Only a handful of places left for our September 2026 cohort. Applications close soonApply now →

MSTA
Your future
starts here.

Tranexamic acid

Also known as: TXA, tranexamic, tranexamic acid serum, Cyklokapron

Tranexamic acid is an antifibrinolytic medicine also used for melasma by oral, topical and intradermal routes. Oral use for melasma is off-label and prescriber-led. Topical formulations beat vehicle and match hydroquinone in trials, but recurrence after stopping is common by any route.

Evidence status

Moderate

Moderate-to-high for short-term topical use in melasma against vehicle, and moderate for equivalence with hydroquinone with fewer adverse events. Moderate for short-term oral use in selected patients. Durable remission after stopping, rare-event safety in months-long regimens, and efficacy for PIH all remain insufficiently established.

What it is#

Tranexamic acid is an antifibrinolytic medicine — it interferes with the breakdown of blood clots, which is why it is licensed in the UK for bleeding indications such as heavy menstrual bleeding and haemorrhage.[1]

Its use in pigmentation is a second life, and melasma is not a licensed indication for the oral tablet anywhere in that product information.

The proposed pigment mechanism runs through the plasmin pathway — inflammatory, vascular and melanocyte signalling. Worth knowing that the mechanistic detail comes from cultured endothelial cells and melanocytes, not from patients.[1, 10]i

Route changes everything#

More than almost any other active, tranexamic acid cannot be discussed as one ingredient. The same molecule by three routes gives three different conversations:

RouteUK status for melasmaEvidence
TopicalCosmetic formulations availableReviews conflict; no difference detected vs hydroquinone
OralOff-label prescribingShort-term benefit, relapse on stopping
IntradermalOutside the injectable's licensed routeLimited; not a licensed use

Topical: what the evidence shows#

Topical tranexamic acid has better evidence than it is usually given credit for — including, we should say plainly, in the first draft of this entry.

Against vehicle — and the review that disagrees

One randomised trial of a 2.5% cataplasmin 81 patients found lower end-of-treatment MASI than vehicle: mean difference −11.43 (95% CI −12.49 to −10.37), rated high certainty.[6]

That single trial is what the favourable review rests on — and a later route-stratified meta-analysis of 28 randomised trials found topical tranexamic acid not significantly betterthan comparators at 12 weeks (SMD 0.66, 95% CI −0.10 to 1.42, p=0.09), with substantial heterogeneity. Its authors say data are still required for the topical and intradermal routes.[17]

Two systematic reviews of the same literature disagree, and this entry carried only one of them. Neither establishes a class effect for whatever tranexamic serum is on the shelf.

Against hydroquinone

No difference was detectedin MASI across two 12-week trials and 160 patients (MD 0.05, 95% CI −0.29 to 0.38, moderate certainty) — and significantly moreadverse events on the hydroquinone side (RR 0.16, 95% CI 0.05–0.46).[6, 8]

The review calls them equally effective. We are not going to repeat that in our own voice: the trials set no equivalence or non-inferiority margin, so what happened is that a difference was not detected in 160 people. The tolerability difference was detected, and is the more solid half of the finding.

That is a genuinely useful finding for UK practice, given hydroquinone is prohibited in cosmetic skin-lightening products here and prescription-only in medical use.

Against triple combination cream

Inferior: MD −3.34 (95% CI −4.30 to −2.38) in 40 patients.[6]

So the accurate position is not “the evidence is mixed”. It is that topical tranexamic acid is a well-tolerated middle option: better than nothing, comparable to hydroquinone with fewer side effects, weaker than triple combination.

The caution that doessurvive is about formulation. Those results belong to a 2.5% cataplasm and to 5% gels and solutions, studied mostly in Asian and Indian populations — and one 5% preparation showed no difference from vehicle in a 21-completer Thai trial. A retail serum listing tranexamic acid somewhere in its INCI has not inherited any of this.[6, 7, 8]

Oral: off-label prescribing#

Short-term oral tranexamic acid improves melasma alongside sunscreen, and as an adjunct to hydroquinone it lowered mean MASI by 1.8 points against hydroquinone alone.[3, 4, 14]

Scope those trials properly: three months, with sunscreen throughout, in a predominantly Hispanic cohort at a single US centre and a Jakarta cohort respectively. And the hydroquinone-adjunct trial was notplacebo-controlled — the control arm received hydroquinone with no placebo capsule, and it was assessor- and analyst-blinded rather than double-blind.

