Mania et al. 2021 — Lead, cadmium, arsenic, mercury, and tin in Polish fruit and fruit products (2015 monitoring)

This paper reports results from a 2015 Polish national monitoring program, analyzing approximately 600 samples across multiple fruit categories (fresh, frozen, dried, preserved, and canned) for Pb, Cd, As (total), Hg (total), and Sn (canned products only). Contamination levels were generally low, broadly comparable to other EU member states, and estimated dietary exposures for all five elements were below EFSA and JECFA tolerable doses for both adults and children. The highest single-sample values were Pb in dried raisins from Iran (0.150 mg/kg) and Sn in canned pears from China (206 mg/kg, still below the EU ML of 250 mg/kg for canned fruit). At 90th-percentile (MB) contamination of fresh fruits, children’s lead intake reached 31.5% of the BMDL₀₁ for neurotoxicity.

Key numbers

Sample composition: 269 fresh fruits (165 berries/small fruits — raspberries, strawberries, currants, blackberries, gooseberries, cranberries, blueberries, grapes — and 104 other fruits), 52 dried, 59 frozen (39 berries and small fruits; 20 other), 222 preserves (67 berry preserves, 83 other-fruit preserves, 72 canned fruits). Pb determined in 530 samples, Cd in 603, tAs in 554, tHg in 559, Sn in 69 (canned products only).

Lead, fresh berries and small fruits: mean (LB–UB) 0.008–0.017 mg/kg; P90 (LB–UB) 0.026–0.038 mg/kg. Fresh fruits other than berries: mean (LB–UB) 0.006–0.015 mg/kg; P90 (LB–UB) 0.022–0.029 mg/kg. Frozen berries and small fruits: mean (LB–UB) 0.011–0.022 mg/kg; P90 reported as “0.037–0.003 mg/kg” in the source — the upper-bound figure is an apparent typographical error (UB cannot fall below LB; the value is preserved verbatim with this caveat). Frozen fruits other than berries: mean (LB–UB) 0.004–0.015 mg/kg; P90 (LB–UB) 0.018–0.026 mg/kg. Dried fruits: mean (LB–UB) 0.021–0.034 mg/kg; P90 (LB–UB) 0.042–0.099 mg/kg. Domestic preserves from berries and small fruits had higher mean lead (0.018 mg/kg MB) than imported (0.008 mg/kg MB). Preserves from other fruits: domestic mean 0.009 mg/kg, imported 0.017 mg/kg MB. Canned fruits: domestic mean 0.020 mg/kg, imported 0.015 mg/kg MB. Highest values: red grapes from Peru 0.084 mg/kg; dried raisins from Iran 0.150 mg/kg.

Cadmium, fresh berries and small fruits: domestic mean (LB–UB) 0.005–0.006 mg/kg; imported 0.000–0.001 mg/kg. Fresh fruits other than berries (domestic): mean (LB–UB) 0.002–0.003 mg/kg. Frozen berries and small fruits and frozen other fruits: mean (LB–UB) 0.010–0.011 mg/kg. Dried fruits: domestic mean (LB–UB) 0.007–0.008 mg/kg (P90 0.018 mg/kg); imported mean (LB–UB) 0.0014–0.004 mg/kg (P90 LB–UB 0.0001–0.003 mg/kg). Preserves from berries: mean (LB–UB) 0.002–0.003 mg/kg (P90 0.007 mg/kg). Preserves from other fruits: mean (LB–UB) 0.001–0.002 mg/kg (P90 LB–UB 0.001–0.003 mg/kg). Canned products: mean (LB–UB) 0.0003–0.002 mg/kg; P90 (LB–UB) 0.000–0.003 mg/kg.

Arsenic (total), full-range across categories: mean (LB–UB) 0.001 mg/kg in fruit preserves to 0.021 mg/kg in frozen berries and small fruits; 90th percentile range 0–0.050 mg/kg. Dried fruits: domestic mean (LB–UB) 0–0.016 mg/kg; imported 0.007–0.021 mg/kg; P90 (LB–UB) domestic 0–0.026 mg/kg, imported 0.023–0.032 mg/kg. Highest single-sample values: frozen domestic strawberries 0.123 mg/kg; fresh kiwi from Spain 0.09 mg/kg.

