Heavy metals in diaper creams

Abstract

The finished-product evidence for heavy metals in diaper creams is thin and one-sided. Four single-unit laboratory reports, all grey literature, measured lead, cadmium, arsenic and mercury by ICP-MS in zinc oxide diaper creams, ointments and pastes as received. Across the four units lead ran 1244.1, 1401.0, 3303.8 and 4657.1 ppb and cadmium 45.26, 55.50, 473.1 and 509.5 ppb; lead and cadmium were each above the reporting value in all four units. Arsenic was below 10.00 ppb and mercury below 5.00 ppb in every unit. Each figure is one retail unit of one lot, bought and shipped by a consumer advocate, and none of the reports prints a sample mass, digestion, calibration, recovery, detection limit or accreditation. These are leads, not a market rate.

No peer-reviewed survey of metals in finished diaper creams is on file. No non-zinc-oxide diaper cream has been measured for metals at all, so the clean-versus-contaminated comparison the product taxonomy asks for cannot be made from measured data. The zinc-pathway studies on file measure zinc, the major element of the active ingredient, not lead or cadmium, and no study on file measures lead or cadmium crossing skin from a diaper cream. That zinc evidence now divides sharply by barrier state: on intact adult skin a single 100 gram application of a 40 percent zinc oxide ointment moved mean serum zinc by 8.8 micrograms per decilitre, which its authors report as not statistically significant, while two separate single-infant reports describe serum zinc reaching 490 and 427 micrograms per decilitre during zinc oxide ointment use on eroded or dermatitic diaper skin. Two cases are not a rate, and the elements involved are not the elements the creams were found to contain. The regulatory position is now read from the operative United States order rather than from a proposal: the skin-protectant monograph in force permits zinc oxide at 1 to 25 percent of the product and above 25 to 40 percent only in an ointment, and sets no heavy-metal contaminant limit for the finished product. The limits that do exist attach to the zinc oxide ingredient, to a daily cutaneous drug dose, or to a product marketed as a cosmetic, each on its own basis. This synthesis reports what those documents support. It does not rank products now on the shelf, it does not convert the measured concentrations into a dose, and it does not set a certification limit.

Product anatomy

A diaper cream is a topical barrier product applied to the diaper area to prevent or treat diaper dermatitis. The dominant active class is zinc oxide, formulated as a cream, an ointment or a stiffer paste. The strengths that appear in both the regulatory text and the clinical literature are the same two clusters the measured units fall into: a review of diaper dermatitis describes marketed ointments at 13 percent and at 40 percent zinc oxide applied at diaper changes ¹⁶, and the operative United States order permits 1 to 25 percent in any dosage form and above 25 to 40 percent in an ointment only. Zinc oxide is the element source that the pathway evidence below is about: it is the ingredient present in bulk, and its principal element is zinc. That review measures no impurity metal, and its authors are employed by a consumer-products company, which is recorded because it bears on independence rather than because the strengths are in doubt.

A second class of products uses no zinc oxide. Plain petrolatum and plant-based preventatives, described on the non-zinc-oxide product page (Diaper Cream, Non-ZnO), occupy this row. They are relevant to a synthesis of metals in diaper creams for one reason and with one caveat: they are the natural comparison group for any claim that zinc oxide carries a metal load, and no metal measurement of any non-zinc-oxide diaper cream is on file. The comparison cannot be made from data. It can only be named as a gap.

One ingredient inventory now says something about what would have to be measured in that comparison group. A tally of labelled ingredients in 41 topical diaper preparations found botanical extracts in 30 of them and fragrance in 18 ¹⁷. That inventory measures no metal and was built to count potential allergens, not contaminants, so it establishes nothing about metal content. It does mean that a non-zinc-oxide diaper preparation is not usually a simple two-ingredient petrolatum base, and that a botanical extract is a plausible route by which lead could enter a cream with no mineral active at all, which is the mechanism developed for supplements on Dietary supplements and traditional herbal preparations are a distinct, under-surveilled heavy-metal exposure pathway whose contamination magnitudes reach thousands of times conventional food limits under materially weaker enforcement. Treating that as a hypothesis to test is the whole of what this inventory supports.