The important part for a UK practitioner is regulatory, not statistical. Melasma is not a licensed indication, so this is off-label prescribing. That is entirely lawful, and it places responsibility on the prescriber for the evidence base, for explaining the off-label status, and for monitoring and follow-up.[1, 2]

Which means it is not a decision an aesthetic practitioner can make, influence or supply around.

The thrombosis question#

This is the safety crux, and it deserves to be stated without leaning in either direction.

A meta-analysis of 22 randomised trials and 49,538 non-surgical patients found no statistically significant increase in venous or arterial thrombosis. But those were bleeding indications, at different doses, for different durations, in different patients — and an absence of a detected increase is not proof of no increase, a standard we apply to our own negative findings elsewhere on this page.[5, 12, 13]i

And the strongest observational signal points the other way. This entry omitted it, which is not defensible on a page that promises not to lean in either direction. In a Danish nationwide cohort of women aged 15–49, venous thromboembolism incidence during oral tranexamic acid use was 11.8 per 10,000 person-years (95% CI 4.6–30.2) against 2.5(2.4–2.6) during non-use — an adjusted incidence rate ratio of 4.0(1.8–8.8). Arterial thrombosis was not significantly increased (IRR 1.3, 0.4–4.2).[18]

Hold the absolute scale alongside the ratio: the estimated number needed to harm was 78,549women using it for five days. A fourfold increase on a very small baseline is still a very small number of events. And the population was overwhelmingly short courses for heavy menstrual bleeding, most commonly 1 g three times daily for five days — not months of low-dose use for melasma.

So the two evidence sets differ in design, dose, duration and indication, and neither answers the melasma question. Do not claim zero risk. Do not claim a melasma-specific rate either, because no source here provides one.[5, 12, 13, 18]i

Against that: one deep-vein thrombosis was recorded in a 561-patient melasma series — in a patient with familial protein-S deficiency. One event does not establish a rate, but it does illustrate exactly why screening for thrombotic risk factors and a personal or family history sits with a prescriber.

Intradermal and injectable#

The UK injectable product is licensed for slow intravenous use only. Intradermal delivery for melasma is outside its licensed route.[11]

On microneedling delivery, one 20-person split-face trial found no significant added benefit from tranexamic acid after whole-face needling.[9]

Read that honestly in both directions: it is a real negative result, and with twenty participants it cannot exclude a genuine effect either. What it certainly does not do is validate needling a retail serum into the face.

What happens when you stop#

Recurrence is the defining feature, and it is where expectations most often break down.

In a 561-patient series, relapse among responders was 27.2%. In the randomised adjunct trial, relapse over three months of follow-up was not significantly differentbetween the treated and control groups — 30% versus 26%.[5, 14]

Read the second finding carefully, because we misread it. Relapse proportions did not differ significantly. But mean MASI at six months was still 1.8 points lowerwith tranexamic acid plus hydroquinone than with hydroquinone alone (95% CI 0.36–3.24, p=0.015). A non-significant difference in how many people relapsed is not the same as no measurable advantage remaining — and this page previously said the second.

Note the route, too. Both of those sources concern oral tranexamic acid. Neither tells you what happens after stopping a topical or intradermal course; nobody here has measured that.

PIH and other pigmentation#

Tranexamic acid is increasingly marketed for post-inflammatory hyperpigmentation. The prominent report behind that is an uncontrolled practice series of around 82 cases with no controlled outcome data.[15]

That is a reasonable prompt for further study. It is not a basis for telling a client it works for their post-acne marks, and the melasma evidence does not transfer to every cause of a dark mark.

Pregnancy#

Pregnancy data for systemic tranexamic acid covers roughly 230 reported exposures, mostly acute late-pregnancy use for medical indications. Within that data, no increase in malformations has been reported, and animal studies show no evidence of teratogenicity.[16]

Both halves of that matter. It is not an alarming signal — and it is also nothing like a safety dataset for a months-long cosmetic regimen in a healthy pregnant person, which is a completely different exposure from a single acute obstetric dose.

