Alharbi et al. 2023 — As, Cd, Pb in Saudi Arabian baby foods (NFMP 2020)

Saudi Food and Drug Authority (SFDA) study quantifying total arsenic, cadmium, and lead by ICP-MS in 111 baby food products from the 2020 National Food Monitoring Program, then calculating estimated daily intake (EDI), target hazard quotient (THQ), hazard index (HI), and margin of exposure (MoE) for infants 0–6 months and 7–12 months. Samples were drawn from pharmacies and main markets in Riyadh, Jeddah, and Dammam, with the sampling frame weighted by market share. Concentrations are reported in µg/kg as-sold (dry weight for powders and cereal meals). Left-censored data are handled with the Codex CCCF lower-bound / middle-bound / upper-bound (LB/MB/UB) substitution scheme. The paper concludes that exposure from the foods tested poses low chronic noncarcinogenic risk (all THQ and HI < 1.00) and lead exposure for 0–6-month infants on stage 1 formula alone reaches the EFSA neurodevelopmental BMDL01 at the upper-bound estimate (MoE = 1.00).

Key numbers

Mean concentrations by food category (middle-bound, µg/kg as-sold; Figure 1 reports mean ± SD):

Per the Abstract and Results §3.1, the highest mean As was in cereal-based meals (15.5 µg/kg), the highest mean Cd was in biscuits (8.76 µg/kg), and the highest mean Pb was in biscuits (53.8 µg/kg).

Middle-bound concentration ranges by category (µg/kg as-sold), Results §3.1:

• Stage 1 infant formula: As 0.35–15.0; Cd 0.50–60.0; Pb 3.80–119.
• Stage 2 infant formula: As 0.35–8.20; Cd 0.50–7.30; Pb 3.80–104.
• Cereal-based meals: As 0.35–179; Cd 0.50–28.0; Pb 3.80–255.
• Biscuits: As 0.35–53.0; Cd 0.50–21.9; Pb 3.80–223.

Statistical comparisons (Kruskal–Wallis across groups; Mann–Whitney pairwise):

• Stage 1 vs stage 2 infant formula: no significant difference for any metal (As p[MB] = 0.258; Cd p[MB] = 0.335; Pb p[MB] = 0.858).
• Across all four baby food groups: As p[MB] = 0.120 (NS); Pb p[MB] = 0.091 (NS); Cd p[MB] < 0.001 (significant).
• Cd pairwise significance: infant formula vs cereal-based meals p[MB] < 0.001; infant formula vs biscuits p[MB] < 0.001; cereal-based meals vs biscuits p[MB] = 0.007.

Estimated Daily Intake (EDI), µg/kg bw/day, LB–MB–UB (Table 2):

Target Hazard Quotient (THQ) and Hazard Index (HI), Table 3:

All THQ and HI values are below 1.00; the paper interprets this as “low potential chronic risks” for noncarcinogenic effects of As and Cd.

Margin of Exposure (MoE) for Pb vs EFSA BMDL01 = 0.50 µg/kg bw/day (developmental neurotoxicity), Table 4:

The upper-bound MoE of 1.00 for the 0–6-month group means upper-bound Pb intake from stage 1 formula alone equals the BMDL01 for developmental neurotoxicity. EFSA considers MoE ≥ 10 indicative of low risk; every MoE in this dataset falls below that threshold.

Per-food-group contribution to total dietary exposure (7–12 mo group):

• As: cereal-based meals 29.1 % of total intake; stage 2 formula 9.18 %; biscuits 3.32 %.
• Cd: cereal-based meals 8.08 %; stage 2 formula 5.89 %; biscuits 2.19 %.
• For 0–6-month infants fed stage 1 formula exclusively, formula contributes 19.5 % of total As intake and 21.4 % of total Cd intake at the scenario consumption rate.

Reference values cited by the paper:

• USEPA RfD: As (inorganic) 0.30 µg/kg bw/day (cardiovascular); Cd 1.00 µg/kg bw/day.
• EFSA BMDL01 for Pb: 0.50 µg/kg bw/day (developmental neurotoxicity, embryos / infants / children ≤7 years).
• JECFA: Cd PTMI 25 µg/kg bw/month (≈0.83 µg/kg bw/day); Pb PTWI withdrawn 2011 (associated with ≥3-IQ-point decrement in children); As PTWI 3 µg/kg bw/day withdrawn (no longer health-protective vs lung-cancer BMDL0.5).
• IARC classifications: As and Cd Group 1 carcinogens; Pb Group 2B.