The finished product is what a caregiver applies, and it is the basis on which the four measurements were made. It is not the same object as the zinc oxide raw material, which carries its own ingredient specification, and it is not a daily dose. Keeping those three bases apart, finished product, ingredient, and daily exposure, is what the rest of this page depends on.

Measured evidence

Four finished-product datasets were opened, and they are all the same kind of document: a single retail unit of one product, relabelled with an internal lot number by a consumer-advocacy business, sent to a commercial laboratory, and analysed by ICP-MS for lead, cadmium, arsenic and mercury. The report image is appended to a blog article. The matrix is the finished zinc oxide cream, ointment or paste, as received. The results are total-element concentrations in the finished product, in ppb, where 1 ppm equals 1,000 ppb equals 1 mg/kg. Arsenic is total arsenic and mercury is total mercury; neither is speciated. These are grey literature, evidence tier C.

The four reports are single units, so they are listed, not pooled. A single unit is a point, not a distribution, and four points chosen by an advocate are not a sample of the category.

Across the four units, lead ran from 1244.1 to 4657.1 ppb and cadmium from 45.26 to 509.5 ppb. Lead was above the reporting value in four of four units and cadmium in four of four. Arsenic was below the 10.00 ppb reporting value and mercury below the 5.00 ppb reporting value in every unit. The lowest lead result was in the 13 percent product and one of the two highest was also a 40 percent product, so within these four units lead did not order itself by zinc oxide percentage. Four points cannot establish that relationship or rule it out.

The same advocate later published a comparison chart of eleven diaper products, one unit each. Four of the eleven charted values match the four laboratory reports above, rounded. The other seven have no laboratory report on file, the chart’s cadmium and lead figures for those seven are not verifiable against a report, and for several of them the chart does not state whether zinc oxide is present. The chart is therefore not used here for any concentration, range, detection rate or comparison, and it charts no product without zinc oxide, so it cannot compare zinc-oxide and non-zinc-oxide products either. The four verified units are the whole of the measured evidence.

The limitations are the same for all four reports and are load-bearing. Each figure is one unit from one lot. The unit was selected and shipped by an advocate, which is a purposive rather than a random sampling frame. None of the reports prints a sample mass, a digestion procedure, a calibration, a recovery, a detection limit beyond the bare reporting value, or a laboratory accreditation. The “less than” entries are left-censored reporting values, not measurements of zero. A single non-detect on total arsenic or total mercury does not establish that either metal is absent below the reporting value, and four detects of lead and cadmium in four purposively chosen units do not establish a category detection rate. Read as what they are, the reports show that lead and cadmium were present, at the printed concentrations, in these four specific units.

Zinc dermal pathway evidence

None of the pathway studies on file measures lead or cadmium. They measure zinc, the principal element of the zinc oxide active. They bear on a narrower question than the one a caregiver asks: whether a zinc oxide vehicle on skin delivers its element into or through skin, and whether barrier damage changes that. They do not answer whether lead or cadmium in a diaper cream reaches or crosses infant skin, because none of them measures lead or cadmium, and none of them uses a diaper cream on infant diaper-area skin.

On intact adult skin the delivered fraction is small, for systemic zinc as well as for skin zinc. Derry and colleagues ⁶ applied 100 grams of a 40 percent zinc oxide ointment in a white petrolatum base to intact skin in healthy men and measured serum zinc by flame atomic absorption: the mean moved from 107.3 plus or minus 5.32 to 116.1 plus or minus 5.02 micrograms per decilitre one hour after application, a rise of 8.8 micrograms per decilitre or 8.2 percent, which the authors report as not statistically significant. In the petrolatum-only control arm the mean fell from 115.2 plus or minus 5.85 to 103.5 plus or minus 5.32 micrograms per decilitre over the same hour. A second phase applied 15 grams daily for eight days to an occluded area of 10 by 15 centimetres; six patients on total parenteral nutrition had a pretreatment mean of 88.6 plus or minus 5.85 micrograms per decilitre, and the three who completed dosing are described as remaining relatively constant over ten days, without printed day-by-day means. This is the largest single zinc oxide dose in the file, on the strength that the highest-strength measured products use, and it did not move serum zinc detectably.

