At 5%, urea is put into creams to help dry skin hold water. At 40%, it can soften an infected toenail until the diseased part of the nail plate can be removed. Concentration charts turn the space between those uses into a tidy handover: low urea moisturises, high urea exfoliates, and 10% is where the ingredient swaps sides.

Urea hasn’t read the chart.

There is no biochemical border at 10%. Water handling, keratin softening and effects on epidermal cells can overlap. As the concentration rises, one outcome may become more conspicuous, but the others don’t have to leave the room. The vehicle, contact time, body site and starting state of the tissue all help decide what you see.

The percentage changes the balance. It doesn’t issue the molecule with a new identity.

The Molecule Is Already in Your Skin

Urea isn’t foreign to the barrier. It is already present in the stratum corneum among the small, water-soluble compounds grouped as Natural Moisturising Factor, or NMF. In that setting, it helps the keratin-rich corneocyte retain flexibility as the surface meets much drier air than the tissue beneath it.

Calling urea a humectant is correct, although it can leave you picturing a tiny bucket that catches water and sits there. The material response is a lot more interesting. Enamul Haque Mojumdar and colleagues examined isolated stratum corneum and corneocytes at controlled humidity in 2017. At 80% relative humidity, adding urea at 20% of the dry tissue weight increased the spacing and mobility within keratin towards the state seen in untreated samples at 96% humidity.

Some water had disappeared, but urea helped the keratin-rich material behave as though its surroundings were wetter. It was not only attracting more water. It was partly standing in for what water normally did to the structure.

This was porcine tissue studied outside the body, not a trial of 20% cream on human skin. The percentage was also calculated against dry tissue rather than describing a finished product. Those coordinates should stay with the result. What the experiment gives us is a physical mechanism for why urea can preserve softer, more mobile keratin at reduced humidity without becoming a different species first.

At higher topical concentrations, enough softening and loosening of keratin-rich material can occur for the effect to be called keratolytic. This is more precise than saying urea has morphed into an exfoliating acid. The molecule is working on a different scale and in a different material.

Forty Per Cent Meets a Toenail

The far end of the range is wonderfully difficult to mistake for a face cream. In 2013, Morad Lahfa and colleagues enrolled adults with fungal infection of a great toenail. One group applied a 40% urea ointment under a plastic dressing each day for up to three weeks; the comparison group used a bifonazole-urea preparation.

Blinded evaluators judged the infected target area to have been completely removed in 61.2% of the 40% urea group, compared with 39.2% of the comparison group. By the later assessment, however, complete cure rates were not significantly different. The study supports the ointment’s ability to help remove a diseased nail. It does not show that 40% urea is a universally stronger moisturiser, or define the exact concentration at which ordinary skin begins to loosen.

Pierre Fabre Dermatologie sponsored the trial, manufactured the Onyster product under study and employed three of the authors. That involvement deserves some attention. The active comparison and blinded endpoint assessment strengthen the result, while the open-label treatment, company role and product-specific design keep the conclusion narrow.

A diseased nail plate, covered with ointment and plastic for days, is a very different biological situation from your cheek after a 5% cream. A number that sounds intense in one context can be exactly the point in another. The tissue hasn’t become a footnote to the percentage. It’s half of the equation.

Twenty Per Cent Refused Its Assigned Role

A conventional chart would place 20% urea firmly on the keratolytic side of the line. Human skin has produced a less orderly answer.

In 2012, Susanne Grether-Beck and colleagues treated twenty-one healthy volunteers once daily for four weeks with placebo, 10% urea and 20% urea at fixed sites on the arms. At the 20% site, transepidermal water loss, or TEWL, fell from 10.0 to 6.9 grams per square metre per hour, a reduction of 31%. The 10% site did not show a statistically significant TEWL change.

This does not establish 20% as the ideal moisturising concentration. The sample was small, each concentration occupied a different fixed body site, and a fall in one barrier measurement does not describe every aspect of a skin condition. One preparation in one study cannot draw the concentration chart for everyone else.

The paper did go further into mechanism. In cultured human keratinocytes, radiolabelled urea entered through transporter-sensitive routes. Urea exposure increased expression of proteins involved in epidermal differentiation, enzymes involved in lipid synthesis, and the antimicrobial peptides LL-37 and beta-defensin-2. Treated human biopsy sites showed similar increases in several of the same markers, supporting an effect on epidermal regulation beyond surface water handling.

The interesting part is the direction of travel. At 20%, urea was associated with lower TEWL, a result we would normally read as improved control of water loss rather than a barrier being progressively dismantled. Whatever else 20% urea was doing, it hadn’t cleanly crossed into a purely keratolytic role.

