Lordan and Zabetakis 2022 — Cadmium in the brown crab industry and human health risks

Lordan and Zabetakis review cadmium accumulation in brown crab (Cancer pagurus), the divergent regulatory limits applied in the European Union versus the People’s Republic of China and Hong Kong, the dietary-intake context that determines whether crab consumption contributes meaningfully to total cadmium exposure, and the clinical effects of chronic cadmium exposure on humans. The review is most useful to the wiki as compiled occurrence evidence for cadmium in brown-crab white meat, brown meat, and hepatopancreas across European fisheries, and as a synthesis of the EU/PRC regulatory divergence that has shaped live-export trade.

Key numbers

All concentration values below are reported as the review’s Table 1 summary of cadmium in brown crab on a wet-weight basis. Values are reproduced as the review presents them, with the underlying primary-study reference number from the review preserved in the right-hand column; consult the primary studies before using any single value in a routed pool.

Cadmium in brown crab tissues (Table 1, mg/kg wet weight)

Regulatory limits and dietary reference values

• EU Commission Regulation (EU) No 488/2014 amending (EC) No 1881/2006: cadmium maximum level of 0.5 mg/kg applies to white crab meat from appendages only [18–20].
• People’s Republic of China and Hong Kong: cadmium maximum level of 0.5 mg/kg applies to the combined white and brown crab meat of all imported crabs [9,17,21]. PRC public consultation has been opened to raise the brown-crab cadmium limit from 0.5 mg/kg to 3.0 mg/kg; an update on the consultation had not been released at the time of review.
• EFSA tolerable weekly intake (TWI) for cadmium: 2.5 µg/kg body weight per week.
• JECFA provisional tolerable monthly intake (PTMI) for cadmium: 25 µg/kg body weight per month, derived from a urinary cadmium excretion rate of 5.24 µg/g creatinine or 0.8 µg/kg per day as a nephrotoxicity threshold [148–150]. (The review itself uses the unprovisional shorter form “tolerable monthly intake” without the “provisional” qualifier; the JECFA-canonical designation is PTMI and is used here to match the regulation slug.)
• EFSA critical β2-microglobulin urinary endpoint: 300 µg/g creatinine; EFSA-adopted cadmium excretion rate as nephrotoxicity threshold: 1 µg/g creatinine; uncertainty factor 0.36 µg/kg body weight per day for 50 years as benchmark dose [151].

Dietary-intake context

• Lifetime cadmium dietary exposure for the European population is estimated at approximately 2 µg/kg body weight per week (within EFSA’s TWI of 2.5 µg/kg body weight per week).
• Ireland: estimated adult weekly cadmium intake of 1.1–2.5 µg/kg body weight per week (44–62% of the EFSA TWI); 95% of National Adult Nutrition Survey participants had urinary cadmium below the 1 µg cadmium/g creatinine value EFSA has deemed safe [101,102]. Cereals contributed 39%, vegetables 36%, dairy 12%, and fish and shellfish approximately 1% of dietary cadmium [101].
• United States (NHANES 2007–2012, n=12,523): average dietary cadmium intake 4.6 µg/day, or 0.54 µg/kg body weight per week (≈22% of TWI); leading contributors were cereals and bread (34%), leafy vegetables (20%), potatoes (11%), legumes and nuts (7%), and root vegetables (6%); leading individual foods were lettuce (14%), spaghetti (8%), bread (7%), and potatoes (6%) [103]. Fish and shellfish contributed 1%.
• South Korea (cohort n=1,245): estimated average cadmium intake 20.8 µg/day [26]. Leading contributors were rice (40%), squid (12%), eel (11%), crab (9%), shellfish (4%), seaweed (3%), and kimchi (3%); crab specifically contributed 8.6% of cumulative cadmium intake in this cohort.
• People’s Republic of China general population: mean dietary cadmium exposure 15.3 µg/kg body weight per month (30.6 µg/day for a 60 kg adult). Table 2 contributor percentages for the general Chinese population: rice 55.8%, wheat flour 11.8%, leafy vegetables 10.5%, shellfish 4.8%, meat 2.6%, seaweed 2.4%, other vegetables 2.4%, other cereals 2.1%, root and stalk vegetables 2.0%, mushrooms 1.1%, fish 1.1%, legumes 0.9%, fruits 0.6%, eggs 0.6%, nuts 0.4%, offal 0.4%, other 0.5% [108].
• People’s Republic of China high-exposure subpopulation (≥95th percentile of mean dietary cadmium exposure): rice 58.6%, shellfish 13.2%, leafy vegetables 9.2%, seaweed 6.4%, wheat flour 2.0%, meat 2.0%, root and stalk vegetables 1.7%, mushrooms 1.5%, other vegetables 1.4%, fish 1.0%, other cereals 0.9%, legumes 0.6%, fruits 0.4%, eggs 0.2%, nuts 0.4%, offal 0.2%, other 0.3% [108].