Skin zinc does rise, which is the distinction worth holding. Ågren ⁷ applied a 25 percent zinc oxide occlusive dressing to intact adult forearm skin for 48 hours and measured epidermal zinc of 35 micrograms per gram dry weight on control skin against 187 under the dressing, with blister-fluid zinc of 0.27 against 0.95 micrograms per millilitre, and an average flux to skin of 5 micrograms per square centimetre per hour. Release from that dressing in vitro was strongly pH dependent, falling by 0.54 plus or minus 0.01 milligrams per square centimetre at pH 5.4 against 0.04 plus or minus 0.04 at pH 7.4 over 48 hours. In a separate study Ågren ⁸ applied occlusive zinc oxide dressings to the forearms of five adults for 48 hours and measured epidermal zinc of 49 micrograms per gram dry weight under a hydrocolloid vehicle and 127 under a gum-rosin vehicle, with blister-fluid zinc of 0.29 and 0.68 micrograms per millilitre. Khabir and colleagues ⁹ dosed excised human abdominal skin with a 20 percent zinc oxide nanoparticle formulation for 48 hours and found exogenous zinc in the viable epidermis of 3.1 micrograms per gram dry weight, against an endogenous 13, with the dermis indistinguishable from blank and the particles themselves retained in the stratum corneum; zinc entered as ions, not particles. The European Scientific Committee on Consumer Safety ¹³, reviewing nano zinc oxide as a sunscreen ingredient, adopted a dermal-absorption fraction of 0.03 percent of the applied dose from its in vitro human-skin data. Read together, these say that a zinc oxide vehicle loads the epidermis several-fold above background while very little reaches blood through an intact barrier. All of them describe intact adult skin and a sunscreen or dressing vehicle, not a diaper cream on an infant. The pH dependence is worth carrying forward for a different reason: diapered skin runs at a higher pH than adjacent undiapered skin, so the release condition in the diaper area is not the condition these dressings were tested under.

Barrier damage is where the sources disagree, and the disagreement is left standing rather than resolved. Ågren’s rat model showed much greater systemic zinc uptake through full-thickness back wounds than through intact skin: serum zinc rose to 4.06 µg/mL under one vehicle against 0.95 µg/mL in controls, and the vehicle ranking that held on intact skin reversed on wounds. In a hairless-mouse model, UVB injury increased skin zinc uptake from topical zinc oxide nanoparticles ¹⁰, though the numeric values sit only in a supplementary figure not on file. Against these, the Scientific Committee reported no significant difference between damaged and intact skin, and no enhancement of zinc penetration on UVB-damaged skin, in the pig and human models it reviewed.

Two infant case reports now sit in this section, and how they should be read needs stating before the numbers. Each is one patient. Two cases are not an incidence, not a rate, and not a distribution, and nothing in either allows a statement about how often hyperzincemia follows zinc oxide ointment use. What two independent cases do is remove the possibility that the first was a reporting artefact, and one of them removes the confounder that limited the first.

Yasugi and colleagues ¹¹ followed a preterm infant with eroded diaper dermatitis whose serum zinc reached 427 µg/dL during concurrent zinc oxide ointment and oral zinc, then fell to 52 µg/dL after both were stopped, with a Naranjo causality score of 9. Both the damaged skin and the concurrent oral dose bear on that reading, and the oral dose means the topical route cannot be isolated from that case alone.

Fujita and colleagues ¹² reported a late-preterm infant, birth weight 1420 grams, treated with a 10 percent zinc oxide ointment at each diaper change from about day 12. Serum zinc measured on days 22, 29, 38, 44 and 49 of life was 77, 119, 490, 194 and 82 micrograms per decilitre, against a normal range the paper states as 58 to 74. The infant was fully breastfed and received no zinc supplements. That is the case’s significance: with no oral zinc to account for the rise, and a fall once topical use ended, the topical route is the remaining explanation in this patient. The ointment was 10 percent, the lowest strength in the file, and the peak was the highest serum zinc of the two cases. Copper was not measured, no zinc measurement of the ointment beyond its stated strength was made, and the stated normal range is given alongside the day-22 result without a further citation.

