Chengappa et al. 2025 — Systematic review of heavy metals in over-the-counter toothpastes (11 in vitro studies)

This Frontiers in Dental Medicine PRISMA 2020 systematic review compiles in vitro evidence on the presence and concentration of heavy metals in over-the-counter toothpastes from 11 primary studies published between 2013 and 2023, screened from 9,409 database records via Google Scholar, PubMed, Scopus, Springer Link, Embase, and Web of Science. All 11 included studies reported the presence of one or more heavy metals — toxic (Cd, Pb, tHg, tAs, Ag) and essential (Co, Zn, Cr, Cu, Fe, Ni) per the review’s classification — in the toothpastes analysed. Eight of the eleven studies used atomic absorption spectrophotometry (AAS; several specified flame AAS in Table 1); two used ICP-MS; one used microwave plasma-atomic emission spectrometry (MP-AES). Quality assessment with the Quality Assessment Tool for in vitro Studies (QUIN) returned a high risk of bias for seven of the eleven studies and a medium risk for the remaining four (no study met the low-risk threshold). The review identifies regulatory-limit fragmentation across jurisdictions and a near-absence of toothpaste-specific occurrence data from North America, Europe (outside Malta), East Asia, and South America as the primary literature gaps.

Key numbers

Systematic-review identification and screening (PRISMA 2020 flow diagram, Figure 2, p. 4):

• Records identified from databases: 9,409 (Google Scholar 8,730; PubMed/MEDLINE 376; Scopus 210; Springer Link 69; Embase 12; Web of Science 12). Registers: 0.
• Records removed before screening: 315 duplicates (0 by automation tools; 0 for other reasons).
• Records screened (title): 9,094. Excluded at title screen: 9,035.
• Reports sought for retrieval: 59. Reports not retrieved: 41.
• Reports assessed for eligibility: 18. Reports excluded for not meeting PICOS criteria: 7.
• Studies included in review: 11.
• Other-methods pathway (websites, organisations, citation searching): 0 records identified, 0 included.
• Inter-rater reliability (Cohen’s kappa): 1.0 (perfect agreement between the two primary reviewers KC and AR; third reviewer RS resolved disagreements).

Eligibility criteria (PICOS, p. 3):

• Population: over-the-counter toothpaste samples.
• Intervention/exposure: none.
• Comparison: any or no comparison.
• Outcome: presence and types of heavy metals.
• Study design: in vitro studies.
• Exclusions: non-English studies, studies lacking full texts or abstracts, randomized clinical trials and other non-PICO designs.

Characteristics of the 11 included studies (Table 1, p. 5):

Per-study concentration findings as reported in §5.3 Heavy metals identified (pp. 4-5):

• Odukudu et al. 2013, Nigeria, n=11 (Table 1; narrative p. 4): Cd 0.035 ppm and Pb 0.02 ppm reported as the toxic-metal concentrations. Zn, Cr, Cu, Fe, Ni reported present as essential metals.
• Ideriah et al. 2016, Nigeria, n=9 (narrative p. 4): Fe present in all toothpastes; Cr detected in 5 of 9 with concentration <0.006 ppm.
• Orisakwe et al. 2016, Nigeria, n=35 (narrative p. 4): Pb detected in all 35 samples; Cd detected in 26 samples. Daily intake of Cd estimated as “high for one sample.” Essential metals: Co and Ni in all 35 samples; Cr in 24 samples.
• Salama 2016, Saudi Arabia, n=4 (narrative p. 4): Cd 2.008–55.28 ppb across the four toothpastes; Pb 1,856–6,313 ppb; tAs 0.6–26.94 ppb; tHg 3.34–13.14 ppb in three of four samples. Essential metals Co, Cr, Cu, Ni present in all four samples.
• Vella & Attard 2019, Malta, n=9 (narrative p. 4): Pb 2.37–12.04 ppm in all samples; Ag 2–5.12 ppm in all samples. Essential Cr, Cu, Ni in all 9 samples; Zn in 8 of 9 samples. Cd and Hg not detected in any sample.
• Arshad et al. 2020, Pakistan, n=18 (narrative p. 4 and Discussion p. 6): Cd detected in 11 of 18 samples; concentration range 0.0125–1.392 ppm (p. 4) / 0.0037–1.39 ppm (p. 6). The two ranges differ in the lower bound (0.0125 vs 0.0037 ppm) — see Verification notes.
• Paul et al. 2020, Bangladesh, n=10 (narrative p. 4 and Discussion p. 6): Pb 0.27–2.12 ppm in all toothpastes; tAs 0.02–0.637 ppm in all toothpastes; Cu found in all toothpastes (range not stated in the review).
• Ogidi & Agbo 2021, Nigeria, n=5 (narrative pp. 4-5): Zn range 1.19–3.08 ppm in three toothpaste samples; very high values of 81.27 ppm and 84.67 ppm in the other two samples.
• Almukainzi et al. 2022, Saudi Arabia, n=2 (narrative p. 5): Cd 8.8–9.19 ppm; Pb 75.86–78.31 ppm; tAs 209.33–221.96 ppm in both samples (the review reports these values without unit qualification beyond ppm; see Verification notes regarding consistency with the primary study). Essential Co, Zn, Cr, Cu, Fe, Ni at “very high” concentrations in both samples.
• Valentine & Ozioma 2022, Nigeria, n=10 (narrative p. 5; the review’s Table 1 reports n=5, body text reports n=10): Cd 0.01–0.13 ppm in two of the toothpastes; Pb 0.01–0.02 ppm in two of the toothpastes. Zn and Fe in 8 of 10 samples; Cr and Ni in 6 of 10 samples; Cu in 5 of 10 samples. (Table 1 vs body text n discrepancy — see Verification notes.)
• Lawi et al. 2023, Iraq, n=10 (narrative p. 5): Pb 1.0–12.05 ppm in all 10 samples; Zn 1.59–402.34 ppm; Fe 36.75–654.24 ppm.