Human-health and biomarker numbers

• Cadmium biological half-life in the kidney: 10–30 years [81,135].
• Critical kidney cadmium cortex concentration: approximately 200 µg/g wet weight (200 ppm); the urinary cadmium value corresponding to this kidney concentration is estimated at 10 µg/g creatinine [138,140,141].
• Cadmium gastrointestinal absorption: 3–5% of ingested cadmium [20,118].
• Bioaccessibility comparison from one cited study: cadmium 84% from fish versus 73% from shellfish [147].
• Long Island Study of Seafood Consumption (n=252 avid seafood consumers): salmon intake associated with log-transformed blood cadmium (β = 0.20; p = 0.001); regular seafood intake not associated (β = −0.01; p = 0.11) [182].
• Norwegian Fish and Game study (n=179): median urinary cadmium 0.16 µg/L (creatinine-corrected); median blood cadmium 0.45 µg/L; 24% of designated high-cadmium consumers and 8% of low-cadmium controls had intakes above the EFSA TWI [183].
• Crab-consumer cohort: regular consumers averaging 447 g brown crab meat per week over 16 years showed no elevation in long-term cadmium exposure markers or kidney-toxicity biomarkers relative to low-consumption controls [143].
• French seafood-consumer cohort: the review prints the mean dietary cadmium ingestion verbatim as “2.4 ± 3.3 mg/kg bodyweight/week” and the mean urinary cadmium verbatim as “0.65 ± 0.45 mg/g creatinine” (PDF p.16); the printed mg units are approximately three orders of magnitude above the EFSA TWI and almost certainly a units typo for µg in the review itself. Values are preserved exactly as the review prints them; see Verification notes for the typo flag, and do not propagate either interpretation without consulting the underlying primary study. Urinary cadmium was significantly higher in women than men.
• US national geometric mean blood cadmium for non-smoking adults: 0.47 µg/L; smoker mean approximately 1.58 µg/L [98].

Methods (brief)

This is a narrative review. It compiles cadmium concentration tables for brown crab tissues from prior primary studies (Table 1) and summarizes dietary-exposure surveys, regulatory limits, biomarker thresholds, and toxicological evidence from European, US, South Korean, and PRC literature. The review does not generate new analytical measurements; concentration values reported in Table 1 are attributed to underlying primary references [13,14,30,33,64,65,66] and used the detection methods (FAAS or ICP-MS) reported by those primary studies. Cooking states (raw, steamed, boiled) and tissue types (white claw meat, brown meat, hepatopancreas) follow the primary-study sampling; basis is wet weight as reported by the underlying studies. The review notes that cadmium detection in biological samples is most commonly performed today by ICP-MS (limit of detection as low as 0.003 µg/L); graphite furnace AAS has a sample detection limit of 0.4 µg/L [67,68,70]. The review reports no de novo speciation or analytical work of its own.

Implications

Certification: The review is a useful entry point for HMTc cadmium-in-crustacean threshold work because it (a) consolidates wet-weight brown-crab concentration evidence across multiple European fisheries with the cooking state and tissue type preserved, (b) documents the EU/PRC regulatory divergence on what tissue the 0.5 mg/kg limit applies to (white meat only in the EU; combined white and brown meat in the PRC and Hong Kong), and (c) clarifies that the brown meat and hepatopancreas tissue compartments carry cadmium concentrations approximately one to two orders of magnitude higher than the white claw meat from the same animal. Routed-pool decisions should consult the underlying primary studies (Wiech et al. 2017, Wiech et al. 2020, Ervik et al. 2020, Maulvault et al. 2012, Barrento et al. 2009, and the Birsay Scotland study) for the raw observation data; this review row should not be pooled as if it were a primary observation.

Courses: Useful teaching example for explaining why “the cadmium limit for crab” is not a single number — the EU limits white meat to 0.5 mg/kg while ignoring brown meat, the PRC and Hong Kong apply the same 0.5 mg/kg limit to combined white plus brown meat, and the brown meat itself can run 8–30+ mg/kg in European fisheries. The disconnect between the regulatory framing and the tissue distribution of cadmium is the heart of the live-export trade dispute documented in the review.

App: Route as broad shellfish and seafood occurrence evidence for cadmium with the crab-specific brown-meat-versus-white-meat distinction preserved. Do not use the review’s compiled Table 1 values as if they were primary measurements when computing ingredient contamination_profile blocks; defer the pooled occurrence values to the underlying primary studies once those are ingested. Crab is not a separate ingredient slug in the current taxonomy; this review is therefore best treated as shellfish/seafood context until a crab-specific routing question forces a taxonomy decision.

