Pregnancy-Safe Skincare: Stop, Keep, Still Debated

What This Page Can and Cannot Tell You

Starting here, because everything below depends on it.

This page reports what regulators and specialist bodies have published about skincare ingredients in pregnancy, and names the source for each one. It does not tell you what is safe for you to use. That is a decision involving your history, your trimester, what else you take, and a clinician who knows all three.

There are three reasons to be careful with any page like this, including this one.

The evidence is mostly absence rather than presence. Pregnant women are not enrolled in trials of cosmetic ingredients. What exists is systemic absorption data, registry and cohort observation, and inference. "No increased risk was observed" is a different statement from "this is safe", and the gap between them is where most of the confident advice online lives.

The regulatory shorthand everyone uses no longer exists — the next section is about that.

And the risk is asymmetric. The benefit of a cosmetic active over nine months is small. The cost of being wrong is not. That asymmetry, rather than any specific piece of evidence, is what drives most of the caution below — and saying so is more honest than dressing up precaution as data.

The Letter Categories You Are Reading Were Withdrawn in 2015

Almost every pregnancy-skincare page sorts ingredients by a letter: category A, B, C, D or X. That system is gone.

The FDA states it plainly in its labelling resources: the Pregnancy and Lactation Labeling Rule "removes pregnancy letter categories – A, B, C, D and X". The labelling changes took effect on 30 June 2015, with new submissions using the new format immediately and existing prescription labels phased in gradually. Three narrative subsections replaced the letters: Pregnancy, Lactation, and Females and Males of Reproductive Potential.

The letters were removed because a single character cannot carry a risk assessment. They were read as a grading scale — B is safer than C — which is not what they meant.

And there is a second problem specific to skincare. The FDA is explicit: "Labeling for over-the-counter (OTC) medicines will not change; OTC drug products are not affected by the final rule." The rule governed prescription drugs. Most of what is in a skincare routine is a cosmetic or an OTC drug, which means a letter category for a moisturiser was never a real classification in the first place — it was borrowed language.

This matters enough that it reached our own library: an earlier article on this site reported azelaic acid as "pregnancy category B", repeating a published review. That phrasing has been corrected, and the azelaic acid article should be read with this section in mind.

If a page sorts ingredients by letter, it is working from a system the regulator retired a decade ago.

Retinoids: Where Oral and Topical Are Genuinely Different Questions

This is the clearest entry and also the most misreported, because two different drugs get discussed as one.

Oral isotretinoin is a known teratogen. A specialist infant-risk centre describes oral retinoids as "absolutely contraindicated in pregnant women", causing craniofacial defects, heart defects and nervous system defects. There is no ambiguity here and no debate to report.

Topical is a different quantity. The same source puts topical benzoyl peroxide absorption at 3% and topical clindamycin at 4 to 5%, and on topical retinoids says controversy exists and they "should be avoided during pregnancy" pending larger studies.

The UK Teratology Information Service is more granular, and its wording is worth reading twice because it holds both halves at once. On adapalene, UKTIS states: "An increased risk of malformation is unlikely where topical adapalene is used at therapeutic doses as per the manufacturer's instructions during pregnancy. However, due to the known teratogenicity of retinoids and inadequate pregnancy data for adapalene, its use during pregnancy cannot be recommended."

Unlikely to cause harm. Cannot be recommended. Both, simultaneously — and that is not a contradiction, it is what an honest position looks like when absorption is low, data are thin, and the downside is severe.

UKTIS adds the part that matters most to someone reading this after the fact: women inadvertently exposed "should be reassured that systemic absorption of the drug is very low and that existing studies show no evidence of a teratogenic risk for topical retinoids as a group." It also records one case report of anophthalmia following topical exposure — a single report, noted rather than weighted.

A clinician's guide to topical retinoids reaches the same conclusion by a different route. Although the risk may be minimal with agents such as tretinoin, "the risk still outweighs the benefits, and all retinoids should be avoided during pregnancy."

Note what is being weighed. Not evidence of harm — the benefit side. A cosmetic retinoid for nine months has very little to offer against even a small unquantified risk. That is the actual argument, and it is a good one.

A detail that proves the previous section: that clinician's guide, published recently, still sorts retinoids as "category C" and tazarotene as "category X" — categories the FDA withdrew in 2015. Even the clinical literature has not finished letting go.

Generally Considered Acceptable

"Generally considered acceptable" is doing deliberate work here. It is what the sources say. It is not a clearance.

Two observations about that table.

The reasoning is absorption-led, not outcome-led. These ingredients are described as acceptable because very little of them gets in — not because anyone has run a trial in pregnant women and watched what happened. That is a reasonable basis and a limited one.

And azelaic acid's entry contains its own caveat: acceptable despite a lack of pregnancy studies. It is the ingredient most often recommended for pregnancy acne and melasma, which makes the thinness of the data worth knowing rather than glossing.

Genuinely Debated

The column other pages do not print.

