LENGTH MATTERS: WHICH CERAMIDES SUPPORT THE SKIN BARRIER?

The stratum corneum — the outermost layer of the epidermis — is often compared to a brick wall: corneocytes play the role of bricks, while the intercellular lipid matrix acts as the mortar holding them together. The main components are ceramides, cholesterol, and free fatty acids, with ceramides forming its backbone. Together, these three lipid classes retain water within the skin and limit the entry of irritants and microorganisms from the outside.

A ceramide is a compound of a sphingoid base (an amino alcohol, such as phytosphingosine) and a fatty acid linked by an amide bond. Ceramides are enormously diverse — this diversity is determined by chain length, degree of unsaturation, and the position of hydroxyl groups. Barrier function depends both on how many ceramides are present in the skin and on which particular molecules they are.

It is precisely this nuance — the length of the fatty-acid chain within the ceramide molecule — that a study by South Korean researchers, published in 2026 in the Journal of Cosmetic Dermatology, brings into focus [1]. The authors asked: does the length of the fatty-acid chain of a ceramide applied to the skin matter?

Not all ceramides are alike: molecular structure and why it matters

Figures such as "C18" or "C24" in a ceramide's description refer to the number of carbon atoms in the fatty-acid chain of the molecule, not in the molecule as a whole. And this is no mere formality.

In subclass notation, the first letter indicates the fatty acid type, and the second indicates the sphingoid base type.

Fatty-acid types (first letter):

N — non-hydroxy;
A — alpha-hydroxy;
EO — esterified omega-hydroxy.

Sphingoid-base types (second letter):

S — sphingosine;
DS — dihydrosphingosine (sphinganine);
P — phytosphingosine;
H — 6-hydroxysphingosine.

Examples decoded:

CER NP — non-hydroxy fatty acid + phytosphingosine;
CER AP — alpha-hydroxy fatty acid + phytosphingosine;
CER NH — non-hydroxy fatty acid + 6-hydroxysphingosine;
CER AH — alpha-hydroxy fatty acid + 6-hydroxysphingosine;
CER AS — alpha-hydroxy fatty acid + sphingosine.

Chain length and its link to barrier status

In human skin, ceramides are for the most part long-chain: according to the data cited in the paper, 80–90% of stratum corneum ceramides carry fatty-acid chains longer than C24 [2].

A whole body of observations accumulated long before this study points to a link between chain length and barrier status. Researchers first noticed this pattern in 1985: analysis of amide-linked fatty acids in human stratum corneum ceramides showed that C24 and C26 chains predominate in most ceramide subclasses, while the C30 chain is characteristic mainly of ω-esterified ceramides. Later lipidomic studies refined the picture: C26 and C28 chains are especially abundant in the CER NP, CER NH, CER AP, and CER AH subclasses, whereas the AS ceramide carries predominantly short chains, such as C16.

Clinical observations complete this picture. In patients with atopic dermatitis, ceramides in affected skin have, on average, shorter chains than in healthy skin [3]. And in people with dry skin, levels of long-chain ceramides — primarily C26 and C28 — are reduced compared with skin without marked dryness [4]. Differences between body sites are also telling: the face, shins, scalp, and lips each have their own "ceramide profiles" that correspond to the barrier properties of each area.

On this basis, the authors suggest that deviations in ceramide chain length may serve as a biomarker of barrier impairment in various skin conditions. This is a hypothesis, not a fact.

The authors explain the mechanism through lipid packing: longer chains promote the formation of a dense, well-ordered lipid lamellar structure that retains moisture better and allows less passage of substances. This explanation draws on earlier laboratory and modeling work; in the present study, the structure of the lipid layers was not measured directly.

Three creams, one concentration: 0.5%

To test the effect of chain length, the authors prepared creams with the same ceramide concentration — 0.5% (by weight) — but with different compositions. All ceramides belonged to a single subclass, CER NP, that is, ceramides based on phytosphingosine and non-hydroxy fatty acids.

Four formulations were compared: a base without ceramides or sphingoid bases (control), a cream with a single C18 ceramide (positive control — the classic "short" ceramide), a blend of ceramides with C16–C24 chains, and a blend with ultra-long C24–C30 chains.

One nuance matters here: the C16–C24 blend actually consisted mostly of C18, whose share was 86%, while C24 accounted for only 0.3%. The C24–C30 blend, by contrast, was genuinely long-chain: C24 — 39%, C26 — 19%, C28 — 32%, C30 — 10%. In other words, the two "mixed" creams differed fundamentally.