Cited regulatory limits the review compiles (Discussion, pp. 5-6):

Quality assessment (QUIN tool, Table 2, p. 6):

• 9 of 12 QUIN criteria applied (the review excluded #4 details-of-comparison-group, #6 operator-details, and #7 randomization as irrelevant to in vitro samples without comparison groups).
• Score range across the 11 studies: 33.33%–55.55%.
• Risk-of-bias distribution: 7 high-risk (Ideriah 44.4%, Orisakwe 38.8%, Salama 38.8%, Arshad 38.8%, Paul 38.8%, Ogidi & Agbo 33.3%, Almukainzi 38.8%), 4 medium-risk (Odukudu 55.55%, Vella & Attard 55.55%, Valentine & Ozioma 55.55%, Lawi 55.55%). No low-risk study.
• Cohen’s kappa between the two reviewers: 1.0.

Methods (brief)

Design: PRISMA 2020-compliant systematic review of in vitro studies on heavy-metal occurrence in over-the-counter toothpastes, prospectively registered at OSF Registries (https://doi.org/10.17605/OSF.IO/V9P2D) under Institutional Research Committee approval 685/2024 (Manipal College of Dental Sciences Mangalore, Manipal Academy of Higher Education).

Search: six databases — Google Scholar, Embase, Springer Link, PubMed, Web of Science, Scopus — plus manual reference-list searching. Boolean search string: (“heavy metals” AND toothpastes) OR (“heavy metals” AND dentifrices). Filters: English language; sort by relevance. Full search strategy in Supplementary Table 1.

Screening and data extraction: two reviewers (KC, AR) screened titles and abstracts and extracted data into a Covidence-customized template, with a third reviewer (RS) adjudicating disagreements. Data items extracted: presence, type, and concentration of heavy metals; methodology; instrumentation; and the analytical method per study.

Quality assessment: Quality Assessment Tool for in vitro Studies (QUIN; Sheth et al. 2022, ref 16), which assigns 12 criteria each scored 0 (not specified), 1 (inadequately specified), or 2 (adequately specified). For this review, 9 of 12 criteria were applied; criteria 4 (details of comparison group), 6 (operator details), and 7 (randomization) were excluded as irrelevant to in vitro samples without a comparison arm. Final percentage scores classified as <50% high-risk, 50–69.9% medium-risk, 70%+ low-risk (the review uses the QUIN-tool defaults).

Limitations the authors declare (§6.1, p. 7): English-language-only inclusion (language bias); evidence available only from Nigeria, Malta, Bangladesh, Pakistan, Saudi Arabia, and Iraq; 7 of 11 studies high-risk-of-bias; in vitro design limits the ability to assess real-world human-body effects; absence of brand-name reporting limits applicability of the findings to consumer guidance.

This source is a secondary synthesis (systematic review) and not a primary occurrence study; the concentration values it transports from the 11 included studies should be cross-checked against the primary publications when used for HMTc threshold work. Two such primary studies are independently ingested in the wiki and supersede the review’s summary numbers for synthesis purposes: chengappa2025-toothpaste-india-aas (the same lead author’s companion AAS primary study on 20 Indian toothpastes, doi 10.1038/s41598-025-12748-3) and almukainzi2022-saudi-topical-cosmetics-icpms (the Almukainzi 2022 ICP-MS study).

