What Percentage of Tranexamic Acid Is Effective? 2%, 3% and 5% Compared
Topical tranexamic acid has been studied on skin at 2%, 3% and 5%. A table of every trial by concentration, what each found, why the formula matters, and how long results take.
Published skin studies have used topical tranexamic acid at 2%, 3% and 5%, and all three lowered pigment scores over 8 to 12 weeks. No trial has compared those strengths head to head, so the evidence does not show that 5% works better than 2% or 3%. Below: a table of what each concentration did in real skin trials, why the formula and the weeks matter as much as the number, and what the studies say about irritation.
What percentage of tranexamic acid is effective on skin?
In published trials, 2%, 3% and 5% have all been effective. Each of those strengths reduced the look of dark patches in at least one controlled or comparison study.
Most of that research is on melasma, the patchy brown pigment that often shows up on cheeks and forehead. Researchers grade it with the Melasma Area and Severity Index, or MASI, a score dermatologists use to rate how much of the face is affected and how dark it is. A lower score means a better result.
If you search for tranexamic acid doses, most results are about bleeding. That is because tranexamic acid began as a medicine. In the US it is a prescription tablet for heavy menstrual bleeding, sold as 650 mg tablets. Those numbers are milligrams of a pill. They do not tell you anything about a percentage on your face. For the background on the ingredient, see tranexamic acid in skincare.
Tranexamic acid 5% vs 3% vs 2%: what did each study find?
Each strength has its own small set of trials. Here is what the published skin studies tested and found, grouped by concentration.
| Strength | Study | Who and how long | What it found |
|---|---|---|---|
| 2% | Kim 2016 | 23 people with melasma, whole face, 12 weeks, no control group | Scores improved in 22 of 23. Biopsies showed less melanin in the outer skin. |
| 2% (with 2% niacinamide) | Lee 2014 | 42 Korean women, twice daily, 8 weeks, against a plain cream | Reduced the look of uneven pigment more than the plain cream, with sunscreen in both groups. |
| 2% (with 2% vitamin C) | Kaikati 2023 | 10 women with stubborn melasma, nightly, 8 weeks, pilot | Average MASI fell from 12.76 to 3.39. No major side effects. |
| 3% | Ebrahimi 2014 | 50 women, split face, 12 weeks, against hydroquinone plus a steroid | Both sides improved with no significant difference. Fewer side effects on the tranexamic acid side. |
| 3% | Yasnova 2024 | 20 people, split face, 8 weeks, against 4% hydroquinone | Scores dropped on both sides by week 4 and week 8. Results were similar. |
| 5% | Kanechorn Na Ayuthaya 2012 | 23 women, split face, 12 weeks, against the same gel without the active | Both sides lightened. The 5% side was not better than the plain gel, and it caused more redness. |
| 5% | Atefi 2017 | 60 women, twice daily, 12 weeks, against 2% hydroquinone | Both groups improved by a similar amount. No side effects in the tranexamic acid group. |
| 5% (liposomal) | Banihashemi 2015 | 30 women, split face, 12 weeks, against 4% hydroquinone | Both sides improved. The larger drop on the tranexamic acid side was not statistically significant. |
| 5% | Heidary 2025 | 50 women, 12 weeks, 5% cream against an oral tablet | MASI fell 50.88% with the cream and 58.86% with tablets. The gap was not significant. |
| 5% | Sobhan 2023 | 60 acne patients with brown marks, 12 weeks, against 20% azelaic acid | Both improved by a similar amount. Fewer side effects with tranexamic acid in the first month. |
Two patterns stand out. First, every strength improved scores from where people started. Second, the trials compared tranexamic acid with other products, not with other strengths of itself.
Is a higher percentage of tranexamic acid better?
The published evidence does not show that. We could not find a trial that tested 2%, 3% and 5% against each other in the same people.
The one placebo-style trial of 5% is a useful warning. In the 2012 split-face study, the 5% gel and the plain gel both lightened skin, and the difference between them was not significant. The authors noted more redness on the tranexamic acid side. Meanwhile, the 2014 trial at 2%, paired with niacinamide, did beat its plain cream.
That does not mean 2% is stronger. The studies used different people, formulas and measuring tools. What it does mean is that a bigger number on a label is not proof of a bigger result.
Large reviews point the same way. A 2023 meta-analysis of topical melasma agents found topical tranexamic acid as effective as hydroquinone, azelaic acid and cysteamine overall. A 2023 review of delivery routes found tablets worked best and topical use alone the least, though topical use was better tolerated than hydroquinone.
Does the formula matter as much as the percentage?
Yes. Tranexamic acid is highly water-loving, and a 2025 study in volunteers described its entry into skin as poor and dependent on the formula. Using light-based scanning on living skin, it measured how much entered from plain 1% and 3% formulas and from a modified version. More entered from the 3% formula than the 1%, and the modified version beat both.