Study results

The authors assessed how the different formulations affected three measures of barrier status — the rate of recovery after damage, hydration, and the strength (cohesion) of the stratum corneum — and then compared these results with changes in the skin's ceramide profile.

Recovery rate: what tape stripping revealed

The functional part of the study used a double-masked, base-controlled design with intra-individual comparison. It included 21 healthy young women (mean age 24.7 ± 4.7 years, with no history of dermatological disease), using the inner forearms as test sites.

First, the researchers induced acute barrier damage: surface layers of the stratum corneum were repeatedly stripped with adhesive tape until transepidermal water loss (TEWL) reached a set level. The creams were then applied immediately, and barrier recovery was assessed at 4 and 8 hours.

At 8 hours, the cream with ultra-long-chain C24–C30 ceramides produced 76% barrier recovery, compared with 65% for the classic C18 ceramide. The difference was statistically significant. The C16–C24 blend did not differ significantly from C18. Thus, the formulation with long chains showed the fastest effect. One participant did not complete the measurements in this test, so the barrier-recovery data are reported for 20 participants.

Hydration: four weeks of observation

In the second part of the study, the creams were applied to the forearms twice daily for four weeks, and skin hydration was measured with a corneometer at weeks 2 and 4.

At four weeks, relative to the untreated control site, C18 produced 21.6%, the C16–C24 blend 39%, and the C24–C30 blend 47%. Both blends significantly outperformed C18 in hydration (p < 0.05 and p < 0.01, respectively), but the two blends did not differ significantly. Notably, already at week 2, only the C24–C30 formulation significantly exceeded C18.

These percentages reflect the difference from the untreated site, not an increase from each site's baseline.

Stratum corneum strength: a cohesion test

Cohesion is the ability of the stratum corneum to remain intact under mechanical stress. After four weeks of use, eight randomly selected participants were assessed for how TEWL changed after 15 consecutive tape strippings: the smaller the rise in TEWL, the more resilient the barrier.

All ceramide-containing creams improved cohesion compared with the base. The C24–C30 formulation showed a marked improvement over C18, whereas there was no significant difference between C18 and the C16–C24 blend.

A look inside: what lipid analysis showed

To understand what happens at the molecular level, the authors collected stratum corneum samples with adhesive tape and measured individual CER NP species by liquid chromatography–tandem mass spectrometry (LC–MS/MS).

The results are consistent with the functional data: after application of the creams with C18 and the C16–C24 blend, levels of short-chain ceramides (C16, C18, C20) rose, while after the C24–C30 blend, levels of long-chain ceramides (C24–C30) rose. However, the analytical part is very small: the figure caption states n = 3, and the authors do not disclose how the replicates were organized. Moreover, increased ceramide content in the samples does not, by itself, prove that the applied molecules incorporated into lipid structures or stimulated the skin's own ceramide synthesis.

What this means for your practice

The word "ceramides" on an ingredient list should not be treated as something uniform: molecules differ in subclass and in fatty-acid chain length, and this can affect their effect. On healthy skin, a blend of ultra-long-chain C24–C30 ceramides outperformed classic C18 in barrier recovery rate, hydration, and cohesion, making such formulations an interesting direction for barrier-focused cosmetics.

The practical difficulty is that chain length is nearly impossible to judge from the label: the ingredient list typically shows, at best, the subclass (for example, Ceramide NP), not the fatty-acid chain length. At the same time, the most common ceramide — CER NP — is produced predominantly with the short C18 chain, whereas in the skin's own stratum corneum, 80–90% of ceramides carry chains longer than C24. In other words, the presence of a ceramide in a product does not by itself mean it is structurally close to the native one. Chain length should be clarified directly with the manufacturer — standard labeling does not reveal it.

The study's limitations should also be kept in mind. The sample was small; it included only healthy young women, the forearms were observed over a short period, and some measurements were carried out on small subgroups: 8 participants for cohesion and n = 3 for the lipid analysis. No tolerability or adverse-event data are reported in the paper — their absence does not mean an absence of such events has been proven. This work therefore does not provide ready-made care protocols after peels, laser procedures, or in dermatoses.

In summary: the fatty-acid chain length of a ceramide is a detail that deserves attention both in formulation and in evaluating barrier products. The results on healthy skin support further study of blends of ultra-long-chain C24–C30 ceramides, but universal clinical recommendations require additional data — above all in larger samples, in dermatoses, and in longer-term observations.