Mercury (total), fresh berries and small fruits: mean (LB–UB) 0.0004–0.0021 mg/kg; P90 (LB–UB) 0.001–0.005 mg/kg. Other fresh fruits: mean (LB–UB) 0.0003–0.0017 mg/kg; P90 (LB–UB) 0.0001–0.004 mg/kg. Dried fruits: mean (LB–UB) 0.0009–0.003 mg/kg; P90 (LB–UB) 0.002–0.005 mg/kg. Highest value: imported dried apples 0.016 mg/kg. Left-censored data were 100% below LOD for mercury in frozen fruits.

Tin (canned products only, n=69): mean (LB–UB) 69.7–70.1 mg/kg; P90 (LB–UB) 140–140 mg/kg; maximum 206 mg/kg in canned pears from China (EU ML for canned fruit and vegetables = 250 mg/kg); minimum 2.39 mg/kg in sliced tomatoes of domestic origin. Left-censored data were 7% below LOD for tin in canned fruits.

Exposure estimates (mean MB intake, fresh fruits): lead 3.3% of BMDL₁₀ (adult nephrotoxicity), 1.4% of BMDL₀₁ (adult cardiovascular), 14.6% of BMDL₀₁ (child neurotoxicity). At 90th-percentile (MB) contamination of fresh fruits, lead intake reached 7.13%, 2.9%, and 31.5% of those three BMDL values respectively. Lead from fruit preserves at mean contamination: 0.5% of BMDL for adult cardiovascular/nephrotoxicity; 1.3–2.1% of BMDL₀₁ for children. Lead from fruit preserves at 90th percentile: 0.12% (adult cardiovascular) to 1.3% (child neurotoxicity) of BMDL₀₁.

Cadmium from fresh fruits at mean (MB): 2.4% of TWI (adults, EFSA) to 8.5% of TWI (children); 1.04–3.6% of PTMI (JECFA). At 90th percentile (MB): 6.8% TWI (adults) to 23.7% TWI (children); 2.9–10.2% of PTMI. Cadmium from fruit preserves at mean: 0.3–1.2% TWI; 0.1–0.5% PTMI. At 90th percentile: 0.7% TWI (adults) to 2.5% TWI (children); 0.3–1.1% PTMI.

Arsenic from fresh fruits at mean (MB): 0.5% of BMDL₀.₅ (adults), 1.6% of BMDL₀.₅ (children). At 90th percentile (MB): 0.9–3.2% of BMDL₀.₅. From fruit products: ≤0.5% of BMDL at mean; 0.2–0.5% at 90th percentile.

Mercury (treated as inorganic for exposure calculation): mean MB intake from fresh fruits 0.5% of TWI (adults) to 1.4% of TWI (children). At 90th percentile (MB): 1.1–3.2% of TWI. Below 0.5% of TWI for fruit products at both mean and 90th percentile.

Methods (brief)

Pb and Cd determined by flame atomic absorption spectrometry (FAAS) or by flameless electrothermal atomization AAS (GFAAS) depending on matrix and concentration range. Total arsenic determined by hydride generation AAS (HGAAS). Total mercury determined by cold vapor AAS (CVAAS). Tin measured only in canned products (those packaged in metal cans, where Sn migration from packaging is the relevant exposure route). Sample preparation followed the European Standard PN-EN 13804:2013. State Sanitary Inspection laboratories with accredited and validated methods participated; internal QC used reference materials with certified metal content; external proficiency was checked through interlaboratory tests organized by the Food Safety Institute NIH NIH-NRI. Left-censored data (LC, below LOD/LOQ) were treated by lower-bound (0), middle-bound (½ LOD/LOQ), and upper-bound (LOD/LOQ) substitution following EFSA’s left-censored-data guidance; figures in the paper present middle-bound (MB) values.