Body weights and label-recommended consumption used in EDI (Methods §2.3.1, §2.3.2):

• 0–6 mo: 5.45 kg; stage 1 formula 112 g/day (5 scoops × 4.50 g × 5 feeds; recommended 5–6 feeds/day).
• 7–12 mo: 8.90 kg; stage 2 formula 129 g/day (7 scoops × 4.60 g; recommended 3–5 feeds/day); cereal-based meals 50 g/day (one meal); biscuits 8 g/day (one 8-g stick).

Methods

Sampling. SFDA National Food Monitoring Program (NFMP), 2020. 111 commercial baby food products, selected by market share, drawn from pharmacies and main markets across three Saudi cities (Riyadh, Jeddah, Dammam). Domestic vs imported split: 21 / 90. Samples preserved at room temperature and sent to corresponding SFDA laboratories in each region.

Sample preparation. 200 g per sample homogenized to fine powder with a Knief Mill GM 200 (Retsch) titanium homogenizer. 0.50 g aliquots weighed into quartz vessels with 5 mL ultrapure 65 % HNO₃ (Romil Pure Chemistry), digested in a Milestone UltraCLAVE pressure-controlled microwave system (60 → 200 °C ramp over 30 min, 120 bar). Digestate diluted to 50 mL with Milli-Q grade 1 water (Millipore), giving a 100× dilution. A 6 mL aliquot was pipetted to disposable ICP-MS vials without filtration.

Analysis. Quadrupole ICP-MS with collision-cell technology (Agilent 7700). RF power 1500 W; plasma gas 15.0 L/min; auxiliary gas 1.00 L/min; helium collision gas 4.50 mL/min. Monitored masses: As 75, Cd 111, Pb 208. Both no-gas and helium modes evaluated; results approved in helium mode. Multi-element standard 10.0 mg/L (As, Cd, Pb) and internal standard 100 mg/mL from Agilent. Validation to CSN EN 15763 and BS EN 13804:2013 (European Standards 2023a, 2023b).

Method validation (n = 8 replicates per fortification, Table 1): LOD As 0.70 µg/L, Cd 0.90 µg/L, Pb 6.90 µg/L (in solution); LOQ As 2.40 µg/L, Cd 3.00 µg/L, Pb 23 µg/L. Linear dynamic range 0–100 µg/L with R² = 1.00 for each analyte. Recovery: As 116 %, Cd 106 %, Pb 108 % (within the Eurachem 70–120 % acceptance band). RSD: As 6.42 %, Cd 3.28 %, Pb 3.83 % (within the < 20 % acceptance band). Selectivity assessed via blank samples (no analyte response). QC controls per batch: solvent blank, fresh calibration standards, spiked samples, certified reference material.

Speciation. Total As was measured; the paper does not separate inorganic from organic arsenic. THQ for As was computed using the USEPA inorganic-arsenic RfD (0.30 µg/kg bw/day) applied to total-arsenic intake — the paper treats this as a conservative simplification.

Risk-assessment equations (Methods §2.3):

• EDC (g/day) = scoops × scoop size (g) × feeds per day.
• EDI (µg/kg bw/day) = (C [µg/kg] × CONS [g/day]) / (1000 × BW [kg]).
• THQ = EDI / RfD; HI = ΣTHQ across the assessed analytes.
• MoE = BMDL01 / EDI.

Censoring. Codex CCCF (CL 2021/78-CF) LB/MB/UB substitution: LB treats < LOQ or < LOD as zero; MB treats < LOQ as ½ LOQ and < LOD as ½ LOD; UB treats < LOQ as LOQ and < LOD as LOD.

Statistics. Microsoft Excel 2016 for descriptive statistics; SPSS Version 21 (IBM) for inferential tests. Levene’s test for homogeneity of variance; Kruskal–Wallis (data not following normal distribution) for across-group comparisons; Mann–Whitney for pairwise. Significance threshold p ≤ 0.05.

Implications

Cereal-based meals and biscuits drive heavy-metal exposure more than infant formulas in the Saudi market sample. The paper identifies cereal-based products as the food group “commonly associated with elevated Pb contents,” and reports the highest mean concentrations for As (cereal-based meals 15.5 µg/kg), Cd (biscuits 8.76 µg/kg), and Pb (biscuits 53.8 µg/kg).

Stage 1 infant formula is the dose-limiting product for 0–6-month-olds. Upper-bound Pb EDI from stage 1 formula alone reaches 0.50 µg/kg bw/day — equal to the EFSA developmental-neurotoxicity BMDL01. The corresponding MoE of 1.00 falls an order of magnitude below the EFSA “low risk” threshold (MoE ≥ 10). LB-MB MoEs (2.38 / 1.67) are also below that threshold.