The barrier state, not the zinc oxide percentage, is what separates these results. A 40 percent ointment at a 100 gram dose on intact adult skin did not move serum zinc significantly; a 10 percent ointment on eroded infant diaper skin accompanied a serum zinc of 490 micrograms per decilitre. Age, dose per kilogram, occlusion and skin integrity all differ between those two settings and none of the sources separates them, so this is a contrast to explain rather than a demonstrated mechanism. These results are about zinc, in wounds, erosion, UVB injury and intact skin, in rats, mice, adults and two infants. They are a reason to measure the dermal route for the metals that were actually found in the creams. They are not a measured fraction for lead or cadmium, and this synthesis does not treat them as one.

A narrative review of the infant skin barrier ¹⁴ supplies the background that makes the diaper area a plausible place to ask the question: the infant barrier approaches adult properties at about two years, the diaper area is occluded and over-hydrated, and an infant’s skin surface area is large relative to body weight. The review also notes that data on the effectiveness of zinc oxide ointment in preventing diaper dermatitis are limited. It reports no metal concentration and no dermal-absorption fraction for any metal.

A pooled analysis of four diaper trials in 1601 infants ¹⁵ puts measurement behind two of those conditions. Diapered skin was wetter, as transepidermal water loss, and ran at a higher pH than adjacent undiapered skin on the same infants, and rash grades were higher where wetness or pH sat in the elevated group, with wetness the stronger association at each grading period. That paper reports no metal concentration. It is cited for one inference only: the diaper area differs from the forearm skin used in every intact-skin absorption study above on both hydration and pH, so those studies were not run under diaper-area conditions. The direction that difference would push is not the obvious one. The single pH-dependence measurement in the file has zinc release from a zinc oxide dressing falling from 0.54 to 0.04 milligrams per square centimetre as pH rises from 5.4 to 7.4, the second figure carrying an uncertainty equal to its own value, so release was near the limit of what that experiment could distinguish from none. Higher pH means less zinc released, not more, and diapered skin is the higher-pH condition. Whether any of that holds for lead or cadmium in a cream is unmeasured, no source here tests it, and elevated pH and elevated hydration could plainly act in opposite directions. The point is that the diaper area is a different release environment, not that it is a worse one.

Toxicology

Toxicology here is the hazard each measured metal is known for, and what, if anything, is known about that metal on infant skin from a diaper cream. A reference dose or a classification derived from ingestion or inhalation is not a dermal dose from a cream, and none of the four reports measured absorption, blood levels or any biological endpoint. The reports are concentrations in a product.

Lead is a developmental neurotoxicant, and the endpoint that dominates infant and child assessment is the developing nervous system. Lead was the metal found at the highest concentrations in the four units, from 1244.1 to 4657.1 ppb. What is not on file is the step from that concentration to a dose: no measurement of how much cream is applied, how much lead leaves the vehicle, or what fraction crosses intact or dermatitic infant skin. Cadmium’s long-term dietary endpoint is kidney tubular dysfunction, on an oral, cumulative record; it was detected in all four units, from 45.26 to 509.5 ppb, and again with no dermal-transfer measurement on file. Arsenic’s regulated hazard is carried by the inorganic species, and mercury’s neurodevelopmental hazard by methylmercury; the reports measured total arsenic and total mercury, both below the reporting value in all four units, so neither a speciated result nor a detected total is available to carry forward. A non-detect on a total method is not proof of absence below the reporting value, and it is not a speciation result.

The exposure surface deserves its own sentence, without a number. The diaper area is occluded, often wet, and often already inflamed, and the infant barrier is immature in early infancy; those are reasons the dermal route could matter more here than for an adult cosmetic. They are not a measured absorbed fraction. No study on file reports the share of lead or cadmium that crosses infant skin from a diaper cream.

Exposure framing

A dermal exposure from a diaper cream has at least four inputs: how much metal is in the product, how much product is applied, how long it stays on the skin, and what fraction is absorbed. The four reports supply the first input for two metals, in four units. They supply none of the other three. No document opened for this work provides, together, an applied mass, a use frequency, a contact time and a dermal-absorption fraction for lead or cadmium from a diaper cream.

Because those inputs are missing, no illustrative dose is offered. Multiplying a concentration by an assumed application and an assumed absorption would produce a number that looked like a dose and was not one. The measured concentrations are concentrations. The honest statement for a caregiver is that lead and cadmium were present in four specific units at the printed levels, that the amount reaching a child from those or any other units has not been measured, and that stopping or continuing a product is a decision this page does not make.

