Rusin et al. 2021 — Cd and Pb across fresh, frozen, dried, and processed Polish fruits and vegetables

Rusin and colleagues measured cadmium and lead in 370 retail samples of six fruit species (apple, cranberry, grape, pear, raspberry, strawberry) and four vegetable species (beetroot, celery, carrot, tomato) sourced from the Polish market, partitioned across four processing states (fresh, frozen, dried, processed). Concentrations were determined by graphite-furnace atomic absorption spectrometry against the TYG006RM vegetable-puree certified reference material under ISO/IEC 17025 conditions. Mean concentrations of both metals were highest in dried products on a dry-weight basis, and twelve samples (2 fruit, 10 vegetable) exceeded the EU Cd maximum permissible level while three vegetable samples exceeded the EU Pb maximum, with a single frozen tomato reaching 1074% of the 0.050 mg/kg Pb limit.

Key numbers

Sample structure (Table 1):

Below-LOQ tallies (Results, p. 4): In fruits (n=242 total), Cd was below the LOQ in 87 samples and Pb in 96 samples. In vegetables (n=128 total), Cd was below the LOQ in 31 samples and Pb in 69 samples.

LOD / LOQ and CRM recovery (Table 4):

Cd and Pb in fruits — mean, SD, min, max (Table 6). Values in mg/kg fresh mass except dried rows in mg/kg dry weight (marked DW). A dash indicates the mean was below the LOQ for that sub-group.

Cd and Pb in vegetables — mean, SD, min, max (Table 7). Same units convention as Table 6.

Exceedances of EU maximum permissible levels (Table 5). Cd MPLs per Commission Regulation (EU) 488/2014 amending (EC) 1881/2006; Pb MPLs per Commission Regulation (EU) 2015/1005 amending (EC) 1881/2006. Percentages are the sample concentration expressed as a fraction of the MPL.

Totals: Cd MPL exceeded in 12 samples (2 fruit, 10 vegetable); Pb MPL exceeded in 3 vegetable samples. The 1074%-of-MPL Pb result is a single frozen tomato sample at approximately 0.537 mg/kg fresh mass (matching the Table 7 frozen-tomato Pb maximum).

ANOVA across processing states (Table 8). Below-LOQ samples were excluded from each analysis; the number of excluded samples is in parentheses.

Outlier analysis (Figure 1, narrative on p. 6). Dried fruits Cd: two outliers at 0.277 and 0.210 (strawberry max and likely raspberry/strawberry). Fresh vegetables Cd: six outliers at 0.203, 0.670, 0.260, 0.690, 0.300, 0.712 (concentrated in beetroot and celery rows). Processed fruits Pb: two outliers at 0.127 and 0.047 (grape processed maximum and a pear processed value). Frozen vegetables Pb: one outlier at 0.537 (frozen tomato).

Tukey HSD post-hoc summary (Figure 2 and p. 8). For Cd in fruits, dried-fruit means differ significantly from fresh, frozen, and processed means. For Cd in vegetables, fresh and processed means differ; frozen versus fresh is significant only at the 10% level. For Pb in fruits, dried-fruit means differ significantly from fresh, frozen, and processed. For Pb in vegetables, processed-versus-dried is highly significant, with fresh-versus-dried and processed-versus-frozen also significant.

Methods (brief)

Graphite-furnace AAS on a SavantAA Sigma spectrometer with PAL3000 autosampler and a GF3000 graphite furnace (GBC, Australia), with background correction, wavelengths 228.8 nm (Cd) and 217 nm (Pb), and lamp currents 5.0 mA (Cd) and 0.3 mA (Pb) (Table 3). Sample preparation followed consumption practice (washing, peeling, removal of inedible parts) before homogenisation on a T18 Digital Ultra-Turrax (IKA, Germany) for fresh and frozen samples and on a Testchem PZS 01 vibratory grinder (Poland) for dried samples; processed liquids were mixed and syrup/marinade products were separated and homogenised. Sample weights: 2 g for fresh/frozen, 1 g for dried, 2 g for processed, 10 g for liquids; 8 mL HNO₃ (65%, Suprapur, Merck, Germany) was added to all samples and dried samples received an additional 1 mL of 30% H₂O₂. Mineralisation used a four-phase microwave program on a Magnum II microwave digester (Ertec, Poland) (Table 2). Calibration used SPEX CertiPrep 1000 mg L⁻¹ Cd and Pb standards in 2% HNO₃ (R² = 0.999 for both metals; Table 3). Method performance was verified against TYG006RM vegetable-puree CRM. LODs were estimated as mean blank signal + 3 SD and LOQs as mean blank signal + 6 SD (Cd LOD 0.2 µg/L, LOQ 0.35 µg/L; Pb LOD 1.7 µg/L, LOQ 3.5 µg/L). Sampling and analysis followed Regulation (EC) 882/2004 and Regulation (EC) 333/2007. Statistics used ANOVA at α = 0.05, boxplot-based outlier analysis, and Tukey HSD post-hoc tests with Bonferroni correction for multiple comparisons; below-LOQ values were excluded from the variance analyses.

The paper reports Cd and Pb only; no other metals were measured. Cadmium and lead are both reported as total elemental concentrations (no speciation). Cranberry samples were drawn from Vaccinium macrocarpon Aiton (the American cranberry). Concentrations for fresh, frozen, and processed products are on a fresh-mass basis; dried-product concentrations are on a dry-weight basis and are not directly comparable to the EU MPLs (which are framed on a fresh-mass basis), so the per-row exceedance counts in Table 5 are restricted to fresh, frozen, and processed samples.

