Using balms while pregnant or breastfeeding, judged on the real evidence
Most anhydrous balms are low risk because absorption from an occlusive base is small. Retinoids are the clear avoid, and salicylates deserve care.
Pregnancy turns an ordinary bathroom shelf into a list of questions that nobody labelled the answers to. This page separates the balm ingredients with a genuine reason behind the caution from the ones avoided out of habit, gives the numbers where numbers exist, and says where the evidence runs out. It also flags the two situations where a symptom that looks like dry skin needs a midwife rather than a jar.
Most plain wax and oil balms are low risk in pregnancy and breastfeeding, because a fat base with no water, no surfactant and no penetration enhancer delivers very little through skin. Retinoids are the one clear avoid. High strength methyl salicylate rubs deserve real care, particularly after 20 weeks. Almost everything else is a judgement about fragrance, nausea and comfort.
Why a wax and oil base is a poor delivery vehicle
The answer to most of these questions is reassuring for a reason that has nothing to do with balms being natural: they are physically bad at getting anything through skin, and three properties work against absorption at once.
Molecular size. The stratum corneum passes small molecules and obstructs large ones. Bos and Meinardi set out the widely quoted 500 dalton observation, that compounds crossing intact skin in meaningful amounts are almost all below about 500 g per mole. The bulk of a balm sits far above it: a triglyceride from shea, olive or sunflower runs around 850 to 890 g per mole, beeswax monoesters cluster at roughly 590 to 700, and petrolatum is a hydrocarbon mixture largely above C20. None of that is absorbed in a quantity worth calculating, as lipid penetration into skin sets out. It stays on the surface and in the upper layers, which is what an occlusive is for.
Partitioning. A small active still has to leave the vehicle before it can enter skin, and a lipophilic molecule dissolved in 4 g of fat has little thermodynamic drive to do so. The same molecule in a light lotion sits at much higher effective activity and moves more readily, which is why the percentage on the pack is a poor proxy for the dose delivered.
No enhancers. Ethanol, propylene glycol, surfactants and several ester solvents all raise skin permeability, and an anhydrous balm contains none of them by definition. What it does have is occlusion, which raises stratum corneum hydration and can increase penetration of whatever is dissolved in the base. That is the one mechanism running the other way, and it is why the sensible questions here are about area, dose and frequency rather than the percentage on the label.
| Format | Area covered | Product per application | What decides the exposure |
|---|---|---|---|
| Lip balm | about 10 cm2 | 10-50 mg | Ingestion, not skin. Take 50 mg a day as a working figure for a heavy user |
| Hand salve | about 400 cm2 | 0.4-0.8 g | Frequency. Six to ten applications a day is normal for wet work |
| Bump or body balm | 1,200-1,800 cm2 | 1.5-3 g | Area, and months of twice daily use |
| Muscle or chest rub | 300-800 cm2 | 2-5 g | Active concentration. Monograph strengths are drug doses, not cosmetic ones |
| Nipple balm | under 20 cm2 | 0.1-0.3 g | Infant ingestion of traces, not maternal absorption |
The list of things worth avoiding is short
Marketing has produced long pregnancy avoid lists, most of them assembled from ingredient reputation rather than exposure data. The list that survives contact with the evidence is short, and it groups by why the concern exists rather than by how natural the material sounds.