Regulatory landscape

A diaper cream is not one legal object, and which instrument applies depends on how the product is marketed. The limits below are cited in their own unit and on their own basis. None of them is placed beside the measured ppb results as a score, because none of them is a finished-product total-metal limit on the same basis as those results.

In the United States a zinc oxide diaper-rash product is an over-the-counter skin-protectant drug, and the order actually in force is cited here. Over-the-Counter Monograph M016, Final Administrative Order OTC000005, was effective 27 March 2020 on enactment of the CARES Act and posted 24 September 2021 ¹⁹. Its section M016.10 permits zinc oxide at 1 to 25 percent, and zinc oxide above 25 to 40 percent only in an ointment dosage form. It states no heavy-metal contaminant limit. The 40 percent strength carried by three of the four measured units is expressly permitted by the operative order in an ointment, rather than resting on a 1990 proposal (55 FR 25204 ²¹) that was never finalised; and the absence of a finished-product metal limit is now a reading of the current order rather than an inference from the codified monograph (21 CFR 347 ²⁰) plus a proposal. The result that matters for this synthesis is unchanged and now better supported: the United States skin-protectant framework sets no finished-product heavy-metal limit for a diaper cream. Two lead figures that a consumer advocate cited against the products, a Washington State cosmetics figure and a general cosmetics figure, are characterised inconsistently within the advocate’s own pages and their applicability to a product that is a drug is not established, so they are not used here.

One further point belongs with the order rather than with the products. M016 sets permitted concentration ranges for several mineral actives beyond zinc oxide, including talc at 45 to 100 percent, kaolin at 4 to 20 percent, calamine at 1 to 25 percent and aluminium hydroxide gel at 0.15 to 5 percent, and sets no contaminant limit for any of them. The pattern, rather than any single figure, is what a regulator should take from this page: a framework that specifies how much of a mineral must be present and nothing at all about what else that mineral may carry.

The limits that do exist attach to something other than the finished drug product. The United States colour-additive specification for zinc oxide (21 CFR 73.1991 and 73.2991 ²²) limits the zinc oxide ingredient, when used as a colour, to cadmium 15, mercury 1, arsenic 3 and lead 20 ppm, with purity not less than 99 percent; it governs the ingredient, not the drug active, and it is an ingredient specification, not a measurement of any product. In Canada, licensed zinc oxide diaper-rash products are natural health products under the 2024 Diaper Rash Products Monograph ²³, whose finished-product specifications follow the Quality of Natural Health Products Guide ²⁴; the Guide’s topical limits are arsenic 3, cadmium 3, lead 10, total mercury 1 and antimony 5 ppm of the finished product. A product marketed as a cosmetic falls under the cosmetic guidance instead: Health Canada’s cosmetic-impurity guidance ²⁵ treats lead above 10, arsenic above 3, cadmium above 3, mercury above 1 and antimony above 5 ppm of the finished product as technically avoidable, and the German orientation values ²⁶ are lead 2.0, cadmium 0.1, mercury 0.1, arsenic 0.5 and antimony 0.5 mg/kg for cosmetics in general. The European zinc oxide active-substance monograph, as recorded in the pharmacopoeial knowledge database ²⁸, retains lead and cadmium tests at limits that were not obtained, adds a thallium test at 5 ppm, and deletes the arsenic test; those are ingredient-grade tests, and the updated lead and cadmium values are not on file.

A cutaneous drug-product guideline, the International Council for Harmonisation Q3D(R2) ²⁷, sets permitted daily exposures for elemental impurities in cutaneous drug products on a µg/day basis, with a dose-dependent concentration column. They are named here for completeness and are not applied to the measured results. Converting any of these limits onto the ppb finished-product basis of the four reports would require an assumed zinc oxide fraction or an assumed daily applied mass, and the reports do not supply either; any such conversion would be a derived figure, not a measurement, and none is made on this page.

Implications

For parents

Four specific units of zinc oxide diaper cream, ointment or paste were tested, one unit each, and lead and cadmium were found in all four, with lead the higher of the two, from about 1.2 to 4.7 ppm, and cadmium from about 0.045 to 0.51 ppm. Arsenic and mercury were below the tests’ reporting values in every unit. These are four units chosen by an advocate, not a survey of the shelf, and each was tested once with no published quality control, so they cannot say how common these levels are or how a given product on the shelf compares.