Implications

Certification: Contributes Polish retail-market occurrence data for Cd and Pb across ten produce commodities in four processing states (n=370), with explicit MPL-exceedance counts against the EU 488/2014 (Cd) and 2015/1005 (Pb) limits. The fresh-beetroot Cd distribution (mean 0.235 mg/kg f.m., max 0.67) and the fresh-celery Cd distribution (mean 0.152 mg/kg f.m., max 0.712) are the highest within-group means in the vegetable dataset and are relevant to upper-tail percentile estimation for root-and-tuber and stem vegetable product categories. The 0.537 mg/kg frozen-tomato Pb outlier is a single sample but contributes to upper-percentile sensitivity for frozen vegetable matrices.

Courses: Useful case study for two distinct teaching points — (i) how dry-weight reporting can make dried products look anomalously high when juxtaposed against fresh-mass MPLs, requiring explicit basis conversion before any threshold comparison; and (ii) how single-sample exceedance can dominate a small-n subgroup (the 0.537 mg/kg frozen tomato Pb result is the entire signal for the frozen-tomato Pb category). Methodologically, the paper is a reproducible example of ISO 17025–accredited GFAAS measurement with CRM-verified recovery.

App: Contributes occurrence data for Polish-origin fresh, frozen, dried, and processed forms of apple, pear, grape, raspberry, strawberry, cranberry, beetroot, celery, carrot, and tomato to the cadmium and lead profiles for each commodity ingredient. Processing-form must be tracked separately in any per-ingredient model because dried-versus-fresh concentrations differ by roughly the water-removal factor.

Verification notes

Page rewritten 2026-05-19 from a 2026-05-12 revision against the source PDF (DOI 10.1038/s41598-021-91554-z). Corrections applied:

• Authors: previously listed as [Rusin M, Romanowska-Duda Z, Brylonek M, Rogulski M, Jasinski M]; the PDF (p. 1) names Monika Rusin, Joanna Domagalska, Danuta Rogala, Mehdi Razzaghi & Iwona Szymala. Four of the five prior co-author names were fabricated. Corrected to the actual author list.
• Title: previously "Cadmium and lead contents in fresh and processed fruit and vegetable products and the resulting dietary risk assessment"; the PDF title is "Concentration of cadmium and lead in vegetables and fruits". The prior title was fabricated. Corrected.
• Key numbers section: the prior revision invented entire findings that the paper does not contain — “dried herbs” and “dried mushrooms” with values up to 2–5 mg/kg DW, “leafy vegetables Cd mean 0.05–0.15 mg/kg fresh weight”. The paper measures only ten commodities (apple, cranberry, grape, pear, raspberry, strawberry, beetroot, celery, carrot, tomato); there are no herbs, no mushrooms, and no leafy vegetables among the species sampled. All fabricated values removed and the Key numbers section rewritten from Tables 1, 4, 5, 6, 7, and 8 of the source.
• Methods section: prior version was a two-sentence summary that did not reflect the instrument model, the sample-prep workflow, the four-phase microwave program, the per-form sample weights, or the analytical-method statute basis. Rewritten from the Material and Methods section (pp. 2–3) and Tables 2–4.
• ingredients: previously [leafy-greens, carrot, apple, fruit] — leafy-greens is not a commodity the paper measured. Replaced with the slugs for the actual ten species, minus cranberries (not in the current taxonomy snapshot; freq-1 here — flagged below for provisional-scaffold consideration).
• products: previously [fresh-fruit] only — the paper covers fresh, dried, canned/jarred (jam, marmalade, syrup), and 100%-juice fruit products and fresh/frozen/dried/processed vegetable products. Expanded to [fresh-fruit, dried-fruit, canned-fruit, fruit-juice-not-canned, root-tuber-vegetables, canned-vegetables].
• matrices: previously listed product-shaped strings; replaced with the standard matrix vocabulary used elsewhere in the corpus: [fresh-fruit, frozen-fruit, dried-fruit, processed-fruit, fresh-vegetable, frozen-vegetable, dried-vegetable, processed-vegetable].
• raw_handle: added the per-paper handle MFK_concentration-of-cadmium-and-lead-in-vegetables-an (the prior page had no raw_handle field).
• raw_path: updated from the stale raw/studies/... location to the canonical Manual Fetch Kimi path (the file at the prior location has the same SHA-256 and is a copy of the MFK PDF; the MFK location is the ingest source).
• raw_sha256: field renamed from sha256: (value preserved verbatim: 75277429df4ab9cae5b91c476c9437ed795d7ad3ca3b7a458098fdbda1810bae).
• sample_population: added; previously absent.
• Implications section: prior version contained fabricated claims about “dried herbs and mushrooms as high-Cd ingredients” — removed. Rewritten to reflect what the paper actually contributes (Polish retail occurrence data for ten named commodities across four processing forms, with specific MPL exceedances).
• Legacy ## Wiki pages updated on ingest section removed per the current Part 6 source-page template.

Freq-1 ingredient flag: cranberries / cranberry is not in the current taxonomy snapshot (no wiki/ingredients/cranberry.md or cranberries.md exists). The paper measures 13 cranberry samples across fresh, dried, and processed forms; a provisional ingredient scaffold via tools/autonomy/create-provisional-ingredient-scaffold.mjs would let this page route to a cranberry ingredient destination on a future ingest. Flagged for the autonomy daemon to handle on its next provisional-scaffold sweep; not creating during this ingest per the per-skill constraint that source-page ingest does not create new ingredient pages by hand.

Wiki pages this source may touch

• cadmium
• lead
• apple
• pear
• grapes
• raspberries
• strawberries
• beet
• celery
• carrot
• tomato
• fresh-fruit
• dried-fruit
• canned-fruit
• fruit-juice-not-canned
• root-tuber-vegetables
• canned-vegetables
• eu-1881-2006-contaminants-superseded

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.