Limitations stated by the authors and implied by the methods: total arsenic was measured (not inorganic arsenic); HGAAS does not speciate, so the iAs exposure calculation assumes the total-As reading is a conservative proxy for iAs. Total mercury was measured (not methylmercury); for exposure calculation the authors assume mercury in fruits is essentially inorganic, supported by EFSA’s position that organic mercury contributes negligibly outside fish and seafood matrices. Sn was measured only in metal-packaged products. Sample geography is Poland-only (domestic plus imports landed in Poland).

Implications

Certification: Pb in dried fruits (mean up to 0.034 mg/kg MB; P90 up to 0.099 mg/kg) is the most significant contamination signal in this dataset; the single highest value was 0.150 mg/kg in dried raisins from Iran. Sn in canned fruit (mean ~70 mg/kg, P90 140 mg/kg, max 206 mg/kg in canned pears from China) is well below the EU ML of 250 mg/kg for canned fruit but is non-trivial for cumulative exposure modeling. At 90th-percentile MB contamination of fresh fruits, child lead intake reached 31.5% of the BMDL₀₁ for neurotoxicity, which is the steepest BMDL utilization figure in the paper.

Courses: Useful baseline monitoring dataset for EU-origin fresh, frozen, dried, preserved, and canned fruit. Illustrates the total-vs-inorganic-arsenic measurement problem for regulatory contexts where HGAAS is used (total-As reading as a proxy for iAs exposure). Also illustrates the EFSA left-censored-data convention (LB/MB/UB substitution).

App: Mean and P90 values across multiple fruit categories and processing states (fresh, frozen, dried, preserved, canned) for Pb, Cd, tAs, and tHg; Sn for canned only. Suitable for ingredient-level risk estimates in EU-origin product modeling. Berries and small fruits are tracked as a distinct subgroup throughout, which is useful for variant-level estimates.

Wiki pages this source may touch

• lead
• cadmium
• arsenic
• mercury
• tin
• strawberries
• raspberries
• blueberries
• grapes
• apple
• pear
• raisins
• fruit
• fresh-fruit
• canned-fruit
• dried-fruit
• eu-1881-2006-contaminants-superseded
• eu-2023-915-cadmium
• efsa-lead-contam-2010
• efsa-cadmium-twi
• jecfa-cadmium-ptmi
• efsa-mercury-twi
• efsa-arsenic-contam-2009

Verification notes

• 2026-05-20 enhancement: corrected truncated raw_path to the actual PDF filename; removed invented slugs (ingredients/currants, ingredients/dried-fruit, ingredients/apples, products/fruit-preserves, products/frozen-fruit) that are not in the taxonomy; added ingredients/apple (singular form, paper reports imported dried apples 0.016 mg/kg Hg, highest mercury value), ingredients/pear (paper reports canned pears from China 206 mg/kg Sn, highest tin value), ingredients/raisins (paper reports Iranian dried raisins 0.150 mg/kg Pb, highest lead value), and umbrella ingredients/fruit. Replaced legacy ## Wiki pages updated on ingest section with the current ## Wiki pages this source may touch form. Expanded ## Key numbers to include the previously-omitted preserves-by-origin breakdown, the Sn minimum, the 90th-percentile exposure figures for Pb in fruit products, the Cd and Hg 90th-percentile exposure figures, and the explicit imported-vs-domestic dried-fruit Cd breakdown.
• Paper-internal anomaly preserved verbatim: Figure 1 / text reports “frozen berries and small fruits Pb P90: 0.037–0.003 mg/kg” — UB below LB is a typographical error in the published paper. Flagged but not silently corrected; the underlying figure values cannot be recovered from the PDF.
• Matrices frozen-fruit and fruit-preserves are not in the common-matrices list from the GPT system prompt; preserved because they accurately describe the form measured and the matrices vocabulary is open per the system prompt’s guidance. New-term flag for taxonomy review.
• Mercury treated as inorganic for exposure calculation is the authors’ explicit modeling choice (paper page 359), consistent with EFSA’s position that organic mercury contributes negligibly to non-fish dietary exposure. The metals: field uses tHg because measurement was total-Hg by CVAAS, not speciated.

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.