Lima de Paiva et al. 2017 — Cd, Pb, Sn, total Hg, and MeHg in canned tuna from São Paulo, Brazil

Paiva, Morgano, and Milani determined cadmium, lead, tin, total mercury, and methylmercury in 30 samples of canned tuna (five anonymised brands, three batches each, conserved in water or oil) acquired from supermarkets in Campinas, São Paulo state, Brazil, in 2015. Cd, Pb, and Sn were measured by ICP-OES after closed-vessel microwave HNO₃/H₂O₂ digestion; total Hg was measured by thermal decomposition–amalgamation atomic absorption (TDA AAS, DMA-80); MeHg was measured by toluene/HCl extraction in a closed microwave vessel, transferred to L-cysteine, then quantified on the DMA-80 with the organic-phase mercury treated as MeHg in full. Twenty per cent of the oil-conserved samples exceeded the Brazilian (ANVISA RDC 42/2013) and EU (Regulation 1881/2006) maximum level of 100 µg/kg for cadmium in fish; no sample exceeded the 300 µg/kg lead limit; tin was below the 94 µg/kg detection limit in every sample. Total Hg ranged 44–402 µg/kg wet weight in the oil conserve and 51–460 µg/kg in the water conserve; MeHg ranged 35–393 µg/kg in the oil conserve and 41–460 µg/kg in the water conserve. The MeHg/total-Hg ratio sat between 82 % and 99 % across all 30 samples. Using a 135 g drained-weight portion and a 60 kg adult, the authors estimated that four cans per week (540 g) of water-conserved tuna and five cans per week (675 g) of oil-conserved tuna would reach 100 % of the JECFA provisional tolerable weekly intake for MeHg (1.6 µg/kg body weight per week).

Key numbers

All concentrations are µg/kg wet weight unless noted. nd = below the method LOD: 8.5 µg/kg for Cd, 30 µg/kg for Pb, 94 µg/kg for Sn (Table 1). Total Hg LOD 0.5 µg/kg; MeHg LOD 1.4 µg/kg.

Table 2 — water-conserved canned tuna (mean and range, three batches per brand, triplicate analysis):

Table 2 — oil-conserved canned tuna (mean and range, three batches per brand, triplicate analysis):

Cross-conserve summary (text, p. 188 and p. 190):

• Cd: oil-conserve highest brand mean about 85 µg/kg (Brand 2, with the 213 µg/kg upper-range sample driving that mean); water-conserve typical brand mean about 20 µg/kg; overall range nd–213 µg/kg.
• Pb: per p. 188, oil-conserve highest mean content about 44 µg/kg (Brand 4 in Table 2); water-conserve “smaller mean contents” reported as <30 µg/kg. Note that Table 2 brand-level means range 34–46 µg/kg for water-conserve and 31–44 µg/kg for oil-conserve, so the source’s “<30” water-conserve narrative does not match the per-brand means in its own Table 2; the brand-level table values are the authoritative numbers above. Overall range nd–59.2 µg/kg; no sample exceeded the 300 µg/kg ML.
• Sn: not detected in any sample (<94 µg/kg). Authors infer the can-interior lacquer is performing as intended.
• Total Hg: 44–402 µg/kg across the 15 oil-conserved samples; 51–460 µg/kg across the 15 water-conserved samples. Pooled means (p. 190): 169 µg/kg in the oil conserve and 173 µg/kg in the water conserve.
• MeHg: 35–393 µg/kg across the 15 oil-conserved samples; 41–460 µg/kg across the 15 water-conserved samples. Pooled means (p. 190): 154 µg/kg in the oil conserve and 160 µg/kg in the water conserve.

MeHg / total Hg ratio (Figure 1): 82–99 % across all 30 samples. The most-toxic mercury species dominates every brand × conserve combination.

Twenty per cent Cd exceedance: 20 % of the 15 oil-conserved samples exceeded the 100 µg/kg ML for Cd in fish set by ANVISA RDC 42/2013 and EU Regulation 1881/2006. None of the water-conserved samples exceeded.