Wiki pages this source may touch

• shellfish
• seafood
• shellfish
• seafood
• cadmium
• eu-1881-2006-contaminants-superseded
• efsa-cadmium-twi
• jecfa-cadmium-ptmi

Verification notes

Identifying metadata: title and authors taken from the PDF first page; DOI 10.3390/toxics10100591 taken from the PDF citation block; publication year and volume/issue from the running header (Toxics 2022, 10, 591). License is CC BY 4.0 as printed on the PDF first page (Creative Commons Attribution 4.0 license). No identity collision was detected: zero existing wiki source page returns on the DOI grep, the KADC_ raw_handle grep, and the lordan2022 cite-key grep.

The Table 1 cadmium values were transcribed directly from PDF page 5; the table preserves the review’s two-column “Spring Caught / Summer Caught” header for the Portugal block, and the “Mean/range” or “yearly median concentration range” header structure for the remaining blocks. Where the review prints ”-/0.08-0.27” for the Birsay Scotland white meat row, the leading hyphen indicates that no mean was reported by the underlying study, only a range; the row preserves that convention. The Norwegian coast Cd brown meat value 8.7 / 0.24–43.0 is attributed to refs [33,65] as printed; the row labels for Senja and Kvaløya are attributed to refs [33,66] as printed.

The review reports the French seafood-consumer dietary cadmium ingestion as 2.4 ± 3.3 mg/kg bodyweight/week and the mean urinary cadmium as 0.65 ± 0.45 mg/g creatinine (PDF p.16); both figures as printed are approximately three orders of magnitude above their plausible scale (the dietary intake is ~1000x the EFSA TWI of 2.5 µg/kg body weight per week; the urinary value at 0.65 mg/g creatinine is ~1000x typical population means in the µg/g creatinine range), which strongly suggests a units typo (likely µg, not mg) in the review itself. Both values are preserved on this page exactly as the review prints them and flagged here; downstream synthesis should not propagate either interpretation without consulting the underlying primary source.

2026-06-03 audit application (fresh-context Claude Opus 4.7 audit subagent, verdict REVISE):

• Finding 1 (⚠️ Check 1, JECFA TMI/PTMI label mismatch with the regulations/jecfa-cadmium-ptmi slug): verified against PDF p.14 (“The JECFA established a tolerable monthly intake of 25 µg/kg/bodyweight per month”) — the review uses the non-canonical short “TMI” form while the JECFA standard designation and the wiki regulation slug both use “PTMI” (provisional tolerable monthly intake). Applied: the body now uses “provisional tolerable monthly intake (PTMI)” to match the slug, with an explicit parenthetical noting that the review itself omits the “provisional” qualifier.
• Finding 2 (⚠️ Check 1, French cohort body-vs-Verification-notes inconsistency): verified against PDF p.16 — the review verbatim prints “2.4 ± 3.3 mg/kg bodyweight/week” and “0.65 ± 0.45 mg/g creatinine”, while the previous body text silently converted both to µg, contradicting the Verification notes’ own claim that the values were “preserved here exactly as the review prints it.” Applied: the body now reproduces both values verbatim as the review prints them (mg) with an inline flag pointing to the typo discussion in these notes.
• Finding 3 (⚠️ Check 1 minor, Table 2 high-exposure PRC integer-vs-one-decimal stylistic normalization for wheat flour 2.0/meat 2.0/fish 1.0): verified — the PDF Table 2 prints these as integers (2, 2, 1) where the wiki normalizes to one decimal; the numeric values are identical. Not applied — judged stylistic, not a fidelity error.
• All Check 2/3/4/5 findings clean per subagent. Routing audit re-run after fix: 0 blocking changes.

Brand firewall (Part 12): the review does not name commercial brands; no brand-firewall edits were required. Scientific-method vendor names are not present in the underlying primary studies as cited in Table 1 (only the analytical-method abbreviation: FAAS or ICP-MS). The CAS-system reference materials and instrument-vendor identifiers cited in some of the underlying primary studies should be captured when those primary studies are ingested individually, not retrofitted from this review.

Wiki/HMTc firewall (Part 2): the review’s “Conclusions” section is the kind of synthesis-from-other-literature statement that does not belong on a single-source wiki page; this page reports only what Lordan and Zabetakis tabulated and documented, not whether their interpretation of the literature is correct. No HMTc threshold proposals were imported from the review.

Speciation: the review measures (or summarizes) total cadmium only, which is the practice across the underlying primary studies. Cadmium is not a speciated analyte in the HMTc vocabulary; no speciation flag applies.

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.