For topical specifically, we found no study assessing use in pregnancy at all.[no source found]

In professional practice#

  • Ask which routebefore anything else. The word “tranexamic” on its own tells you nothing.
  • Oral belongs to an appropriately qualified prescriber acting within competence — which some aesthetic practitioners are and some are not. If you are not one: refer, and do not sell or supply. Discussing the evidence and supporting a referral is neither supply nor prescribing.[2]
  • Be honest that the topical reviews conflict. One rests on a single cataplasm trial; a route-stratified review of 28 trials found no significant effect at 12 weeks.[17] Check the formulation, not the ingredient list.
  • Set expectations at control, not cure. Relapse after stopping was no different from control in one trial.
  • Do not needle a retail serum. The one trial of delivery through microneedling found no added benefit.

What remains uncertain#

  • Whether any topical formulation maintains benefit long term, or only during use.
  • Rare-event safety of months-long oral regimens in healthy people — the reassuring meta-analysis studied bleeding indications.
  • Whether it does anything useful in PIH.
  • Whether the negative microneedling result reflects no effect or an underpowered study.
  • Anything at all about topical use in pregnancy.

Common misconceptions#

“The topical evidence is weak.”

It is more contested than weak. One cataplasm trial beat vehicle at high certainty;[6] a route-stratified review of 28 trials found no significant effect at 12 weeks.[17] And against hydroquinone, no difference was detected with fewer adverse events.[6, 8] The real caveat is formulation specificity and two reviews that disagree.

“It’s a safer hydroquinone.”

Better tolerated, on the trial evidence. But it also lost to triple combination cream, so “safer” is not “stronger”.[6]

“Oral tranexamic acid is basically a supplement.”

It is a prescription-only medicine used off-label, with a thrombotic-risk assessment attached.[1, 2]

“It’s proven not to cause clots.”

No significant increase was detected in bleeding-indication trials. That is not the same as proof of no risk in a different dose, duration and population.[5, 12, 13]i

“It works for post-acne marks.”

That rests on an uncontrolled series of about 82 cases.[15]

Frequently asked questions#

Can I retail a topical tranexamic acid serum?

Topical cosmetic formulations are available, and the category has genuine vehicle-controlled evidence. Ask the supplier which formulation and concentration was studied — the results belong to specific preparations, and one 5% product showed no benefit.[6, 7, 8]

A client asks about oral tranexamic acid. What do I say?

That it is a prescription medicine used off-label for melasma, that it works short-term, that it requires a thrombotic-risk assessment, and that the decision belongs with a prescriber who will monitor them.[1, 2]

How does it compare with hydroquinone?

No significant MASI difference across 160 patients, with significantly fewer adverse events on the tranexamic acid side.[6, 8]

Will the melasma stay away afterwards?

Often not. Relapse among responders was 27.2% in one series, and in a randomised trial relapse at three months was no different from control.[5, 14]

Can it be injected or needled in?

The UK injectable is licensed for slow intravenous use only. Intradermal delivery for melasma sits outside that, and the one microneedling-delivery trial found no added benefit.[11]

References#

Each source is graded by evidence tier. Tier 4 material (manufacturer documents, expert consensus, practitioner experience) is useful for protocol and context, and is never presented as equivalent to independent clinical evidence.