Exposure estimate (authors’ calculation)

The authors take a 135 g drained-weight portion (one can) and a 60 kg adult; the PTWI for MeHg is 1.6 µg/kg body weight per week (CCCF/CAC 2015) and the PTWI for total Hg is 4 µg/kg body weight per week.

• Cd: at the highest sample value (213 µg/kg), an adult would need to consume about 52 cans (~7 kg) in a month to reach the JECFA PTMI of 25 µg/kg body weight per month. Authors conclude Cd exposure from canned tuna is not the binding concern at this contamination level.
• Pb: every sample falls below the 300 µg/kg ML, so consumption-based PTWI characterisation is not the binding control; CAC withdrew the Pb PTWI in 2010 because the prior value (25 µg/kg bw/week) was associated with a 3-point IQ loss in children and elevated adult blood pressure.
• Total Hg: at the oil-conserve pooled mean of 169 µg/kg, a 60 kg adult would have to eat 11 cans/week (~1.5 kg) to reach 100 % of the total-Hg PTWI; at the water-conserve pooled mean of 173 µg/kg, 10 cans/week (~1.4 kg).
• MeHg: at the oil-conserve pooled mean of 154 µg/kg, 5 cans/week (~675 g) would reach 100 % of the MeHg PTWI; at the water-conserve pooled mean of 160 µg/kg, 4 cans/week (~540 g) would reach 100 %.

The four-cans-per-week threshold (water conserve) is the headline number. Adults eating roughly one can every other day of water-conserved canned tuna from this market reach the MeHg PTWI ceiling. The headline does not assume tail-end brands; it uses the pooled mean.

Methods notes

• Acquisition: three batches × five brands × two conserve types = 30 samples; brands were the “most commonly consumed and available in the largest markets” in the Campinas region. Brand identities are anonymised as Brand 1–5 throughout.
• Sample handling: drained from water/oil, ground in a domestic processor, stored at −18 °C, analysed in triplicate.
• Cd, Pb, Sn — closed-vessel microwave digestion (Start D, Milestone): 1 g sample + 4 mL HNO₃ + 4 mL deionised water overnight, then 2 mL 30 % H₂O₂; four-ramp programme to 170 °C, hold 25 min; 25 mL final volume in 5 % HNO₃; Agilent 5100 VDV ICP-OES, axial mode, Cd 214.439 nm, Pb 220.353 nm, Sn 189.925 nm.
• Total Hg — TDA AAS (DMA-80, Milestone): 60 mg sample (water-conserved) or 80 mg (oil-conserved); dry 200 °C/60 s, decompose 600 °C/180 s, desorb 850 °C, λ = 253.7 nm.
• MeHg — closed-microwave toluene/HCl extraction (Paiva et al. 2016 method): 1 g sample + 8 mL toluene + 1 mL water + 0.75 mL 30 % HCl, ramp to 110 °C in 10 min and hold 5 min; 4 mL of the organic phase transferred to 2 mL 2.5 % L-cysteine and centrifuged 6 min at 3500 rpm; 100 mg of the L-cysteine phase quantified on the DMA-80 (dry 120 °C/60 s, decompose 300 °C/180 s, desorb 850 °C/12 s, λ = 253.7 nm). Authors treat the organic-phase Hg as MeHg in full.
• Calibration: 0.002–0.5 mg/L for Cd and Pb; 0.05–10 mg/L for Sn; r² ≥ 0.999 across all curves.
• QC reference materials (Table 1): TORT-2 hepatopancreas for Cd and Pb (recoveries 87 ± 1 % and 87 ± 2 %), NIST SRM 1566b oyster tissue for total Hg (97 ± 1 %), DORM-4 fish protein for MeHg (111 ± 2 %). Sn does not have a certified value in the materials used; three fortification levels (100, 500, 5000 µg/kg) gave recoveries 89 ± 0.8 % to 92 ± 1.2 %.
• Precision (16 repetitions on a water-conserved sample): 3.7 % RSD (Cd), 4.1 % (Pb), 5.5 % (total Hg), 9.0 % (MeHg).