The absorption figure is the whole argument: 30 to 40% after topical use — an order of magnitude above benzoyl peroxide's 3%. The source recommends avoiding it in pregnancy, and that recommendation follows directly from the number rather than from any observed outcome. Worth noting alongside what the specialist position on hydroquinone actually is outside pregnancy, which is less alarmed than its reputation.

Genuinely unsettled, and the sources sit on different sides. Low-concentration topical use is described as commonly used in pregnancy — and the same source notes salicylic acid is used systemically in pregnancy at low doses for a different purpose entirely. What is not established is where topical use stops being negligible: concentration, surface area and frequency all matter, and none of them have a published threshold. A 2% leave-on and a salicylic peel are not the same exposure.

Widely discussed, and none of the specialist sources consulted here resolves it. The infant-risk overview says sunscreens do not contain known harmful ingredients for pregnant women and does not address individual filters. Mineral filters are frequently preferred on the reasoning that they sit on the surface, which is mechanistically sensible and not the same as being demonstrated. If a page tells you confidently which filters to avoid in pregnancy, ask what it is citing.

Covered above, and they belong here too — "unlikely to cause harm" and "cannot be recommended" is the definition of unsettled.

Melasma: The Problem Pregnancy Actually Causes

Most people arrive at this subject because of melasma, and it has an unkind structure.

Melasma is a bilateral, blotchy, brownish facial pigmentation, most common in people who tan easily or have naturally brown skin — Fitzpatrick phototypes III and IV — and predominantly in women aged 20 to 40. Pregnancy and hormonal contraception are implicated in a quarter of affected women.

Here is the problem. DermNet reports the most effective topical treatment as a combination of hydroquinone, tretinoin and a moderate-potency topical steroid, achieving 60 to 80% improvement.

Read that against the two sections above. The single most effective treatment for the pigmentation pregnancy causes is built on the two ingredients most consistently flagged in pregnancy. That is not a gap in the advice; it is the actual situation, and pages that imply a straightforward alternative are not levelling with you.

What is left is photoprotection, and the detail that makes it work is specific: ultraviolet and visible light both promote melanin production, and DermNet's recommendation is year-round broad-spectrum SPF50+ containing iron oxides. Iron oxides are the part that addresses visible light, which ordinary sunscreen does not. How much sunscreen you actually apply matters more than the number on the bottle, and pigmentation in brown and deep skin tones has its own course worth understanding before treating anything.

That is a genuinely useful, specific, actionable thing — and it is the one recommendation in this article that carries no meaningful risk attached.

"Natural" Is Not a Safety Category

A short section because the point is short.

"Natural", "clean" and "pregnancy-safe" are marketing terms with no regulatory definition. As the wider problem with category language on packaging shows, the FDA maintains no list of approved cosmetic claims and approves no labelling before market. A bottle saying "pregnancy-safe" has met no standard, because there is no standard to meet.

This cuts both ways, and the second direction is the one worth sitting with. Botanical extracts and essential oils are not tested in pregnancy any more than synthetic actives are — frequently less. The absence of evidence is more complete, not less. "Natural" describes an origin, not a risk profile.

Take This to Your Clinician

The honest summary of everything above.

1. Oral isotretinoin is contraindicated. No ambiguity.
2. Topical retinoids are advised against, on the basis that the benefit is small rather than that harm is demonstrated — and inadvertent exposure is explicitly something specialists reassure people about.
3. Hydroquinone's 30–40% absorption is the reason it is flagged, and that number is the argument.
4. Several common ingredients are considered acceptable on absorption grounds, which is a real basis and a narrow one.
5. Salicylic acid and chemical filters are unsettled, and anyone telling you otherwise should be asked for a citation.
6. For melasma, photoprotection with iron oxides is the recommendation that is both effective and uncomplicated.

Take this list to the person responsible for your care. That is not a disclaimer — it is the actual recommendation. A page cannot know your trimester, your history or what else you are taking, and the sources above are written for clinicians to apply, not for readers to self-assign.

Conclusion

The letter categories that organise almost every page on this subject were withdrawn by the FDA in 2015, and over-the-counter products were never covered by them. Any page sorting your moisturiser into category B is using borrowed language from a retired system.

What replaces it is less tidy and more useful. Oral isotretinoin is contraindicated. Topical retinoids are advised against because the benefit is negligible, not because harm is established — and UKTIS says in the same breath that malformation risk is unlikely and that use cannot be recommended. Hydroquinone is flagged on a 30–40% absorption figure. Benzoyl peroxide, clindamycin and azelaic acid are considered acceptable because very little gets in, which is a narrower claim than "safe". Salicylic acid and chemical filters are genuinely unresolved.

And the condition that brings most people here has, as its most effective treatment, the exact combination that pregnancy rules out. What remains is sun protection with iron oxides, because visible light drives melasma and ordinary sunscreen does not block it.

This page reports positions. It does not issue clearance, and no page can. Take the list to your clinician.