Survey of Inorganic Arsenic in Seaweed and Seaweed-Containing Products Available in Australia

Overview

Food Standards Australia New Zealand surveyed inorganic arsenic in 38 dried seaweed and seaweed-containing product samples collected from Australian retail in April–May 2010, undertaken to review the 2004 FSANZ consumer advice to avoid hijiki. Inorganic arsenic in dried seaweed ranged from below the 0.05 mg/kg limit of reporting to 7.80 mg/kg in a single hijiki composite, with all samples below the Australia New Zealand Food Standards Code maximum level of 1 mg/kg (calculated at 85% hydration) except that one hijiki composite, which at 7.80 mg/kg dried equated to 1.4 mg/kg at 85% hydration and was referred to the relevant jurisdiction. Inorganic arsenic was consistently higher in hijiki/Sargassum fusiforme than in wakame, kombu, or nori; among seaweed-containing products only two seaweed-chip samples exceeded the limit of reporting. The survey reproduces the cross-national pattern that hijiki is the inorganic-arsenic outlier among edible seaweeds.

Key numbers

Method limits: limit of reporting (LOR) 0.05 mg/kg; practical quantitation limit (PQL/LOQ) 0.03 mg/kg. Inorganic arsenic defined as arsenite (As³⁺) + arsenate (As⁵⁺), conservatively including any co-extracted MMA. All values dried-seaweed basis unless noted; non-detects assigned 0.05 mg/kg (upper bound) for means.

Table 1 — inorganic arsenic (mg/kg) in dried seaweed (individual and composite samples):

Table 2 — inorganic arsenic (mg/kg) in seaweed-containing products:

Additional source-reported facts:

• All dried-seaweed samples were below the Code ML of 1 mg/kg inorganic arsenic (iAs) except the one hijiki composite at 7.80 mg/kg dried (= 1.4 mg/kg at 85% hydration; hijiki composite moisture 14.6%), which exceeded the ML and was referred to the relevant jurisdiction.
• In this survey hijiki inorganic arsenic was at least 23-fold higher than other brown seaweeds and approximately 48-fold higher than nori.
• Samples labelled hijiki versus Sargassum fusiforme differed by at least 20-fold (7.80 versus 0.32 mg/kg). The authors discuss harvest region, season and labelling uncertainty as possible explanations; the survey does not establish a cause for the difference between those retail composites.
• Cross-national comparison (the survey’s Table 3) reported the dried-seaweed inorganic-arsenic ranges as: wakame 0.16–0.20, kombu 0.16–0.33, hijiki/Sargassum 0.32–7.8, nori 0.09–0.16, sea vegetable 0.10 mg/kg (FSANZ 2012, n=38), with hijiki the highest across all countries compared.
• Dietary exposure (Appendix 1, 23rd ATDS baseline + assumed 10 g/day dried seaweed): best-case scenario 0.3–0.7 µg/kg bw/day across population groups; worst-case 2.2–7.6 µg/kg bw/day; for children aged 2–5 the range across scenarios was 0.7–7.6 µg/kg bw/day. Exposure is far more sensitive to high seaweed arsenic concentrations than to background diet.
• Hazard context: JECFA (2011) BMDL₀.₅ for inorganic arsenic = 3 µg/kg bw/day (range 2–7); the prior PTWI of 15 µg/kg bw/week was withdrawn.

Methods (brief)

Samples were drawn from those already collected for the companion FSANZ iodine survey (retail outlets, all Australian states/territories, April–May 2010): 22 individual samples plus composites (three 3-purchase composites of different brands, one 3-purchase composite of the same brand, two 2-purchase composites of the same brand). Dried seaweed was analysed directly from packaging; seaweed-containing products (soup, tea, etc.) were analysed as purchased. The National Measurement Institute (NMI) performed analysis using a NATA-accredited method of acid extraction followed by hydride-generation inductively coupled plasma mass spectrometry (HG-ICP-MS; Holak and Specchio 1991), LOR 0.05 mg/kg, PQL 0.03 mg/kg. Inorganic arsenic was reported as arsenite + arsenate; because the extraction may co-extract MMA, the conservative total-inorganic figure may slightly overestimate true inorganic arsenic. Concentrations are for dried seaweed except where the 85%-hydration basis is stated (used only for the ML comparison on the exceeding hijiki sample). The report rates occurrence-data uncertainty as low (accredited, validated method) and overall dietary-exposure uncertainty as medium-to-high (driven by sparse Australian seaweed-consumption data).

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.