The statement not to take from this page is a risk verdict. No study on file follows lead or cadmium from a diaper cream across a child’s skin and into the body, so the amount, if any, that reaches a child is unmeasured. Whether to use, change or stop a product is a caregiving decision this page does not make.

One separate matter concerns zinc rather than the contaminants, and it is about broken skin rather than about any product being contaminated. Two published single cases describe infants whose blood zinc rose well above the normal range while a zinc oxide ointment was being used on eroded or inflamed diaper skin, and fell after it stopped. In one of the two the infant was receiving no zinc by mouth. Two cases across the whole literature is not a rate and says nothing about how likely this is for any individual child; on intact adult skin a far larger dose of a stronger ointment moved blood zinc by an amount the authors called not significant. The reason it appears here at all is that it points at broken skin as the condition that matters, which is also the condition under which a caregiver is most likely to be applying the product. Any question about a specific child’s skin, or about how much ointment to use on it, belongs with that child’s clinician and not with this page.

For manufacturers

The finished product and the zinc oxide raw material need separate numbers. The ingredient specifications on file constrain the zinc oxide raw material; they are not a measurement of the finished cream, and a compliant ingredient does not by itself establish a finished-product concentration. A brand that wants to speak to metals in its finished diaper cream has to measure the finished product, on the finished-product basis, across lots rather than in a single unit.

The zinc oxide percentage is not a proxy for the metal load. In the four units on file the lowest lead result was in the lowest-strength product and one of the highest was a 40 percent product, so lead did not track zinc oxide content in this small set. Whether lead and cadmium enter through the zinc oxide ingredient, through another ingredient, or through both is not settled by four units.

For regulators

The gap is the finding, and it is now a reading of the operative order rather than of a lapsed proposal. Monograph M016 sets no finished-product heavy-metal limit for a diaper cream, and the limits that exist elsewhere are on different objects and different bases: an ingredient specification in ppm of the zinc oxide raw material, a cutaneous drug dose in µg/day, and cosmetic guidance in ppm of a finished cosmetic. A number written for one of those objects should not be read as a finished-product drug limit, and the measured ppb results should not be scored against it. Where a government has no finished-product number for metals in this product, the gap should stay visible rather than be filled by borrowing a limit from an adjacent basis.

The absence is a drafting pattern and not an oversight peculiar to zinc oxide. M016 specifies permitted concentration ranges for mineral actives including talc, kaolin, calamine and aluminium hydroxide gel, and carries no contaminant specification for any of them. A manufacturer safety framework for the adjacent product, disposable diapers themselves, shows the same shape: a published safety evaluation covering exposure scenarios and a threshold of toxicological concern reports no concentration of lead, cadmium, arsenic, mercury, nickel, chromium, tin or antimony ¹⁸. Neither document is evidence that metals are absent. Both are evidence that the frameworks in place were not built to look.

Speciation rules that exist for food should not be relaxed because the matrix is a cream. Total arsenic is not inorganic arsenic and total mercury is not methylmercury; the four reports measured totals, both below the reporting value, and that is all they can support.

For researchers

The next useful study is specific. Measure lead, cadmium, arsenic and mercury in the finished diaper cream on the finished-product basis, across multiple lots and multiple products rather than one unit each, with sample mass, digestion, calibration, recovery, detection limits and accreditation reported. Include the non-zinc-oxide products, because the clean comparison group has never been measured and the whole clean-versus-contaminated question rests on it. Speciate arsenic and mercury where a total is detected. Then, and separately, measure the dermal step: how much lead or cadmium leaves the vehicle and crosses intact and compromised infant-relevant skin, so that a concentration can become an exposure. Until that dermal step is measured, a concentration in a diaper cream is a concentration in a diaper cream.

