Sultana et al. 2022 — Heavy metals in root and leafy vegetables, Dhaka, Bangladesh

This study measured concentrations of eight metals (Cr, Cd, Pb, Ni, Cu, Zn, Fe, Mn) in four root vegetables (beet, radish, carrot, turnip) and five leafy vegetables (mustard, cabbage, spinach, coriander, mint) purchased at Kawran Bazar, Dhaka’s largest fresh produce market, across four seasonal sampling phases at minimum 14-day intervals over two months. Leafy vegetables consistently showed higher metal concentrations than root vegetables, with mean concentrations of Cr, Cd, Pb, and Ni exceeding FAO/WHO maximum permissible limits (MPL) in several species. Average daily intake (ADI), hazard quotient (HQ), and hazard index (HI) calculations indicated that four of five leafy vegetables (mustard, spinach, coriander, mint) and one of four root vegetables (turnip) posed potential health risks (HI > 1) at the assumed Bangladeshi adult intake of 166.1 g/person/day.

Key numbers

All concentrations on a dry-weight (dw) basis. Bold values in source Tables 2 and 3 indicate exceedance of the FAO/WHO MPL.

Table 2 — Root vegetables (mg/kg dw; phase 1 → 4 with overall mean):

Notes on Table 2:

• Source-internal anomaly: the Range column for Beet Cr is printed as 0.60–0.90 in the paper’s Table 2, but the per-phase values span 0.60–1.90. The mean 1.20 mg/kg dw is consistent with the per-phase data; the printed range is a paper-internal typo on the upper bound. Preserved here as the per-phase listing.
• The 49.60 mg/kg dw Cr value for turnip in the 4th (late) phase is anomalously high (>21× the MPL). The authors verify “we did the analysis several times and found a similar value” and attribute it to “the high use of preservatives at the end of the season” (page 6). The mean Cr value for turnip is dominated by this single phase.
• All root vegetables except turnip had mean Cd at or above the MPL (paper page 6: “Average cadmium (Cd) concentration exceeded the MPL in all root vegetables except turnip”). Mathematically beet’s mean of 0.20 mg/kg dw equals the MPL rather than strictly exceeding it; the paper’s phrasing is preserved.

Table 3 — Leafy vegetables (mg/kg dw; phase 1 → 4 with overall mean):

Notes on Table 3:

• The paper text on page 8 states that Mn concentrations did not exceed the MPL in any leafy vegetable. This is consistent with the mean values shown but not with the individual phase value of mustard 389.30 mg/kg dw — that phase is also below the 500 MPL, so the statement holds across all phases as well.
• Pb exceeded the leafy-vegetable MPL on a four-phase mean basis in all five leafy vegetables (page 8: “the average concentrations of Pb exceeded the MPL in all leafy vegetables”).

Table 4 — Average daily intake (ADI; mg/person·day) at 166.1 g vegetables/person/day, 60 kg body weight:

ADI exceedances of PMTDI (bold above):

• Turnip Cr: 0.202 mg/person·day vs. PMTDI 0.20 — driven by the late-season Cr spike (49.60 mg/kg dw, 4th phase).
• Mint Fe: 19.681 mg/person·day vs. PMTDI 17.0 — driven by accumulated Fe across all four phases (mean 800.6 mg/kg dw).

All other ADIs are below PMTDI at the 166.1 g/day Bangladeshi consumption assumption.

Hazard quotient (HQ) and hazard index (HI):

The authors compute HQ per metal per vegetable using RfD values (mg/kg/day): Pb 0.004, Zn 0.30, Cu 0.04, Cd 0.0005, Cr 0.0003, Ni 0.02, Fe 0.70, Mn 0.14 (FAO/WHO 2013). HI is the sum of HQs across the eight metals for each vegetable.

Explicit values from paper text (page 9):

• HQ > 1: Cr in turnip = 1.121; Zn in spinach = 1.104. All other per-metal HQs were < 1 in root vegetables. HQ values for individual leafy vegetable–metal pairs are not enumerated in the paper text.
• HI > 1 (potential cumulative health risk; abstract and page 9): turnip 1.541, mustard 1.663, spinach 2.113, coriander 1.925, mint 2.834.
• HI < 1 (text-explicit only): radish 0.362 (lowest among root vegetables); cabbage 0.38 (lowest among leafy vegetables).
• HI < 1 for beet and carrot: the paper text reports the order “mint > spinach > coriander > mustard > turnip > beet > carrot > radish > cabbage” (page 9). This order places beet and carrot HI between turnip (1.541) and radish (0.362). Specific numerical values for beet and carrot HI are not given in the paper text or in Table 4; Figure 2 displays per-vegetable HI graphically but exact bar heights for beet and carrot are not transcribed by the authors. The wiki captures only text-explicit values.

Source-internal contradiction (preserved): the paper states cabbage HI = 0.38 while also stating radish HI = 0.362 and the rank order “radish > cabbage.” If both numeric values are correct then cabbage (0.38) > radish (0.362), contradicting the printed rank. The authors do not reconcile this. Treated as a transcription typo somewhere in their text rather than a data-integrity stop condition.