| Ingredient or class | Why it is raised | Evidence in pregnancy | Practical position |
|---|---|---|---|
| Retinol, retinyl palmitate, retinal, tretinoin | Oral retinoid teratogenicity | Strong for oral, weak and reassuring for topical | Avoid. The precaution costs you almost nothing |
| Methyl salicylate, high strength rubs | Systemic salicylate absorption from large areas | Case reports of toxicity from heavy topical use | Avoid rubs at monograph strength, especially from 20 weeks |
| Salicylic acid | Same class, keratolytic use | Capped in cosmetics at low levels anyway | Fine at cosmetic levels on small areas. Skip whole body use |
| Camphor at rub strengths | Toxicity if swallowed, and a stimulant reputation | Very thin for topical cosmetic levels | Cosmetic traces are not the issue. Keep chest rub strengths off large areas |
| Essential oils generally | Absorption of small aroma molecules | Thin. Most safety data is not pregnancy specific | Cut them, mostly for nausea and sensitisation reasons |
| Beeswax, shea, cocoa, plant oils | Nothing, beyond allergy | Long cosmetic record, no pregnancy restriction | Use freely, subject to your own allergies |
| Petrolatum and mineral oil, USP grade | Origin, not toxicology | Not absorbed. Refined grades are the patch test vehicle | Fine, and the blandest option available |
| Lanolin, highly purified | Used on nipples, so ingestion by an infant | Long use record in this application | Fine, subject to lanolin allergy |
"Pregnancy safe" is not a regulated term anywhere. No authority certifies it, no test supports it, and a brand using the phrase is making a marketing claim rather than a compliance statement. The same is true of clean, non toxic and chemical free. Judge the ingredient list instead, which you can check yourself using reading a balm label.
Retinoids: the clearest advice, built on the weakest topical data
The avoid recommendation for topical retinoids is close to universal in obstetric and dermatological guidance, and what it rests on is not what most people assume. Oral isotretinoin is a proven human teratogen with a characteristic pattern of craniofacial, cardiac and central nervous system malformations, which is why prescribing it involves pregnancy prevention programmes with mandatory testing. High dose oral vitamin A carries the same class of concern. That evidence is not in dispute and it is the reason the whole category is treated cautiously.
The topical evidence is a different matter. Absorption from a cosmetic applied to intact skin is low, prospective cohort studies of first trimester topical tretinoin exposure have not shown an increased rate of major malformations, and no causal link between a cosmetic retinoid and a birth defect has been established. A few case reports describe embryopathy after topical use, and case reports cannot establish causation. The honest position is that the data does not show harm and is not large enough to demonstrate its absence.
What tips the decision is the cost of the precaution. Skipping a cosmetic retinoid for nine months costs you a slower rate of change in fine lines. Nothing else. The European Union has separately capped retinol, retinyl acetate and retinyl palmitate at 0.3 percent retinol equivalent in most products and 0.05 percent in body lotion under Annex III entry 376, with the label line "Contains Vitamin A. Consider your daily intake before use", and that cap exists to control total preformed vitamin A intake rather than to address pregnancy specifically. Formulation detail is in retinol and bakuchiol, including why a hot wax balm is a poor carrier for the material in the first place.
Bakuchiol is regularly promoted as the pregnancy substitute. It is not a retinoid chemically, but there is no pregnancy safety data for it either, so trading an ingredient with reassuring topical data for one with no data is a marketing move rather than a safety one.
Salicylates: the one where dose and area really matter
Salicylates split into two very different exposures that get discussed as though they were one. Cosmetic salicylic acid is tightly limited already. Annex III entry 98 of the EU Cosmetic Products Regulation caps it at 0.5 percent in lip products and 2.0 percent in other products, with a restriction on use in children under three. At those levels, on a small area, systemic exposure is trivial. The caution that circulates is inherited from oral aspirin, and low dose aspirin at 75 to 150 mg daily is actively prescribed in pregnancy to reduce pre-eclampsia risk. The molecule is not the problem. The dose is.
Methyl salicylate is where the concern becomes real. It is a counterirritant active at 10 to 60 percent in the US external analgesic monograph, so a muscle rub can carry more salicylate in a single application than several aspirin tablets. Wintergreen and sweet birch oils are 85 to 99 percent methyl salicylate. Percutaneous absorption from these products is well documented, and there are published cases of systemic salicylate toxicity from heavy or repeated topical use, particularly with heat, occlusion or application to large areas. The EU takes a different route and caps methyl salicylate in cosmetics at 0.06 percent leave-on and 0.03 percent on lips, which effectively removes it from cosmetic balms altogether.