Data gaps and research agenda

No peer-reviewed survey of metals in finished diaper creams is on file. No non-zinc-oxide diaper cream has been measured for metals, so the clean baseline of the product pair has no measured evidence, and the one ingredient inventory that describes those products counts allergens rather than metals. No study on file measures lead or cadmium absorption from a diaper cream, and no dermal-absorption fraction for any metal from a diaper cream was opened. Every absorption measurement in this file is of zinc, the active, and the two that use infant skin are single cases. Inorganic arsenic and methylmercury in a diaper cream: no data found, and the total-arsenic and total-mercury results that exist are non-detects. A market detection rate cannot be calculated from four purposively selected units or from an eleven-item advocacy chart whose unverified values are not used here.

Two further gaps follow because the sources that closed others opened these. No study measures zinc, lead or cadmium release from a diaper cream at diaper-area pH and hydration, although both differ measurably from the forearm conditions every intact-skin study used. And no case series exists for hyperzincemia during topical zinc oxide use in infants; there are two independent single cases, which is enough to establish that the event occurs and not enough to say how often, in which infants, or at what applied dose.

The research agenda follows those gaps, in order. First, finished-product content in mg/kg for lead, cadmium, arsenic and mercury, across lots and products, with full method reporting, including non-zinc-oxide products. Second, speciation of arsenic and mercury wherever a total is detected. Third, a dermal-transfer measurement for lead and cadmium on intact and compromised skin. Fourth, a sampling frame that is not built from advocate-selected units.

Peer review state

This synthesis claim has not yet been evaluated by external reviewers. Verdicts will be added here as named domain experts, listed at Curators and conflict-of-interest disclosure, complete their review. Reviewer verdicts are recorded on this page.

The Heavy Metal Index publishes synthesis claims as preprints — before external review completes — with the review state visibly tracked. Until at least one external verdict is recorded below, a synthesis page is a preprint, not a peer-reviewed work. External review accumulates over time, and the credibility of the claim is partly the cumulative result of that visible review.

How to cite

Pendergrass, K. (2026). Heavy metals in diaper creams. Institute of Contaminant Standards. https://doi.org/10.83376/hmi.heavy-metals-in-diaper-creams