All THQ and HI values for As and Cd remain below 1.00 for both age groups under the paper’s consumption-and-body-weight assumptions, which the paper interprets as low chronic noncarcinogenic risk for these two metals from the foods tested. The paper notes that this conclusion depends on label-recommended portion sizes and Ministry of Health body weights, and may not capture exposure from drinking water used for reconstitution, baby juices, gruels, ready-to-eat purees, teething products, or homemade complementary foods (which were not in the sampling frame).

As contribution from cereal-based meals (29.1 % of total intake at 7–12 months) is the largest single-source contribution among the four food groups assessed. Stage 2 formula adds 9.18 % and biscuits 3.32 %.

Speciation caveat. Because total As was measured and the USEPA iAs RfD was applied to total intake, the paper’s THQ for As is conservative against any RfD that targets inorganic As specifically; it does not, however, allow per-species exposure to be estimated from the dataset.

Wiki pages this source may touch

• arsenic-total
• cadmium
• lead
• infant-formula-powder
• baby-cereals-dry
• infant-formula-powder
• baby-cereals-dry-rice-based
• baby-cereals-dry-non-rice
• teething-and-snacks-rice-based
• teething-and-snacks-non-rice
• efsa-lead-contam-2010
• jecfa-cadmium-ptmi
• jecfa-lead-ptwi-withdrawn
• epa-iris-cadmium-rfd
• epa-iris-inorganic-arsenic-rfd

Verification notes

• Duplicate consolidation (2026-05-18). Two wiki source pages previously covered this paper: this page (alharbi2023-baby-foods-saudi-arabia-heavy-metals.md, prior revision 2026-05-12) and original2020-original-a.md (prior revision 2026-05-14). The latter cite-key was a manifest text-mining artifact (first author parsed as “original” from “ORIGINAL ARTICLE”; year mined incorrectly as 2020 from cited material), and the page itself carried suggested_cite_key_rename: alharbi2023-baby-foods-heavy-metals-saudi-arabia as a rename flag. This page is the canonical surviving cite-key; the duplicate has been removed in the same commit.
• Re-verified against the manuscript PDF (2026-05-18). Source consulted: raw/Manual Fetch Kimi /06_Infant_Foods_Formula/06_Infant_Foods_Formula/Occurrence and dietary exposure assessment of heavy metals in baby foods in the Kingdom of Saudi Arabia.pdf (DOI 10.1002/fsn3.3485, Food Sci Nutr 2023;11:5270–5282). The retired original2020-original-a page transposed the As means between cereal-based meals (15.5 µg/kg) and biscuits (11.1 µg/kg); the correct attribution is restored here.
• Speciation discipline (CLAUDE.md Part 14). Paper measures total As; frontmatter metals: uses tAs, not iAs. The risk-assessment THQ applies the USEPA iAs RfD (0.30 µg/kg bw/day) to total As as a conservative simplification, which is methods detail rather than a speciation measurement.
• Brand firewall. No brand names appear in the manuscript’s results or in this page; SFDA NFMP data are reported in aggregate by product category. Domestic-vs-imported split is reported as a sampling-frame descriptor only.
• Routing scope. Stage 1 and stage 2 formulas are entered under the umbrella products/infant-formula-powder slug because the paper does not separate soy from non-soy. Cereal-based meals and biscuits are entered under both rice-based and non-rice slugs because the paper does not separate by grain type (the discussion notes rice is the dominant grain in Saudi infant cereals but does not stratify the dataset by grain).
• Body-weight and consumption assumptions. Body weights (5.45 kg at 0–6 mo; 8.90 kg at 7–12 mo) cite Ministry of Health Saudi Arabia 2021. Label-recommended consumption was used; the paper notes that consumption data specific to Saudi infants under one year is sparse and that label portions may not reflect actual intake — a documented limitation of the EDI estimates.
• Enhancement scope (2026-05-18, manual-fetch PDF re-ingest). Relative to the 2026-05-12 revision of this page: full LB/MB/UB EDI, THQ/HI, and MoE values added from Tables 2–4; analytical method detail expanded from Methods §2.2 (UltraCLAVE program, ICP-MS RF/gas settings, atomic masses, EN method references, recovery/RSD/LOD/LOQ); statistical-test results added; per-food-group exposure contributions added; reference-value context (JECFA, EFSA, IARC) added; legacy ## Wiki pages updated on ingest heading replaced with ## Wiki pages this source may touch; speciation note clarified per CLAUDE.md Part 14.
• Codex manual-fetch recheck (2026-05-18). Fresh-context audit returned REVISE for three cleanup items: unsupported n = 4 replicates per sample wording, nonstandard matrix terms, and the internal-standard unit. The PDF page image confirms the source prints 100 mg/mL; this page now preserves that unit, normalizes matrices to broad controlled terms, and softens the BMDL01/Pb wording to what the source states.

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.