The third trimester point is about the drug class, not the balm. Systemic non steroidal anti-inflammatory exposure late in pregnancy is associated with premature narrowing of the ductus arteriosus and with reduced fetal urine output, which is why regulators advise avoiding NSAIDs from around 20 weeks unless a clinician has weighed it up. A high strength salicylate rub applied twice a day to a whole back is the only balm format capable of producing that kind of systemic dose. The mechanism, the monograph levels and the poisoning figures are set out in menthol, camphor and other actives, and the formulation consequences in muscle rub balm.
A cosmetic balm containing a trace of wintergreen for scent and a pharmacy muscle rub at monograph strength are different products by two orders of magnitude, and the same word covers both. If a rub says it relieves muscular pain, it is a medicine, the dose is a drug dose, and the question belongs with a pharmacist or midwife who can see the pack. Where the cosmetic line sits is explained in cosmetic versus drug claims.
Essential oils: what is known, and how little of it is pregnancy specific
Essential oil constituents are small, often below 200 g per mole, and they do cross skin. That is the whole mechanistic basis of the concern. What does not exist is human data on cosmetic level topical exposure in pregnancy. Most of what gets quoted comes from oral dosing in animals, from poisoning cases involving ingestion, or from traditional use claims.
The IFRA Standards are the most rigorous limits in this space, and they are not pregnancy specific. They are built on sensitisation, phototoxicity and general population systemic exposure, so a material sitting inside its IFRA limit has satisfied a safety assessment that never asked the pregnancy question. Lip products fall into IFRA Category 1, the strictest category, where individual constituent limits bite hard: citral at 0.11 percent, isoeugenol at 0.019 percent, coumarin at 0.089 percent. The full working, constituent by constituent, is in essential oils in balms.
The oils that appear on cautioned lists in the reference texts are cautioned mainly on constituent grounds: wintergreen and birch for methyl salicylate, sage, wormwood and thuja for thujone, pennyroyal for pulegone, aniseed and fennel for anethole, parsley seed for apiole, and camphor rich rosemary chemotypes. Those lists come from toxicological reasoning about constituents rather than from pregnancy outcome studies, which is a reasonable basis for caution and a poor basis for confident statements in either direction.
There is a more practical argument for leaving them out. Olfactory aversion in early pregnancy is intense and unpredictable, and a scented film over your own torso is not something you can walk away from. Heavy, repeated application over large areas is also the exposure pattern most likely to induce a fragrance allergy you then keep for life. The declared allergens are in fragrance allergens and a base built without them is in fragrance free balm. The functional benefit of fragrance in a bump balm is zero, so any uncertainty is a bad trade.
Nipple balms and whether they need wiping off
The constraints on a nipple product are not maternal absorption. They are that traces will be ingested by an infant, and that a cracked nipple should not have to be wiped clean eight to twelve times a day. Between them those two eliminate essential oils, herbal extracts, flavours, strong tastes and anything with pharmacological activity.
Highly purified anhydrous lanolin is the reference material because it satisfies both. It is a very effective occlusive, it is poorly absorbed, and it has a long record in this specific use, which is why it does not need removing before a feed. Grade matters more than anything else here: ordinary craft lanolin and highly purified medical grade lanolin are not the same product, and purified grades reduce free lanolin alcohols below 1.5 percent, the fraction the historical allergy literature points at. In the United States lanolin is a recognised skin protectant active at 12.5 to 50 percent. Material notes are in lanolin and the minimal formula, plus a lanolin free alternative built on white soft paraffin, is in nipple balm.
Be honest about what the product does. A Cochrane review of interventions for painful nipples in breastfeeding women found the trials small and inconsistent, with no clear evidence that lanolin or any other applied treatment outperformed expressed breast milk or doing nothing at all. A balm keeps damaged skin from drying and reopening between feeds, which is worth having. It does not fix the cause, which is usually mechanical and usually resolves when someone experienced watches a whole feed.