  1. Sovereign Medical. Tranexamic Acid 500 mg Tablets: Summary of Product Characteristics. Electronic Medicines Compendium. Updated 5 May 2026.Tier 1Supports: Product pharmacology, licensed indications, contraindications, thrombotic cautions, pregnancy and lactation information. Melasma is not a listed indication. NOTE: this SmPC does not use the phrase 'synthetic lysine analogue' — do not attribute it here.
  2. Medicines and Healthcare products Regulatory Agency. Off-label or unlicensed use of medicines: prescribers' responsibilities. Drug Safety Update.Tier 1Supports: UK responsibilities for evidence, prescribing, communication, monitoring and follow-up when a medicine is used off-label.
  3. Del Rosario E, Florez-Pollack S, Zapata L Jr, et al. Randomized, placebo-controlled, double-blind study of oral tranexamic acid in the treatment of moderate-to-severe melasma. Journal of the American Academy of Dermatology. 2018;78(2):363–369.Tier 2Supports: Three-month oral regimen plus sunscreen, 44 enrolled and 39 completing, followed three months after stopping. Population was predominantly Hispanic women at a single US centre.
  4. Shihab N, Prihartono J, Tovar-Garza A, et al. Randomised, controlled, double-blind study of combination therapy of oral tranexamic acid and topical hydroquinone in the treatment of melasma. Australasian Journal of Dermatology. 2020;61(3):237–242.Tier 2Supports: Three-month incremental oral effect in 50 completers already using hydroquinone and sunscreen, with three months of post-treatment follow-up. Jakarta cohort.
  5. Lee HC, Thng TGS, Goh CL. Oral tranexamic acid in the treatment of melasma: a retrospective analysis. Journal of the American Academy of Dermatology. 2016;75(2):385–392.Tier 3Supports: Uncontrolled 561-patient Singapore series. Relapse among responders was 27.2%. One deep-vein thrombosis occurred, in a patient with familial protein-S deficiency.
  6. Pennitz A, Kinberger M, Avila Valle G, Passeron T, Nast A, Werner RN. Self-applied topical interventions for melasma: a systematic review and meta-analysis of data from randomized, investigator-blinded clinical trials. British Journal of Dermatology. 2022;187(3):309–317.Tier 1Supports: Concludes: 'Topical TXA proved to be more effective than placebo and equally effective as HQ in treating melasma. However, TXA was inferior to TCC with respect to MASI.' Versus vehicle: MD −11.43 (95% CI −12.49 to −10.37), GRADE HIGH, 81 patients, 2.5% cataplasm. Versus hydroquinone: no significant MASI difference, MD 0.05 (95% CI −0.29 to 0.38), GRADE moderate, 160 patients across two studies — with significantly MORE adverse events on hydroquinone, RR 0.16 (95% CI 0.05–0.46). Versus triple combination cream: TXA inferior, MD −3.34 (95% CI −4.30 to −2.38), 40 patients.Funding / interest: Supported by an unrestricted institutional research grant from Beiersdorf AG; the authors state the sponsor had no influence. T. Passeron disclosed grants and honoraria from Beiersdorf and other companies.
  7. Kanechorn Na Ayuthaya P, Niumphradit N, Manosroi A, Nakakes A. Topical 5% tranexamic acid for the treatment of melasma in Asians: a double-blind randomized controlled clinical trial. Journal of Cosmetic and Laser Therapy. 2012;14(3):150–154.Tier 2Supports: Twelve-week split-face trial in a Thai cohort: 23 enrolled, 21 completed, no significant difference between the studied 5% TXA and vehicle. A single small negative trial, not the weight of the vehicle-controlled evidence.
  8. Janney MS, Subramaniyan R, Dabas R, Lal S, Das NM, Godara SK. A randomized controlled study comparing the efficacy of topical 5% tranexamic acid solution versus 3% hydroquinone cream in melasma. Journal of Cutaneous and Aesthetic Surgery. 2019;12(1):63–67.Tier 2Supports: Twelve-week single-blind comparison in 100 analysed Indian patients. The authors concluded the two were comparable in MASI reduction. Note this is a 5% SOLUTION, not a gel or cataplasm.
  9. Kuster Kaminski Arida D, Rebellato PRO, de Campos GLM, et al. Randomized, double-blind, placebo-controlled split-face trial of the efficacy of tranexamic acid by drug delivery through microneedling in the treatment of melasma. Journal of Cosmetic Dermatology. 2021;20(12):4005–4010.Tier 2Supports: Twenty-person split-face study finding no significant incremental TXA effect after whole-face microneedling. With n=20 this cannot exclude a real effect — an undetected difference is not proof of no difference.
  10. Zhu JW, Ni YJ, Tong XY, et al. Tranexamic acid inhibits angiogenesis and melanogenesis in vitro by targeting VEGF receptors. International Journal of Medical Sciences. 2020;17(7):903–911.In vitroTier 3Supports: Human umbilical-vein endothelial cells and cultured normal human melanocytes. Not a patient outcome or topical absorption study.