Implications

Certification (HMTc): The systematic review’s principal contribution to the toothpaste (Cat 2 Row 11) evidence base is corpus-mapping: it documents that every published in vitro toothpaste-metal study to date (2013–2023, 11 studies, 118 individual toothpaste samples) reported presence of at least one heavy metal, and that no published low-risk-of-bias study exists in the in vitro toothpaste corpus. The per-study concentration ranges the review transports (Pb 0.02 ppm to 78.31 ppm; tAs 0.0006 ppm to 221.96 ppm; Cd 0.0037 ppm to 55.28 ppm) span 3-4 orders of magnitude, primarily attributable to different analytical platforms (flame AAS, ICP-MS, MP-AES) and different sample-preparation procedures rather than to true between-jurisdiction concentration differences. The review’s regulatory-limit compilation (FDA, WHO, EU, Health Canada, India BIS) is useful as a starting point for the toothpaste regulatory-context section of HMTc Cat 2 Row 11, with primary-source verification recommended before any threshold-setting use. No threshold values are proposed on this page.

Courses: Useful as a corpus-mapping teaching case for the Cat 2 toothpaste module, particularly the geographic distribution of toothpaste-metal research (heavy concentration in West Africa, the Middle East, and South Asia; near-absence in North America, Europe outside Malta, East Asia, and South America) and the analytical-platform sensitivity question (3-4 orders of magnitude range across studies, driven by ICP-MS vs flame-AAS sensitivity differences). The QUIN-assessed risk-of-bias distribution (7 of 11 high-risk, 0 low-risk) is a useful teaching anchor for the importance of method validation, replicate analysis, and reference-material reporting in occurrence studies.

App: Not directly applicable. Toothpaste is outside the consumer-app scope (focussed on children’s foods and personal care). The review may inform consumer-app context paragraphs on toothpaste as a daily-use personal-care product with documented heavy-metal contamination across multiple jurisdictions.

Wiki pages this source may touch

• lead
• cadmium
• arsenic-total
• mercury-total
• nickel
• chromium
• cobalt
• copper
• zinc
• iron
• silver
• aluminum
• manganese
• toothpaste
• eu-ec-1223-2009-cosmetics (proposed; not yet in regulation taxonomy)
• india-bis-6356-toothpaste (proposed; not yet in regulation taxonomy)