Stretch mark products: the evidence is thinner than the shelf suggests
The Cochrane review of topical preparations for preventing stretch marks in pregnancy found few trials, most small and at meaningful risk of bias, and no clear evidence that any tested preparation reduced the development of striae against placebo or no treatment. That covers products with research budgets behind them, so a homemade balm claims an effect on an evidence base that never demonstrated one for anything.
There is also a confounder in almost every protocol: participants are asked to massage the product in daily, and where the control arm gets the same massage with a plain base the gap between arms tends to shrink. The predictors that do hold up across observational studies are family history, younger maternal age, higher pre-pregnancy body mass index, greater gestational weight gain and higher birthweight, and a balm changes none of them.
What a balm reliably does is reduce water loss while it is on the skin, so tight, itchy, stretched skin feels better. That is a real and sufficient reason to use one, and it is the basis on which the formula in pregnancy belly balm is built. How much to use over that area is worked out in how much balm to apply. For inner thigh chafe, a separate late pregnancy problem, a drier and firmer stick works better than a rich jar.
Melasma, the lower lip and the sun
Melasma reported during pregnancy ranges from roughly 15 to 70 percent of women across published series, the spread driven by skin type, latitude and how each study defined it. It is provoked by ultraviolet radiation and by visible light, and the visible light part is why a transparent chemical filter alone often disappoints on the face.
A balm is not a substitute for a tested sun product on any part of this. Sun protection factor is a measured value from a defined protocol, not a property you can infer from an ingredient list, and a homemade balm with mineral powder stirred into it has an unknown protection factor that its appearance cannot reveal. What a labelled figure does and does not buy on the lip is in lip balm with SPF. Shade, a brimmed hat and timing beat every product, and on lips the hat is unusually effective because midday exposure is close to vertical.
Where a balm can help is opacity. Iron oxides attenuate visible light in a way transparent filters do not, which is why tinted products perform better in melasma studies than untinted ones, and the pigment side is in iron oxides and mineral pigments. A tinted balm is a reasonable adjunct, not a treatment.
Itch that is not dry skin
Intense itching, particularly of the palms and soles, especially at night and without a rash, needs same day assessment. It can be intrahepatic cholestasis of pregnancy, which is diagnosed by blood tests for bile acids and liver function and monitored by an obstetric team, not moisturised. A new or spreading rash, blistering, itching with feeling unwell, or itching that keeps you awake also belongs with a midwife or doctor. Moisturising an itch of this kind is not merely ineffective, it is a delay.
Several pregnancy specific skin conditions present as itch and get mistaken for dryness. Polymorphic eruption of pregnancy typically appears in the third trimester as itchy papules starting in abdominal striae, atopic eruption of pregnancy comes earlier and in people with an atopic history, and pemphigoid gestationis is rare, blistering and needs specialist care. An emollient may make the skin more comfortable while the diagnosis is made, and that is the whole of its role.
Ordinary pregnancy dryness does respond to a bland occlusive, and the marker is duration and pattern. Itch that follows dry, tight skin over the bump, improves within an hour of applying something greasy and returns gradually is behaving like dryness. Itch that is worse at night, sited on palms and soles, or unrelated to how the skin looks is not. The general limits of what any page here can tell you are in the safety disclaimer.
The decision rule, and what this page cannot settle
For a buyer, the rule is short. Read the ingredient list rather than the front of the pack. Reject anything carrying retinol, retinyl palmitate or another retinoid, and anything at rub strength with methyl salicylate or camphor. Then choose on ordinary grounds: fewest ingredients, no fragrance while your sense of smell is unreliable, and a texture you will actually apply twice a day for four months. Alarming looking names are usually just INCI conventions, and a short home trial before covering a large area is sensible, using the method in how to patch test a balm.
What no general page can settle is your case. Pregnancy interacts with medication you already take, with skin conditions that predate it, and with prescribed topicals whose behaviour changes under an occlusive layer. A midwife, obstetrician, pharmacist or dermatologist can weigh those; the person who made the balm cannot. In the United Kingdom the UK Teratology Information Service publishes patient facing summaries of named exposures through its bumps service, which is a better place to take an ingredient question than any product website.