Methods (brief)

Sampling: Vegetables collected from Kawran Bazar wholesale market, Dhaka, in 4 phases with minimum 14-day intervals over 2 months, capturing early, mid-time, and late harvesting periods. Three replicates per vegetable per phase were homogenized into composite samples for analysis. The market is described as the supply hub for Dhaka’s vegetable demand, sourced 60–70% from Manikganj, Tangail, Narsingdi, Munshiganj, Savar, Keraniganj, and Dhamrai districts.

Sample preparation: Samples washed with clean tap water then three times with distilled water; sliced and air-dried 2 h; dried in induction oven at ~65 °C for ~90 h; reweighed for dry-weight conversion. Moisture content (Table 1) ranged 85.2% (mint) to 95.3% (radish). Dried samples ground in pestle/mortar and sieved through 0.20 mm screen.

Digestion: 1 g oven-dried sample in a 100 mL beaker with 10 mL concentrated HNO₃ (Merck, analysis grade) + 2 mL concentrated HClO₄ (Merck) on hot plate at 150–200 °C; cooled, filtered, made up to 100 mL with DI water; stored in plastic bottles.

Analysis: Atomic Absorption Spectrophotometer (Varian AA 240) with hollow cathode lamps. Wavelengths: Cr 357.9 nm, Cd 228.8 nm, Pb 283.3 nm, Ni 232 nm, Cu 324.7 nm, Zn 213.9 nm, Fe 248.3 nm, Mn 279.5 nm. Reported detection limits (µg/L): Cr 1.7963, Cd 0.1986, Pb 100, Ni 3.5473, Cu 20.0, Zn 5.1260, Fe 30.0, Mn 10.0. Sensitivity checked with intermittent standards. The Pb LOD of 100 µg/L is unusually high for AAS (graphite-furnace AAS typically reaches sub-µg/L for Pb; the paper does not specify atomization mode beyond the instrument model), which is consistent with the high frequency of 0 mg/kg Pb readings across root vegetables and early-phase leafy vegetables.

Risk equations: ADI = (Av_consumption × %DW × C_metal); %DW computed from moisture content per vegetable; Av_consumption = 166.1 g/person·day (Ibrahim and Mollah 2011 household survey); body weight = 60 kg (FAO/WHO 1993). HQ = ADI / RfD. HI = Σ HQ across measured metals.

Statistics: ANOVA with Duncan’s Multiple Range Test (DMRT) in IBM SPSS Version 20, α = 0.05. Significance letters (a, b) shown next to means in Tables 2 and 3 indicate DMRT groupings within each metal column.

Speciation: None — total metals only. Paper does not measure inorganic vs total arsenic, methylmercury vs total mercury, hexavalent chromium, organotins, or any other speciated form. As, Hg, Sn, Al, Sb, U are not measured at all.

Implications

Certification: Root and leafy vegetables sourced from Dhaka wholesale market show Cr, Cd, and Pb mean exceedances of FAO/WHO MPLs in multiple species, with the largest signals in coriander Cd (mean 1.13, range up to 1.80 mg/kg dw), mint and coriander Pb (means 2.25 and 2.00), and the single-phase turnip Cr anomaly (49.60 mg/kg dw, 4th phase). These are seasonal-market means from one wholesale hub and one calendar window; downstream certification work should treat them as occurrence data points for South Asian leafy-green and root-vegetable supply, not as benchmark values for global supply.

Courses: Useful as a worked example of seasonal-harvesting variability — the paper’s late-season (4th phase) values for several metals are markedly higher than early-season values, which the authors attribute to higher use of preservatives and possibly to lower water availability concentrating metals in plant tissues. Demonstrates how a single late-phase outlier (turnip Cr 49.60 mg/kg dw) can drive a four-phase mean above the MPL even when three of four phases are below the limit.

App: Dry-weight basis for all values; to convert to fresh weight (as-consumed), multiply by (100 – %moisture)/100 per vegetable. Spinach and mint show high Cd, Pb, and Fe accumulation; mint specifically drives the only Fe ADI exceedance in the study (Fe ADI 19.681 vs PMTDI 17.0 mg/person·day) at the 166.1 g/day Bangladeshi adult intake assumption. Risk classification at population scales other than Bangladesh requires re-deriving the consumption rate.

Wiki pages this source may touch

• chromium
• cadmium
• lead
• nickel
• copper
• zinc
• iron
• manganese
• root-vegetables
• leafy-vegetables
• leafy-greens
• spinach
• carrots
• beet
• cabbage

Coriander, mint, mustard greens, radish, and turnip do not have dedicated ingredient pages in the current taxonomy. Mustard (Brassica campestris) and mint/coriander route via leafy-greens / leafy-vegetables; turnip and radish route via root-vegetables. Auto-stub at freq ≥ 2 will surface these slugs if the corpus accumulates sufficient species-specific evidence; freq-1 ingredient scaffolds are out of scope for this ingest.