The commercial context is relevant here too. The study was supported by an ISDIN grant; two co-authors disclosed positions on the company’s Scientific Advisory Board, while another was an ISDIN employee. The human comparison, biopsy work and cell experiments give the paper several complementary layers of evidence, but those relationships still belong in the evaluation, particularly when a single formulation is being used to support a broader account of urea.

Ten Per Cent Did Two Jobs at Once

Ingke Hagemann and Ehrhardt Proksch found the overlap from the other direction in 1996. Ten people with psoriasis applied a 10% urea ointment to selected plaques for two weeks, while other sites received the vehicle or no treatment.

Hydration in the urea-treated stratum corneum more than doubled. At the same time, epidermal thickness fell by 29%, the measured proliferation signal fell by 51% relative to the untreated site, and several abnormal differentiation markers moved towards a less psoriatic pattern. The vehicle helped too, but urea produced the larger change in proliferation.

Psoriatic plaque is thickened, hyperproliferative tissue. You cannot carry those percentages across to normal facial skin and call the job done. The study shows something more contained and more interesting: at a concentration routinely filed under moisturising, hydration and tissue remodelling happened together. The effects had not agreed to take turns.

The Vehicle Rewrites the Percentage

If concentration were the full picture, holding it steady should produce roughly the same result. Céline Couteau and colleagues tested that assumption in 2006 by putting 5% urea into six oil-in-water emulsions and ten gels. They applied the formulations to the forearms of ten women and measured skin capacitance one hour later.

The hydration response varied across the bases. A sodium carboxymethyl cellulose gel produced the smallest change, while one polyacrylamide-based gel produced the largest. Across the formulations tested, the gels raised the reading more than the emulsions. Every label could have said 5% urea. The skin did not receive the same intervention.

This was a very small, short experiment, and capacitance is an electrical proxy for water content rather than a certificate of barrier repair. It also cannot separate the effect of released urea from hydration contributed by the rest of each base. This is exactly why the study earns its place here. The percentage remained unchanged, it was the vehicle that shifted the outcome.

Body site, contact time and tissue state continue the same argument. A light 5% gel on an ordinary forearm is not functionally identical to 5% in an occlusive preparation on a fissured heel. Forty per cent on an infected toenail is another intervention again. The number hasn’t become meaningless. It’s become incomplete.

Read the Percentage as a Direction

Concentration bands remain sensible shorthand. Lower percentages are more often formulated where increased hydration is the main aim. Higher percentages are more often used where thickened keratin needs to be softened. The bands tell you which effect is expected to dominate. They do not mark a molecular handover.

So when does urea stop being a moisturiser? There is no single percentage. Its contribution to water handling can persist at concentrations that also change keratin-rich tissue or epidermal signalling. More urea is not automatically more moisturising, and a higher number is not a faster route to barrier repair.

The graph wants sections. The skin gives you a balance.

References

Mojumdar, E.H., Pham, Q.D., Topgaard, D., et al. (2017). Skin hydration: interplay between molecular dynamics, structure and water uptake in the stratum corneum. Scientific Reports, 7, 15712. https://doi.org/10.1038/s41598-017-15921-5

Lahfa, M., Bulai-Livideanu, C., Baran, R., et al. (2013). Efficacy, safety and tolerability of an optimized avulsion technique with Onyster (40% urea ointment with plastic dressing) compared with bifonazole-urea ointment for removal of the clinically infected nail in toenail onychomycosis: a randomized evaluator-blinded controlled study. Dermatology, 226(1), pp. 5-12. https://doi.org/10.1159/000345105

Grether-Beck, S., Felsner, I., Brenden, H., et al. (2012). Urea uptake enhances barrier function and antimicrobial defense in humans by regulating epidermal gene expression. Journal of Investigative Dermatology, 132(6), pp. 1561-1572. https://doi.org/10.1038/jid.2012.42

Hagemann, I. and Proksch, E. (1996). Topical treatment by urea reduces epidermal hyperproliferation and induces differentiation in psoriasis. Acta Dermato-Venereologica, 76(5), pp. 353-356. https://doi.org/10.2340/0001555576353356

Couteau, C., Coiffard, L.J.M. and Sébille-Rivain, V. (2006). Influence of excipients on moisturizing effect of urea. Drug Development and Industrial Pharmacy, 32(2), pp. 239-242. https://doi.org/10.1080/03639040500466361

This article contains no affiliate links, sponsored content or product recommendations. Scientific claims are grounded in the peer-reviewed literature cited above. The Lahfa nail-avulsion trial was sponsored by Pierre Fabre Dermatologie, which manufactured the Onyster product, and included three company authors. The Grether-Beck paper disclosed that Peter Elias and Jean Krutmann sat on ISDIN’s scientific advisory board and Carles Trullas was a an ISDIN employee. These relationships do not erase the findings, but they make the product-specific design, fixed treatment sites, comparators and measured endpoints especially important when deciding how far the conclusions can travel.