Winiarska-Mieczan et al. 2023 — Cd and Pb in Polish culinary herbs and spices

This Polish study measured cadmium and lead in 432 samples of culinary herbs and spices purchased in Lublin, eastern Poland: 100 dried herb samples across 9 species, 184 fresh herb samples across 15 species, and 148 single-component spice samples across 14 species. Analysis was by graphite furnace AAS (GF-AAS). Mean Cd was highest in dried herbs (0.134 mg/kg fresh weight), far above fresh herbs (0.004 mg/kg) and spices (0.017 mg/kg). Mean Pb was also highest in dried herbs (0.548 mg/kg), compared to fresh herbs (0.039 mg/kg) and spices (0.064 mg/kg). All calculated exposure metrics (% TWI, % BMDL, CDI, THQ, HI) were well below 1 at typical Polish consumption of 0.7 g/day of herbs and 0.7 g/day of spices. However, Cd exceeded the EU maximum level in dried coriander (mean 0.419 mg/kg) and estragon (mean 0.427 mg/kg), and Pb exceeded the limit in several fresh herbs: watercress, jiaogulan, celery, basil, and dill.

Key numbers

Concentrations expressed in mg/kg fresh weight:

Dried herbs (n=100):

• Cd: mean 0.134 ± 0.168, max 0.493, median 0.047, p25 0.047; LOQ 0.0002 mg/kg
• Pb: mean 0.548 ± 0.161, max 0.913, median 0.555, p25 0.555
• Cd exceeded 0.3 mg/kg EU limit in: dried coriander (mean 0.419 mg/kg) and estragon (mean 0.427 mg/kg); 20% of coriander and 30% of estragon samples exceeded

Fresh herbs (n=184):

• Cd: mean 0.004 ± 0.007, max 0.028, 19% below LOQ
• Pb: mean 0.039 ± 0.033, max 0.122, 8% below LOQ
• Pb exceeded 0.3 mg/kg fresh herb limit in: watercress, jiaogulan, celery, basil, dill

Single-component spices (n=148):

• Cd: mean 0.017 ± 0.019, max 0.057, 32% below LOQ; no sample exceeded 0.3 mg/kg limit
• Pb: mean 0.064 ± 0.050, max 0.160, median 0.044; no sample exceeded limit
• Highest Cd spices: cumin > lemon pepper > black pepper > cinnamon > ginger
• Highest Pb spices: allspice > cayenne pepper > pink peppercorn > ginger > black pepper

Dried vs. fresh comparison (per dry weight basis): Cd and Pb were both significantly higher (P<0.05) in dried than fresh herbs, attributed to perennial vs. annual plant age, phosphate fertilizer use in commercial growing, and concentration during drying.

Regulatory limits applied (from PDF p. 3572 col 1 and p. 3573 col 2; the paper compares against both national and EU limits):

• Fresh herbs (Polish national MPL, Journal of Laws 13 Jan 2003, ref [38]): 0.05 mg/kg Cd, 0.3 mg/kg Pb
• Fresh herbs (EU, Commission Regulation EC 1881/2006, ref [56]): 0.2 mg/kg Cd
• Dried herbs and spices containing more than 50% herbs (Polish national MPL, Journal of Laws 27 Dec 2000, ref [40]): 0.3 mg/kg Cd, 2.0 mg/kg Pb

Exposure/risk:

• At 0.7 g/day dried herbs: EWI Cd = 6.57×10⁻⁴ µg/kg bw, corresponding to 3.75×10⁻⁴ % of TWI (2.5 µg/kg bw/week); EWI Pb = 2.69×10⁻³ µg/kg bw, corresponding to 3.65×10⁻⁴ % of BMDL01
• THQ for all products < 1; HI (Cd+Pb) < 1 for all categories

Methods (brief)

GF-AAS (graphite furnace atomic absorption spectroscopy, Varian Spectr AA 880). Sample digestion: dry ashing at 450°C for 12 hours with H₂O₂ as oxidant, then dissolved in 1M HNO₃. LOD 0.001 mg/kg Cd, 0.011 mg/kg Pb; LOQ 0.0002 mg/kg Cd, 0.003 mg/kg Pb. Reference materials: INCT-TL-1 (tea leaves) recovery 103% Cd, 99% Pb; INCT-MPH-2 (mixed Polish herbs) recovery 94% Cd, 102% Pb. Replication deviation ≤5% (each analysis run in triplicate); duplicate-measurement deviation 5.8% Cd / 4.5% Pb; reproducibility 10.1% Cd / 13.9% Pb (Table 2). Statistical analysis: ANOVA with Duncan’s test, P<0.05. Fresh herbs were dried at 65°C for 24 h before analysis; reported both as fresh-weight and dry-weight equivalents.

Implications

Certification: Contributes occurrence data for Cd in dried herbs (n=100 across 9 species) and Pb in fresh herbs (n=184 across 15 species) sampled from Polish grocery retail. Two single-species dried herbs — coriander (mean Cd 0.419 mg/kg, 20% of samples above limit) and estragon (mean Cd 0.427 mg/kg, 30% above limit) — exceeded the applicable Polish national MPL of 0.3 mg/kg Cd for dried herbs. Five fresh herbs — watercress, jiaogulan, celery, basil, dill — exceeded the Polish national Pb MPL of 0.3 mg/kg in fresh herbs. The paper attributes elevated Cd in commercial dried herbs to soil phosphate-fertiliser use and to the older mean age of perennial commercial herb plants relative to the annual fresh herbs sampled here.