Verification notes

• 2026-05-20 — Merge-enhance pass against the source PDF. Prior page version (updated 2026-05-13) had several defects that have been corrected here:
  • raw_handle: was the generic manual-fetch-kimi; updated to the specific MFK_heavy-metals-in-commonly-consumed-root-and-leafy-v form per current convention.
  • raw_path: was truncated to …Dh.pdf; corrected to the full filename …and Assessment of Associated Public Health Risks.pdf.
  • metals: previously listed [Pb, Cd, Cr, Ni] — missing Cu, Zn, Fe, Mn, which the paper measures and tables. Expanded to [Pb, Cd, Cr, Ni, Cu, Zn, Fe, Mn] (eight metals per Tables 2, 3, and 4).
  • ingredients: expanded with [[ingredients/leafy-greens]] (the routing destination for mustard greens), [[ingredients/beet]], and [[ingredients/cabbage]] (each exists as a current ingredient slug and is directly measured in Tables 2/3).
  • products: previously empty ([]), triggering the products-missing routing advisory in routing_malformed.csv. Populated with [spinach, leafy-vegetables-other, root-tuber-vegetables] — the three current product slugs that this paper provides direct occurrence data for. No new product pages were created (Step 0 Lock remains Karen’s call).
  • Numerical correction (Check 1, Beet Ni mean): prior page reported Beet Ni mean = 0.58 mg/kg dw. Table 2 of the source (page 5) reports Beet Ni phase values 0.00 / 0.00 / 0.00 / 1.50 with overall mean 0.38 a, which is also the arithmetic mean of the four phase values. Corrected to 0.38.
  • Numerical fragility (HI for non-exceeding vegetables): prior page reported beet HI = 0.56 and carrot HI = 0.34. These values are not given anywhere in the paper text or in Tables 1–4; only Figure 2 displays them graphically and not labelled with numerals. Replaced with text-explicit-only HI values for radish (0.362) and cabbage (0.38). Beet and carrot HI are noted as not enumerated; the paper’s rank order places both between turnip (1.541) and radish (0.362). The prior beet/carrot numerals appear to be Figure 2 visual estimates rather than text-anchored values.
  • Methods AAS mode: prior page asserted “(AAS flame)” for the Varian AA 240. The paper does not state atomization mode beyond model number. The Varian AA 240 family includes both flame (AA240FS) and graphite-furnace (AA240Z) variants; the reported Pb LOD of 100 µg/L is closer to flame-mode performance and the reported Cd LOD of 0.20 µg/L is closer to graphite-furnace performance, so the paper’s instrument operation cannot be unambiguously assigned. Softened to “the paper does not specify atomization mode beyond the instrument model.”
  • Body-table format: prior version compressed Tables 2 and 3 into mean-only summaries. Expanded to full phase-by-phase tables matching the source so downstream synthesis can read every per-phase value and verify which phases drive each exceedance. Added Table 4 (ADI) verbatim, including PMTDI references; prior version paraphrased only the two exceedances.
  • HQ values added: prior version mentioned only HI; added the two explicit HQ > 1 values (Cr in turnip 1.121, Zn in spinach 1.104) per page 9.
  • Source-internal paper anomalies preserved and labelled (not wiki defects):
    • Beet Cr range printed as 0.60–0.90 in Table 2 while phase values span 0.60–1.90 (paper typo on upper bound).
    • Cabbage HI vs radish HI rank-vs-numeric contradiction (rank says radish > cabbage but printed values give cabbage 0.38 > radish 0.362).
    • Spinach Pb mean printed as 0.67 mg/kg dw while phase values (0, 0, 0.68, 0) average to 0.17; preserved as the paper-reported mean with the phase listing intact so the downstream reader can see the discrepancy.
  • Replaced legacy ## Wiki pages updated on ingest heading with ## Wiki pages this source may touch per current schema.
• Brand firewall (Part 12): no brand names appear in the source or this page. The paper samples market-level vegetables at Kawran Bazar without brand or vendor attribution; method instruments (Varian AA 240, Merck reagents, IBM SPSS) are scientific-method vendor names, permitted per the 2026-05-17 locked reading of Check 4.
• Wiki/HMTc firewall (Part 2): Implications section describes the paper’s contribution as occurrence data for South Asian leafy-green and root-vegetable supply and as a worked example of seasonal harvesting variability. No threshold value is proposed; no consumer risk advisory is issued. The prior version’s “Leafy vegetable risk scoring for South Asian sourcing should weight Cr, Cd, and Pb higher than for other origins” was softened in this pass to remove the prescriptive framing while preserving the geographic-context observation.
• Speciation discipline (Part 14): the paper measures total Cr (AAS, no Cr-VI separation) and does not measure As or Hg. metals: [Cr, ...] is correct (not Cr-VI); no speciation flags required.
• Routing: matrices: [root-vegetable, leafy-vegetable] retained as the dominant matrix-slug convention for vegetable-occurrence sources in the corpus (25 / 21 instances vs. 9 / 3 for the plural variants per the 2026-05-20 sweep). products: [spinach, leafy-vegetables-other, root-tuber-vegetables] routes this source as direct occurrence evidence for the current product pages that cover the species measured.

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.