea data, and used to compute Ni intake by multiplying mean food-content by consumption weight. The load factor F was derived per food item as percent-of-Ni-intake divided by percent-of-consumption.
Implications
- Certification: This is the foundational primary occurrence-data source for Ni in foods. Cocoa (9.8 µg/g), soy beans (5.2 µg/g), soy products (5.1 µg/g), walnuts (3.6 µg/g), peanuts (2.8 µg/g), oats (2.3 µg/g), bitter chocolate (1.9 µg/g), hazelnuts (1.8 µg/g), dried legumes (1.7 µg/g), and almonds (1.3 µg/g) are the priority commodities for HMTc Ni-threshold-setting. The load factor F flags oatmeal as the dominant per-gram contributor to dietary Ni in the average diet (F = 24). Whole-grain wheat, potatoes, and root vegetables also contribute meaningfully.
- Microbiome / clinical: Establishes the dietary Ni context for SNAS, low-Ni diet trials (Kaaber 1978, Veien 1993), and clinical-population studies. 150 µg/day Danish-average baseline is the comparator that the ~50 µg/day BraMa-Ni diet (Braga 2013) targets.
- App: The full mean-Ni-content table is the seed dataset for ingredient-page contamination_profile.Ni values. Body-prose synthesis should cite this paper as foundational and supplement with newer occurrence data (Pereira 2020, cereal-grains studies) where available.
- Courses: Standard older reference for Ni-occurrence-in-food. Cited by virtually every subsequent Ni-in-food paper.