Amer et al. 2019 — Heavy metal exposure assessment in Egyptian apples, grapes, and oranges

This study measured Pb, Cd, Cr, Cu, and Ni concentrations in 108 fresh fruit samples (apples, grapes, oranges; 36 per fruit type, 9 per governorate per fruit) collected from four governorates in Egypt (Cairo, Giza, Alexandria, El-Fayoum). Lead and cadmium were not detected in any sample. Copper and nickel were present in all samples, frequently above WHO/FAO maximum residue limits. Chromium was detected only in grape samples from Cairo and El-Fayoum at concentrations more than ten times the EU permissible limit. Washing grapes reduced Cu and Ni by up to ~97%; peeling oranges reduced Cu by up to ~81%. Estimated daily intake for Cu and Ni exceeded reference tolerable daily intake values in most fruit-governorate combinations.

Key numbers

Apples (n=36 total, 9 per governorate; AAS, dry-ashed at 540°C; LODs: Pb 0.1, Cd 0.02, Cu 0.03, Ni 0.1, Cr 0.06 mg/kg):

Grapes (n=36 total, 9 per governorate):

Oranges (n=36 total, 9 per governorate):

Mitigation by washing (grapes): Ni reduced by ~28% (Cairo), ~97% (Giza), ~31% (El-Fayoum), ~75% (Alexandria); Cu reduced by ~12% (Cairo), ~29% (Giza), ~58% (El-Fayoum), ~55% (Alexandria). (Values read from Fig. 1.)

Mitigation by peeling (oranges): Cu reduced by 81.36% (Cairo), 12.56% (Giza), 24.07% (Fayoum), 54.32% (Alexandria); Ni reduced 9.67% (Cairo), 22.61% (Giza), and no reduction in Fayoum or Alexandria (the source states “no reducing effect on Ni” for those two governorates without giving a numerical value). The authors interpret this as Cu being predominantly surface-bound on orange peel and Ni being more systemically absorbed in some regional samples.

Estimated daily intake (EDI) vs. tolerable daily intake (TDI) for adults (assumed BW 70 kg; WHO GEMS/Food Middle East regional consumption: apples 7.5 g/person/day, grapes 15.8 g/person/day, oranges 31.5 g/person/day; reference TDI per USEPA Region III: Cr 0.003, Cu 0.040, Ni 0.020). EDI exceeded TDI in:

• Apples: Cu (Cairo, Alexandria, Giza); Ni (Cairo, El-Fayoum).
• Grapes: Cr (Cairo, El-Fayoum); Cu (all four governorates); Ni (all four governorates).
• Oranges: Cu (all four governorates); Ni (all four governorates).

The paper’s Table 4 reports both EDI and TDI in units it labels “mg/kg/day” but the magnitudes are consistent with µg/(kg body weight·day) given the input data and formula; the EDI > TDI comparisons inside the paper remain internally consistent because both sides use the same magnitude convention.

Methods (brief)

Atomic absorption spectrophotometry (Agilent Technologies 200 Series AA) with cathode-lamp settings: Cd 228.8 nm/0.5 nm slit, Cr 357.9 nm/0.2 nm, Cu 324.8 nm/0.5 nm, Pb 283.0 nm/1.0 nm, Ni 232.0 nm/0.2 nm. Sample prep: 5 g homogenized fresh fruit dried at 105°C for 16 h, dry-ashed at 540°C for 8–12 h, dissolved in 1.0 mL 36% HCl, filtered (Whatman No. 40 ashless), made up to volume, analyzed by AAS. Reference standards from Merck (Darmstadt), prepared from 1000 mg/kg stock in 0.1 N HNO3. Limits of detection (mg/kg): Pb 0.1, Cd 0.02, Cu 0.03, Ni 0.1, Cr 0.06. Basis convention not explicitly stated, but reporting against MRLs in mg/kg and the starting material being fresh homogenized fruit indicates fresh weight (wet weight). Only total metals were measured; no arsenic species (iAs vs tAs) or mercury species (MeHg vs tHg) were reported, and Al, Sn, Sb, U were not in scope.

EDI computed as Ci × IR / BW with Ci in mg/kg, IR in g/person/day, and BW = 70 kg. Statistical analysis: one-way ANOVA with Fisher’s Protected LSD for between-governorate comparison; significance at the 1% level (per table footnotes).

Implications

Certification: Cu and Ni MRL exceedances in Egyptian-origin fresh fruit are relevant for sourcing audits. Pb and Cd were below LOD (0.1 and 0.02 mg/kg respectively) in all 108 samples; this informs but does not by itself establish a regional clean-platform claim for Pb/Cd because the AAS LOD for Pb (0.1 mg/kg = 100 ppb) is well above modern ICP-MS LODs (~1 ppb) used in HMTc-relevant comparator studies. Washing and peeling are documented mitigation levers with substantial but incomplete reduction.

Courses: Useful example of how washing and peeling reduce surface-bound metals (Cu and surface-Ni predominantly), and of the limits of these interventions when metals are systemically absorbed (Alexandria grape Ni at 1.78 mg/kg is not plausibly all surface).

App: Cu and Ni present across all 12 fruit-governorate cells; Pb and Cd not detected; Cr only in grapes from Cairo and El-Fayoum. Within-fruit-type variation across governorates is approximately 10x for Cu and 6x for Ni, large enough that regional sourcing matters for category-level risk modelling.

Verification notes

• 2026-05-20 merge-enhance pass: fixed invalid ingredient slug apples → apple (canonical singular slug per current taxonomy; wiki/ingredients/apple.md exists, apples.md does not); restored full raw_path (previously truncated to frui.pdf); added per-governorate n=9 detail; added explicit wet-weight basis note (source does not state explicitly but starting material is 5 g fresh homogenized fruit reported against fresh-weight MRLs); preserved cite_key, raw_handle, license, DOI, near_duplicates fields; added explicit EDI/TDI exceedance enumeration from Table 4; expanded washing/peeling mitigation numbers from Figs. 1–2. Paper-internal note: the abstract and main text state Giza apples Cu = 0.33 mg/kg whereas Table 1 reports 0.55 ± 0.14 mg/kg; this page follows the table value (table reports per-replicate mean ± SE and is the authoritative datum). The paper’s Table 4 labels EDI/TDI units as “mg/kg/day” but the magnitudes are µg/(kg BW·day); EDI > TDI comparisons remain internally consistent within the paper’s convention.
• 2026-05-20 audit-subagent pass (verdict REVISE): one ⚠️ applied — peeling-oranges Ni line reworded from “0% (Fayoum), 0% (Alexandria)” to “no reduction in Fayoum or Alexandria” to match the source’s “no reducing effect on Ni” phrasing (PDF p. 540 §3.2) without overstating numerical precision. One ❌ rejected as a false positive: the subagent flagged ingredients/oranges as not present in docs/gpt-collaboration/taxonomy-snapshot.md, but the snapshot’s Ingredients line does include oranges (verified: ...orange-juice,oranges,organ-meats,...) and wiki/ingredients/oranges.md exists on disk — slug is valid in both taxonomy and on-disk inventory. Checks 1 (numerical fidelity), 3 (speciation/methods), 4 (brand firewall), 5 (HMTc firewall) all clean.

Wiki pages updated on ingest

• apple
• grapes
• oranges
• fresh-fruit
• copper
• nickel
• chromium

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.