References

1. Consumer-commissioned laboratory test of a 40% zinc oxide diaper rash ointment for lead, cadmium, mercury and arsenic (lot 2025-142)Rubin, T. · Lead Safe Mama, LLC · 2025Consumer advocacy testing
2. Consumer-commissioned laboratory test of a 40% zinc oxide diaper rash paste for lead, cadmium, mercury and arsenic (lot 2025-169)Rubin, T. · Lead Safe Mama, LLC · 2025Consumer advocacy testing
3. Consumer-commissioned laboratory test of a 40% zinc oxide maximum-strength diaper rash paste for lead, cadmium, mercury and arsenic (lot 2025-182)Rubin, T. · Lead Safe Mama, LLC · 2025Consumer advocacy testing
4. Consumer-commissioned laboratory test of a 13% zinc oxide diaper rash cream for lead, cadmium, mercury and arsenic (lot 2025-218)Rubin, T. · Lead Safe Mama, LLC · 2025Consumer advocacy testing
5. Consumer-advocacy comparison chart of lead, cadmium, mercury and arsenic in eleven diaper creams (updated January 2026)Rubin, T. · Lead Safe Mama, LLC · 2026Consumer advocacy testing
6. A study of the percutaneous absorption from topically applied zinc oxide ointmentDerry, J. E., McLean, W. M., & Freeman, J. B. · Journal of Parenteral and Enteral Nutrition · 1983Peer-reviewed
7. Percutaneous absorption of zinc from zinc oxide applied topically to intact skin in manÅgren, M. S. · Dermatologica, 180(1), 36-39 · 1990 · doi.org/10.1159/000247982Peer-reviewed
8. Influence of two vehicles for zinc oxide on zinc absorption through intact skin and woundsÅgren, M. S. · Acta Dermato-Venereologica, 71(2), 153-156 · 1991 · doi.org/10.2340/0001555571153156Peer-reviewed
9. Human epidermal zinc concentrations after topical application of ZnO nanoparticles in sunscreensKhabir, Z., Holmes, A. M., Lai, Y.-J., Liang, L., Deva, A., Polikarpov, M. A., Roberts, M. S., & Zvyagin, A. V. · International Journal of Molecular Sciences, 22(22), 12372 · 2021 · doi.org/10.3390/ijms222212372Peer-reviewed
10. UVB exposure enhanced the dermal penetration of zinc oxide nanoparticles and induced inflammatory responses through oxidative stress mediated by MAPKs and NF-κB signaling in SKH-1 hairless mouse skinPal, A., Alam, S., Chauhan, L. K. S., Saxena, P. N., Kumar, M., Ansari, G. N., Singh, D., & Ansari, K. M. · Toxicology Research, 5(4), 1066-1077 · 2016 · doi.org/10.1039/c6tx00026fPeer-reviewed
11. Early intervention by a ward pharmacist for the hyperzincemia caused by zinc oxide ointment in a infant with diaper dermatitis: A case reportYasugi, M., Ooi, H., Asai, Y., Konishi, T., Otobe, Y., Kitamura, S., Uchizono, H., Takahashi, M., & Takeuchi, M. · Yakugaku Zasshi, 146(1), 85-89 · 2026 · doi.org/10.1248/yakushi.25-00096Peer-reviewed
12. Hyperzincemia due to topical zinc oxide in an infant with diaper dermatitisFujita, T., Kondo, M., Baba, T., Ohishi, A., & Iijima, S. · Pediatrics International · 2021 · doi.org/10.1111/ped.14761Peer-reviewed
13. Opinion on zinc oxide (nano form), COLIPA S 76Scientific Committee on Consumer Safety · SCCS/1489/12, European Commission · 2012Government report
14. Skin barrier function in infants: Update and outlookRahma, A., & Lane, M. E. · Pharmaceutics, 14(2), 433 · 2022 · doi.org/10.3390/pharmaceutics14020433Peer-reviewed
15. Association of skin wetness and pH with diaper dermatitisBerg, R. W., Milligan, M. C., & Sarbaugh, F. C. · Pediatric Dermatology · 1994Peer-reviewed
16. Diaper dermatitis: Etiology, manifestations, prevention, and managementStamatas, G. N., & Tierney, N. K. · Pediatric Dermatology · 2014Peer-reviewed
17. Potential allergens in disposable diaper wipes, topical diaper preparations, and disposable diapersYu, J., Treat, J., Chaney, K., & Brod, B. · Dermatitis · 2016Peer-reviewed
18. Safety evaluation of superabsorbent baby diapersKosemund, K., Schlatter, H., Ochsenhirt, J. L., Krause, E. L., Marsman, D. S., & Erasala, G. N. · Regulatory Toxicology and Pharmacology · 2009Peer-reviewed
19. Over-the-Counter Monograph M016: Skin protectant drug products for over-the-counter human use (Final Administrative Order OTC000005)U.S. Food and Drug Administration · effective 27 March 2020, posted 24 September 2021 · 2021Regulatory text
20. 21 CFR Part 347, Skin protectant drug products for over-the-counter human use (4-1-25 edition)U.S. Food and Drug Administration · 2025Regulatory text
21. Skin protectant drug products for over-the-counter human use; Tentative final monograph for diaper rash drug products; Proposed ruleU.S. Food and Drug Administration · 55 FR 25204 · 1990Regulatory text
22. 21 CFR Part 73, color additives exempt from certification (4-1-2023 edition)U.S. Food and Drug Administration · 2023Regulatory text
23. Diaper Rash Products MonographHealth Canada, Natural and Non-prescription Health Products Directorate · 2024Regulatory text
24. Quality of Natural Health Products GuideHealth Canada, Natural and Non-prescription Health Products Directorate · 2015Regulatory text
25. Guidance on heavy metal impurities in cosmeticsHealth Canada · 2012Regulatory guidance
26. Technically avoidable heavy metal contents in cosmetic productsGerman Federal Office of Consumer Protection and Food Safety (BVL) · Journal of Consumer Protection and Food Safety, 12(1), 51-53 · 2017 · doi.org/10.1007/s00003-016-1044-2Regulatory text
27. ICH harmonised guideline Q3D(R2): Guideline for elemental impuritiesInternational Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use · 2022International guideline
28. Knowledge database: Zinc oxide, European Pharmacopoeia monograph 00252, monograph details and historyEuropean Directorate for the Quality of Medicines & HealthCare · 2026Government dataset

Update history

No substantive edit history is available in this build. The full commit record is available in git.