Verification notes

• Brand firewall (Part 12 strict reading, 2026-05-17): the systematic review itself does not name individual toothpaste brands in its body text; it cites the primary studies’ authors and per-study aggregate findings. The review explicitly notes in its Limitations (§6.1, p. 7): “the absence of specific brand names of toothpastes in some studies could affect the applicability of the results.” This page reports only per-study aggregates and ranges; no brand attribution is reproduced.
• Speciation discipline: the included primary studies report total arsenic and total mercury without speciation. This page records tAs and tHg per CLAUDE.md Part 14. Chromium is recorded as total Cr (no included study reports Cr-VI separately).
• Paper-internal discrepancy on Arshad et al. 2020 Cd range (recorded, not adjudicated): §5.3 (p. 4) reports the Arshad et al. Cd range as 0.0125–1.392 ppm; §6 Discussion (p. 6) reports the same range as 0.0037–1.39 ppm. The review does not flag which is the original primary-study value. Synthesis pass should consult the Arshad et al. 2020 primary publication.
• Paper-internal discrepancy on Valentine & Ozioma 2022 sample size (recorded, not adjudicated): Table 1 (p. 5) reports n=5 for Valentine & Ozioma; the §5.3 narrative (p. 5) describes “their study of ten toothpastes in Nigeria, Valentine and Ozioma (21) reported the presence of the toxic heavy metals cadmium and lead in two of the ten toothpastes.” Five-vs-ten cannot both be right. Synthesis pass should consult the Valentine & Ozioma 2022 primary publication.
• Paper-internal discrepancy on Almukainzi et al. 2022 concentration units (recorded, not adjudicated): the §5.3 narrative reports Cd 8.8–9.19 ppm, Pb 75.86–78.31 ppm, tAs 209.33–221.96 ppm. The independently-ingested Almukainzi 2022 primary study (almukainzi2022-saudi-topical-cosmetics-icpms) is the canonical source for these values; the review’s transport of those concentrations is included here as the review reports them. Synthesis should use the primary source.
• Regulatory-limit transport: the review transports cosmetic and toothpaste regulatory limits from FDA, WHO, EU, Health Canada, and India BIS via secondary citation. These should be verified against the primary regulatory documents before being placed on regulations/* pages or used in HMTc threshold work. The review attributes Pb=20 ppm to FDA cosmetics (ref 26) but the FDA’s published cosmetics testing page does not state a 20 ppm Pb limit for cosmetics generally; this is a transport-error candidate that the synthesis pass should reconcile.
• Evidence tier: A-tier per CLAUDE.md Part 13 (peer-reviewed systematic review, PRISMA 2020-compliant, prospectively registered at OSF). The QUIN-assessed quality of the included primary studies (7 high-risk, 4 medium-risk, 0 low-risk) is a downstream-weighting concern for synthesis, not a reason to downgrade the review itself.
• The lead author Chengappa K of this systematic review is the same author as the primary AAS study Chengappa et al. 2025 (chengappa2025-toothpaste-india-aas, Scientific Reports, doi 10.1038/s41598-025-12748-3). The systematic review (ref 14 in the primary study) was registered before and informed the design of the primary study; both are independently ingested. The primary study supersedes the systematic review’s transported Almukainzi values for the Saudi Arabia data point.
• Empty ingredients: [] and matrices: [cosmetic-personal-care] are intentional: the included studies measure metals as contaminants in finished cosmetic products (toothpastes), not in food-style ingredients. The cosmetic-personal-care matrices vocabulary follows the established convention across sibling source pages (chengappa2025-toothpaste-india-aas, jitareanu2025, rbeida2023, balogun2024, massarsky2025, arshad2020).
• The paper folder location (03_Contamination_Data within June 1 Infant cosmetics metals papers) is a manual-fetch curation choice; the systematic review itself does not specifically address children’s or infant oral-care products (one included study, Odukudu et al. 2013, examined “fluoride and heavy metals” in toothpastes; no included study restricted to children’s toothpaste). Routing to toothpaste (Cat 2) is appropriate.
• Generative AI statement (p. 8): “The author(s) declare that no Generative AI was used in the creation of this manuscript.”
• Funding statement (p. 8): “The author(s) declare that no financial support was received for the research and/or publication of this article.”
• Conflict-of-interest statement (p. 8): “The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.”
• Audit subagent (2026-06-02) flagged the WHO row in the regulatory-limit table as misattributing EU values (tAs 0.1 ppm, Cr 1.0 ppm) to WHO — verified against source p. 6: “WHO for lead at 10 ppm, cadmium at 0.3 ppm and mercury at 1 ppm (27). The EU has set limits of 0.5 for lead and mercury, 0.1 for arsenic and 1.0 ppm for chromium (7).” Corrected: split the WHO and EU values into two separate rows; WHO row now em-dashed for tAs and Cr.
• Audit subagent (2026-06-02) flagged “graphite-furnace AAS” in the opening paragraph as a methods-attribution slip — verified against source: PDF Table 1 specifies “Flame AAS” or generic “AAS” for the eight AAS studies; no included study uses graphite-furnace AAS. Corrected to “atomic absorption spectrophotometry (AAS; several specified flame AAS in Table 1)“.
• Audit subagent (2026-06-02) flagged the metals: frontmatter array as omitting Al and Mn — verified against source p. 5 Table 1 Salama row (“Lead, aluminium, cadmium, cobalt, chromium, copper, manganese, nickel, mercury, and arsenic”): both Al and Mn are present in the included corpus via Salama 2016. Added Al and Mn to the frontmatter metals: array and to the “Wiki pages this source may touch” list.
• Audit subagent (2026-06-02) flagged that [[regulations/eu-ec-1223-2009-cosmetics]] and [[regulations/india-bis-6356-toothpaste]] are not in the current regulation-page taxonomy — confirmed: both regulation pages do not yet exist in wiki/regulations/. Both are bona-fide regulations cited multiple times across the toothpaste corpus (e.g., on the companion source page chengappa2025-toothpaste-india-aas) and are appropriate candidates for new regulation pages via Karen’s regulation-page workflow. Marked as (proposed; not yet in regulation taxonomy) rather than removing the wikilinks, to preserve the routing signal for the future regulation pages.
• Audit subagent (2026-06-02) flagged the opening paragraph’s “essential (Co, Zn, Cr, Cu, Fe, Ni, Sn)” classification as adding Sn to the essential category, where the source p. 4 classifies only “Cobalt, zinc, chromium, copper, iron, and nickel” as essential (Sn is detected by Vella & Attard but not categorically classified in the source’s essential/toxic schema). Corrected to “essential (Co, Zn, Cr, Cu, Fe, Ni) per the review’s classification”.

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.