Frequently asked questions
Is lip balm safe in pregnancy?
A plain wax and oil lip balm carries no pregnancy specific concern. The relevant exposure is what you swallow rather than what crosses your skin, on the order of 10 to 50 milligrams a day for a heavy user. Skip retinol lip treatments, and skip anything medicated with camphor, menthol at high levels or wintergreen. Fragrance free is worth choosing while nausea is unpredictable.
Can I use a retinol product while pregnant?
The standard advice is to avoid all topical retinoids in pregnancy, and it is worth following. Oral isotretinoin and high dose vitamin A are proven teratogens. Absorption from a cosmetic retinoid is low and studies of first trimester topical tretinoin exposure have not shown increased malformation rates, but the data is not large enough to prove safety, and the cost of pausing for nine months is only a slower cosmetic result.
Are essential oils safe in a pregnancy balm?
Nobody can answer that from evidence, because human data on cosmetic level topical exposure in pregnancy barely exists. IFRA limits are built on sensitisation and general population exposure, not pregnancy. The practical argument against them is stronger anyway: fragrance offers no functional benefit in a body balm, nausea makes any scent unpredictable, and months of large area use is how fragrance allergies are acquired.
Does nipple balm need wiping off before a feed?
Highly purified anhydrous lanolin and pharmaceutical white soft paraffin do not need removing, which is a large part of why they are used. Anything with a strong taste, a fragrance, an essential oil, a herbal extract or pharmacological activity fails that test and should not be on a nipple at all. Wiping a cracked nipple ten times a day is worse than the original problem.
Will a belly balm prevent stretch marks?
There is no good evidence that any topical does. The Cochrane review of preparations for preventing stretch marks in pregnancy found few, small, biased trials and no clear benefit over placebo or no treatment. The strongest predictors are family history, maternal age, body mass index, weight gain and birthweight, none of which a balm changes. It does make tight, dry skin feel comfortable, which is a fair reason to use it.
Can I use a muscle rub for back pain in pregnancy?
Not without asking someone who can see the pack. Rubs at monograph strength carry 10 to 60 percent methyl salicylate, which is absorbed through skin and is pharmacologically a salicylate dose rather than a cosmetic one. Systemic non steroidal exposure is advised against from around 20 weeks. Heat, occlusion and large areas all increase the absorbed dose. Ask a pharmacist or midwife.
My skin is itching badly. Is a richer balm the answer?
Not until someone has ruled out the causes that are not dry skin. Intense itching of the palms and soles, worse at night and without a rash, can be intrahepatic cholestasis of pregnancy, which is diagnosed by blood tests and needs same day assessment. A new or spreading rash, blistering or itching alongside feeling unwell also needs a midwife or doctor rather than a jar.
Sources and further reading
- Scientific Committee on Consumer Safety, Opinion on Vitamin A (retinol, retinyl acetate, retinyl palmitate), SCCS/1639/21, European Commission.
- European Commission, Regulation (EC) No 1223/2009 on cosmetic products, EUR-Lex. Annex III entries covering salicylic acid, methyl salicylate and vitamin A.
- Brennan M, Young G and Devane D, Topical preparations for preventing stretch marks in pregnancy, Cochrane Database of Systematic Reviews, 2012, CD000066.
- Dennis C-L and colleagues, Interventions for treating painful nipples among breastfeeding women, Cochrane Database of Systematic Reviews, 2014.
- Royal College of Obstetricians and Gynaecologists, Intrahepatic Cholestasis of Pregnancy, Green-top Guideline No. 43, London, 2022.
- Bos JD and Meinardi MMHM, The 500 Dalton rule for the skin penetration of chemical compounds and drugs, Experimental Dermatology, 2000.
- UK Teratology Information Service, bumps: best use of medicines in pregnancy, patient information monographs, Newcastle upon Tyne.
Reviewed and updated 6 September 2026. Spotted an error? Tell us and we will fix and log it.