Formulators work on this problem in several ways, such as liposomes and niosomes, tiny fat-based carriers. Two of the trials above used them. In a 2025 trial of 99 patients, a niosomal cream with 2% tranexamic acid and 2% niacinamide did as well as 4% hydroquinone over three months. So did a regular cream with 5% tranexamic acid and 4% niacinamide.
The practical point: how a product gets the ingredient into skin can matter as much as how much of it is in the bottle. For more on why serums sit on the surface, see why your serum is not absorbing.
How long does each strength take to show a difference?
Expect about 8 weeks before a clear change, at any strength. Most trials ran 8 to 12 weeks.
- Week 4: some studies saw early drops. The 3% split-face trial and the 2% pilot both reported lower scores by week 4.
- Week 8: a 2024 network meta-analysis found topical tranexamic acid differed significantly from placebo starting at week 8. Tablets showed a difference from week 4.
- Week 12: most 5% trials measured their main result here.
Every one of these trials also used sunscreen. Several assigned a sunscreen to apply each morning. Sunlight makes skin produce more pigment, which can darken existing melasma, according to the AAD. See how long until skincare works.
Does a higher percentage irritate more?
Tranexamic acid is one of the gentler pigment ingredients at the strengths studied. In the 2023 meta-analysis, irritation was reported in 0.8% of people using topical tranexamic acid. The figure was 18.7% for azelaic acid and 50.9% for hydroquinone combinations.
Gentle is not the same as zero. The 5% gel in the 2012 trial caused more redness than its plain gel. In the 2025 cream trial, one person on 5% stopped because of skin sensitivity. If a product stings, burns or leaves skin red for hours, stop and see a dermatologist.
Patch testing first is a simple step. The American Academy of Dermatology suggests applying a new product to a small test spot twice daily for seven to 10 days. See how to patch test.
Which strength should you look for?
Any of the studied strengths, 2% to 5%, has evidence behind it. Pick the one in a formula your skin tolerates, and use it long enough to judge.
- Check that the percentage is stated. A product that names its strength lets you compare it with the studies above.
- Look at the whole formula. Many trials paired tranexamic acid with niacinamide or vitamin C. See tranexamic acid vs niacinamide.
- Plan for 8 to 12 weeks of steady use, with sunscreen every morning.
- See a dermatologist if patches are spreading, very dark, or not changing after three months. Tablets are a prescription option only a doctor can decide on.
Where re:clear-T fits
re:clear-T, the Glow ampoule, contains tranexamic acid at 5% with niacinamide at 2%. That puts it at the top of the range used in published skin studies, paired with the niacinamide that several of those trials also used. It is chosen for uneven tone, dark spots and marks.
It was lab tested in South Korea. UV-induced melanocyte activation was suppressed 68.4% in an ex vivo test (2025). Tyrosinase, a key enzyme in making pigment, was inhibited 51.3% in an enzymatic assay. These are lab results, not a clinical trial on people.
Each 50 mL glass vial is crimp-sealed and drawn with a 1 mL syringe through the rubber seal, so no air or fingers get in. It is USD 25.90. Some people use it with the re:tones Tap Pen about once a week. You can see the ampoule on the re:clear-T page.
Frequently asked questions
What percentage of tranexamic acid is effective?
Published skin trials found 2%, 3% and 5% all lowered pigment scores over 8 to 12 weeks. No trial has compared them directly, so no single strength is proven best. Formula, sunscreen and time matter as much as the number.
Is 5% tranexamic acid better than 2%?
The evidence does not show that. One 5% gel did no better than its plain base in a 2012 trial, while a 2% cream with niacinamide beat its plain cream in 2014. The studies were too different to rank the strengths.
Is 3% tranexamic acid enough?
In two split-face trials, 3% did about as well as hydroquinone-based creams over 8 to 12 weeks. That makes 3% a studied strength, not a weak one. Results depend on the formula and on daily sunscreen.
Can you use more than 5% tranexamic acid on skin?
Almost all published face studies stayed between 2% and 5%. A 10% solution shows up in a single case report on facial redness, so there is little data on higher strengths. More is not shown to be better, and it may irritate more.
How is a skin percentage different from tranexamic acid tablets?
Tablets are a prescription medicine measured in milligrams, used for heavy menstrual bleeding and sometimes prescribed off label for melasma. A percentage on a serum describes how much is in a cosmetic formula. Only a doctor can decide whether tablets are right for you.
How long should you use tranexamic acid before judging it?
Give it at least 8 weeks, and up to 12. A 2024 network meta-analysis found topical use separated from placebo starting at week 8. Take a photo in the same light at the start so you can compare fairly.
