Tranexamic acid wandered into skincare from antifibrinolytic medicine and immediately acquired a very beauty-industry problem: a real signal followed by a much tidier story than the evidence deserves. Topical TXA can help melasma. It is not a casual brightener, and the trials are not unanimous.
Oral and topical are two different medicines wearing one name
Tranexamic acid interferes with plasmin-related signaling involved in pigmentation. Oral, intradermal, microneedling-assisted, and topical routes have all been studied, but they are not the same decision. This file is about topical leave-on use. Oral TXA has systemic contraindications and risk considerations that do not belong inside a serum verdict.
- Systematic reviewWhat a verdict wants: people, randomized, followed long enough for the effect to be real rather than new.
- Randomized trialWhat a verdict wants: people, randomized, followed long enough for the effect to be real rather than new.
- Small human trialWhat a verdict wants: people, randomized, followed long enough for the effect to be real rather than new.
- Case seriesWhat the box usually cites. Real research, but a long way from your face — a reason to look, never proof.
- Animal modelWhat the box usually cites. Real research, but a long way from your face — a reason to look, never proof.
- In vitroWhat the box usually cites. Real research, but a long way from your face — a reason to look, never proof.
Four trials, laid out where you can watch them disagree
| Trial | Compared against | Design | What it found |
|---|---|---|---|
| Janney, 2019 | 3% hydroquinone | Randomized, roughly 12 weeks, topical 5% TXA | 27% MASI reduction versus 26.7%. A tie. |
| Atefi et al., 2017 | 2% hydroquinone | Randomized, double-blind, roughly 12 weeks, topical 5% TXA | Similar MASI improvement in both arms. |
| Na Ayuthaya, 2012 | Vehicle | Split-face, randomized, vehicle-controlled, topical 5% TXA | Both sides improved over time. TXA did not outperform vehicle. |
| Route-focused review | Oral TXA | Review across administration routes | Topical appears useful, but less effective than oral overall. |
Read down the last column and the pattern is not it works. It is: against an active comparator it holds its own; against its own vehicle it stopped being distinguishable. Both of those are true at the same time, and a product page will only ever quote you the first one.
That is not a reason to write the ingredient off. Melasma is stubborn enough that matching hydroquinone twice, with fewer irritation complaints, is a genuine result. It is a reason to stop treating one line of that table as the whole file.
What a retail serum inherits from that table, which is less than you would like
Every trial in it used 5%. Concentration, vehicle, combination ingredients, sunscreen use, and the exact pigment problem all change what a retail serum can reasonably claim. Evidence from microneedling-assisted delivery does not validate a plain topical. Evidence from a 5% study does not automatically validate a lower-strength blend with no finished-product testing.
The commonest version of this on the shelf is a 2% serum standing next to four citations it did not earn. If the brand will not print the percentage, the percentage is not 5.
Months, not a long weekend
Melasma trials commonly run for months. Consistent photoprotection is part of the treatment context because pigment can recur and UV exposure can keep feeding the problem. If the routine cannot support daily use and serious sunscreen, the serum is being asked to do a job the protocol never asked it to do alone.
Buy the sunscreen first. It is the unglamorous half of every number in that table.
Where it fits in a pigment plan
Topical TXA is most defensible for melasma or persistent hyperpigmentation when you want a non-hydroquinone option and accept that the evidence is product- and protocol-dependent. Azelaic acid, retinoids, hydroquinone, and professional treatment plans may be stronger or more appropriate depending on the diagnosis, skin, and tolerance. Pigment is not one problem wearing several names.
So: 5%, from a brand that prints the number without being asked, given the months the trials were given, under sunscreen you would not skip on a gray day. Below that specification you are funding the marketing for a molecule that already had a perfectly decent argument of its own.
Sources: Atefi et al., 2017 randomized double-blind 5% TXA-versus-2% hydroquinone trial (PMID 28748406); Janney et al., 2019 randomized 5% TXA-versus-3% hydroquinone study (PMID 31057273); Na Ayuthaya et al., 2012 split-face vehicle-controlled trial (PMID 22506692); review of TXA administration routes (PMID 36606378). Reviewed August 2026. General education, not individualized medical advice.
Source file
What this article is standing on
- British Journal of Dermatology: Self-applied topical interventions for melasma — a systematic review and meta-analysis (2022)The synthesis that holds the topical tranexamic acid evidence for melasma. It supports a real effect for self-applied topicals while showing how much of the base is small and short. It does not address oral tranexamic acid, which carries a different risk profile entirely.
- FDA: FDA works to protect consumers from potentially harmful OTC skin lightening products (2022)Regulatory context for the pigment category, and why unapproved lightening products are a live safety concern. Aimed at hydroquinone and adulterated products, though — not tranexamic acid specifically.
- FDA: Is It a Cosmetic, a Drug, or Both? (Or Is It Soap?)Why route matters here: a topical cosmetic tranexamic serum faces no premarket review, while an oral or injected form is a drug.
- FDA: Summary of Cosmetics Labeling RequirementsConcentration is never disclosed on a cosmetic label, so "contains tranexamic acid" cannot be checked against a trial dose from the packaging alone.