Courses: Clean illustration of the dried vs. fresh concentration effect: the same herb species shows ~30-fold higher Cd per fresh weight when dried, because moisture is removed but metal content is retained. This is critical for label-comparison literacy — a consumer seeing a mg/kg limit on dried herbs cannot directly compare it to a fresh-herb limit without a water-content conversion.

App: Dried herbs are a distinct ingredient class from fresh herbs in the risk model. The app should not apply fresh-herb contamination profiles to dried herb ingredients. The ingredient matrix field must distinguish these (dried-herbs vs. fresh-herbs), and contamination profiles should be maintained separately.

Wiki pages updated on ingest

• herbs-and-spices
• dried-herbs
• fresh-herbs
• cadmium
• lead
• eu-1881-2006-contaminants-superseded

Verification notes

Merge-enhance pass 2026-05-29 (manual-fetch ingest, MFK condiment_papers/05_PB_Vanilla_Spices):

• Re-verified all Table 3 per-category statistics (mean, max, median, p25, %<LOQ) against the source for dried herbs (n=100), fresh herbs (n=184), and spices (n=148) on Cd and Pb in mg/kg fresh weight; key numbers in this page match the source.
• Re-verified per-species exceedances against Figs. 2–4: Cd > 0.3 mg/kg dry-matter MPL in dried coriander (mean 0.419) and dried estragon (mean 0.427); Pb > 0.3 mg/kg fresh-herb MPL in watercress, jiaogulan, celery, basil, and dill (Fig. 3b); no spice sample exceeded its applicable Pb MPL. Matches source.
• Re-verified exposure metrics against Table 4: EWI Cd 6.57×10⁻⁴ µg/kg bw for dried herbs corresponding to 3.75×10⁻⁴ % TWI; EWI Pb 2.69×10⁻³ µg/kg bw corresponding to 3.65×10⁻⁴ % BMDL01 and 8.72×10⁻⁴ % BMDL10; all THQ and HI(Cd+Pb) < 1. Matches source.
• Re-verified analytical figures of merit and CRM recoveries against Table 2; all values match.
• Frontmatter defects corrected: raw_handle changed from generic manual-fetch-kimi to per-paper MFK_the-content-of-cd-and-pb-in-herbs-and-single-compo; raw_path corrected from truncated ...Use.pdf (file did not resolve) to the full filename ...Used in Polish Cuisine.pdf; raw_sha256, access_url, and tier_rationale added; sampling_locations populated with Lublin grocery retail per the Material and Methods section.
• Paper-internal inconsistencies noted (also captured in tier_rationale) and do not affect downstream synthesis: the methods narrative lists 15 fresh-herb species summing to 187 samples while the abstract and Table 3 report n=184; and the methods narrative names “pink peppercorn (n=8)” for the same single-component spice that Table 1 lists as “Rose pepper / Schinus terebinthifolius (n=10)” (botanically the same species).
• Brand-firewall check: source reports only species-level and category-level means; no brand-by-brand attribution is present in the paper, so no firewall-violating content to remove.

Audit subagent pass 2026-05-29 (v2.0 fresh-context general-purpose Agent, verdict REVISE): four findings applied, three rejected as not actionable.

Applied:

• ❌ Broken slug [[regulations/eu-reg-1881-2006]] corrected to [[regulations/eu-1881-2006-contaminants-superseded]] — verified via ls wiki/regulations/; the eu-reg-* form does not exist for 1881/2006.
• ⚠️ Regulatory attribution corrected: 0.05 mg/kg Cd in fresh herbs and 0.3 mg/kg Pb in fresh herbs are Polish national MPLs (PDF p. 3572 col 1, ref [38]: Polish Journal of Laws 13 Jan 2003); the EU 1881/2006 value for Cd in fresh herbs is 0.2 mg/kg (ref [56]). Prior wiki text incorrectly attributed 0.05 Cd to EU 1881/2006.
• ⚠️ Methods (brief) “Reproducibility ≤5%” was a conflation. PDF Table 2 reports two distinct metrics — duplicate-measurement deviation 5.8% Cd / 4.5% Pb and reproducibility 10.1% Cd / 13.9% Pb — plus PDF p. 3570 text saying replication deviations did not exceed 5%. Methods rewritten to list all three.
• ⚠️ Implications/Certification rewritten to contribute occurrence data and report the paper’s own causal hypothesis (phosphate-fertiliser and plant-age) rather than prescribe HMTc analyte-flagging or supplier-spec levers. Part 2 firewall: literature side, not program-design side.

Rejected (verified false-positive or not actionable):

• ⚠️ “p25 = median” presentational note for dried herbs Cd (0.047) and Pb (0.555). Both percentile values match Table 3 verbatim; the duplication is an artefact of the source distribution, not a wiki error. No edit.
• ⚠️ Matrices vocabulary check (dried-herbs, fresh-herbs, spices). taxonomy-snapshot.md does not define a matrices controlled list; the three slugs are in active corpus use across many sources. Not actionable from this PDF alone.
• ⚠️ No product-page route for fresh/dried herbs (~284/432 samples). This is a routing-coverage gap that would require Karen-approved Part 10 new-page proposals (no dried-herbs or fresh-herbs product page exists); not a per-source defect.

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.