Regulatory comparators cited by the authors

• ANVISA RDC 42/2013 (Brazil, MERCOSUR Technical Regulation on inorganic-contaminant maximum levels in food): Cd ML 100 µg/kg in fish; Pb ML 300 µg/kg in fish.
• EU Regulation 1881/2006: Cd ML 100 µg/kg, Pb ML 300 µg/kg in fish.
• JECFA via CCCF/CAC 2015: PTMI for Cd = 25 µg/kg body weight per month; PTWI for total Hg = 4 µg/kg body weight per week; PTWI for MeHg = 1.6 µg/kg body weight per week. The PTWI for Pb (formerly 25 µg/kg bw/week) was withdrawn in 2010 because it was associated with a 3-point IQ loss in children and elevated adult blood pressure.

Limitations

• Brand anonymisation removes brand-identifying lab data from public view (consistent with the wiki’s brand firewall), but it also means the brand-level Cd outlier (Brand 2 oil-conserve, mean 85 µg/kg, range 15–213 µg/kg) cannot be attributed to a named commercial tuna line for procurement-pattern follow-up.
• Tuna species (skipjack, yellowfin, albacore, bonito) and FAO catch area are not reported. Fresh-vs-canned and species-vs-species comparisons in the discussion are referenced to other studies; the present study cannot itself separate species or geographic-catch effects from canning-process effects.
• “Organic Hg” is operationally defined as Hg in the toluene/L-cysteine extract under the Paiva 2016 protocol; the authors treat this fraction as MeHg in full, which is conventional in fish but may slightly overstate MeHg if minor non-methyl organomercury species are present.
• Sn is reported only as the inorganic Sn total measured by ICP-OES at 189.925 nm; organotin species (TBT/TPT) are not measured. The “tin not detected” finding speaks only to inorganic tin migration from the can; it does not address organotin contamination from environmental routes.
• N is modest (30 samples, five brands). The 20 % Cd exceedance is a within-brand observation driven by Brand 2 oil-conserve and should not be projected to the Brazilian canned-tuna market as a whole without the larger surveys the authors cite (Voegborlo 1999, Ashraf 2006, Ganjavi 2010) which were performed in other countries.

Verification notes

• Audit subagent (2026-06-04) flagged the intro and Key-numbers cross-conserve summary as quoting a composite “44–460 µg/kg” total Hg span and “35–460 µg/kg” MeHg span that mix one conserve type’s minimum with the other’s maximum — no single conserve type spans either range. Verified against the source p. 188: oil 44–402 / 35–393, water 51–460 / 41–460. Both ranges split into per-conserve form on this page.
• Audit subagent (2026-06-04) flagged that the cross-conserve Pb summary had inverted the conserve attribution. Verified against source p. 188: the source’s narrative (“oil … Pb (44 µg/kg) … water … Pb (<30 µg/kg)”) gives the highest Pb mean to the oil conserve and the lower mean to the water conserve. Cross-conserve Pb summary corrected; a one-sentence flag is added because Table 2’s brand-level means (water 34–46, oil 31–44) do not align with the source’s “<30” water-conserve claim, and the table is treated as authoritative for the per-brand numbers.

Routing notes

• Direct evidence for [[ingredients/canned-tuna]] and [[products/canned-fish]], [[products/canned-seafood]]. Broad-product-category context for [[products/seafood]] and [[products/fish-marine-predatory]] (the source measures canned product, not fresh predatory tuna species).
• Direct evidence on Cd, Pb, tHg, and MeHg in canned tuna for the Brazilian (ANVISA RDC 42/2013) and EU (Regulation 1881/2006) regulatory comparators.
• Sn finding is a non-detect across all 30 samples; this is a coverage-relevant negative for canned-tuna Sn migration discussion on [[metals/tin]] and on the canned-fish product page.

Page history

The five most recent substantive edits to this page. 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.