  11. Bowmed Ibisqus Limited. Tranexamic Acid 1 g/10 ml solution for injection/infusion (100 mg/ml): Summary of Product Characteristics. Electronic Medicines Compendium. Text revised 28 April 2026.Tier 1Supports: The eMC record is under Bowmed Ibisqus Limited, with Ibigen S.r.l. as marketing-authorisation holder. The injectable is limited to SLOW INTRAVENOUS use — not intradermal delivery for melasma.
  12. Meaidi A, Mørch LS, Torp-Pedersen C, Lidegaard Ø. Oral tranexamic acid and thrombosis risk in women. EClinicalMedicine. 2021;35:100882.Tier 3Supports: Danish cohort of short-course use for heavy menstrual bleeding. Indirect to months-long melasma regimens.
  13. Chornenki NLJ, Um KJ, Mendoza PA, et al. Risk of venous and arterial thrombosis in non-surgical patients receiving systemic tranexamic acid: a systematic review and meta-analysis. Thrombosis Research. 2019;179:81–86.Tier 1Supports: No statistically significant increase in venous or arterial thrombosis across 22 bleeding-indication RCTs and 49,538 patients. Indirect to melasma duration, dose and patient selection — and an absence of a detected increase is not proof of no increase.
  14. Lajevardi V, Ghayoumi A, Abedini R, et al. Comparison of the therapeutic efficacy and safety of combined oral tranexamic acid and topical hydroquinone 4% treatment vs. topical hydroquinone 4% alone in melasma: a parallel-group, assessor- and analyst-blinded, randomized controlled trial with a short-term follow-up. Journal of Cosmetic Dermatology. 2017;16(2):235–242.Tier 2Supports: Adjunctive oral TXA lowered mean MASI by 1.8 points versus hydroquinone alone (P = 0.015). Relapse over three months of follow-up did not differ significantly (30% vs 26%). IMPORTANT: this is NOT placebo-controlled — the control arm received hydroquinone only, with no placebo capsule — and it was assessor- and analyst-blinded rather than double-blind.
  15. Lindgren AL, Austin AH, Welsh KM. The use of tranexamic acid to prevent and treat post-inflammatory hyperpigmentation. Journal of Drugs in Dermatology. 2021;20(3):344–345.Tier 4Supports: A brief practice report of approximately 82 cases with no controlled outcome data. Insufficient for any broad PIH efficacy claim.
  16. UK Teratology Information Service. Use of tranexamic acid in pregnancy. Version 3, June 2022.Tier 4Supports: Pregnancy evidence for systemic TXA is limited: around 230 reported exposed pregnancies, mainly acute late-pregnancy use for medical indications. UKTIS reports no increase in malformations in the available data and no evidence of teratogenicity in animal studies — but this does not establish safety for months-long oral melasma regimens.
  17. Panchal R, et al. Efficacy of tranexamic acid in the treatment of melasma stratified by route of administration: a systematic review and meta-analysis of randomised controlled trials.Tier 1Supports: THE CONTRADICTORY REVIEW. 28 randomised controlled trials, stratified BY ROUTE. Verbatim on the topical route: 'At 12 weeks, topical TXA did not show a significant change with SMD of 0.66, 95% CI -0.10-1.42, P = 0.09 compared to adjuvant treatment' — with substantial heterogeneity, I-squared 92% (P<0.00001). Its conclusion is that 'data are still required for topical and intradermal routes'. Registered because the entry previously carried only the review that favours topical use; two systematic reviews of the same literature disagree, and a page built on claim-level accountability has to say so.
  18. Meaidi A, Morch L, Torp-Pedersen C, Lidegaard O. Oral tranexamic acid and thrombosis risk in women. EClinicalMedicine. 2021;35:100882.Tier 3Supports: THE STRONGEST OBSERVATIONAL SIGNAL IN THIS DOMAIN, and the entry omitted it. Danish nationwide cohort of women aged 15-49. Venous thromboembolism incidence 11.8 per 10,000 person-years (95% CI 4.6 to 30.2) during oral tranexamic acid use versus 2.5 (2.4 to 2.6) during non-use; adjusted incidence rate ratio 4.0 (1.8 to 8.8). Arterial thrombosis was NOT significantly increased: IRR 1.3 (0.4 to 4.2). ABSOLUTE SCALE: the estimated number needed to harm was 78,549 women using oral tranexamic acid for five days to produce one additional VTE. SCOPE: the population was overwhelmingly short courses for heavy menstrual bleeding - most commonly 1 g three times daily for five days - not months of low-dose use for melasma. It is a real signal at a tiny absolute rate, in a different indication and duration.

Continue learning with MSTA#

Skinipedia is written by the educators at MSTA, the medic-led skincare training academy in Liverpool.