The Eye Area as Skin Ages: What the Evidence Supports, and What Is Cosmetic Camouflage

The Eye Area as Skin Ages: What the Evidence Supports, and What Is Cosmetic Camouflage

The short version

  • Under-eye darkness is not one condition. It can be pigment, blood vessels, lost volume or a mix, and a cream only acts on pigment.
  • Sleep is not the switch. One night without sleep made no measurable difference to under-eye darkness in a randomised study of 181 people.
  • Topicals move texture and hydration by modest amounts over weeks. The clearest numbers come from a three-month vitamin A trial that had no placebo arm.
  • Anything visible in the first fifteen minutes is water and light, not remodelling.
  • Most eye-cream evidence is small, short and written by the company selling the cream. We sell eye products, and no study here tested one of ours.

Dark circles are four problems wearing one name

A 2023 review classifies under-eye darkness as pigmented, vascular, structural or mixed, and says the cause is often multifactorial or difficult to work out in a given person (Samaan, 2023).

A systematic review of 39 studies found fillers and fat grafting most effective when the darkness came from volume loss and creams and peels when the issue was pigment. Given the scarcity of high-quality evidence, it said its own recommendations should be interpreted selectively (Michelle, 2021).

Sleep is not the culprit

Researchers randomised 181 people to one night of total sleep deprivation or a normal night. Neither the deprivation nor their own sleepiness was related to any facial variable measured.

Under-eye darkness did predict how tired other people rated a face as looking (Holding, 2019). No cosmetic company was involved in that work.

Vitamin A has the longest record

In a randomised double-blind trial, 40 Korean women used retinaldehyde cream at 0.1 or 0.05 per cent twice daily for three months. Texture improved 13.7 and 12.6 per cent and hydration rose 10.2 and 6.0 per cent. Only the 0.1 per cent cream shifted the melanin index (Kwon, 2018).

Both arms received an active cream, so there was no placebo. The trial cannot separate the vitamin A from three months of consistent moisturising.

The peptide trial is one molecule from one company

The trial usually pointed to ran 56 days against a retinol serum and a vehicle. The peptide reduced the number, area and roughness of crow's feet, while retinol reduced only the area (Chang, 2025).

The retinol comparator was dosed at 0.002 per cent, far below normal formulation levels, so beating it is a low bar. All eight authors work for Shanghai Chicmax Cosmetic and Shanghai Kans Biotech, and the paper states they declare no conflicts of interest.

Eye creams as a category

One study followed 33 women for 56 days, reporting hydration and surface improvement within fifteen minutes, and wrinkle reduction measured at seven days (D'Arcangelis, 2025). There was no placebo arm among those women, and all authors are full-time employees of Colgate-Palmolive.

The second was open-label over 12 weeks with no control group. Of 40 women aged 35 to 60, 37 completed, reporting improvement in dark circles and puffiness (Rajabi-Estarabadi, 2024). Three of the five authors list Revision Skincare, which makes the cream.

Three of the four topical studies here were authored by the manufacturer's own staff, two of them entirely. That does not make them false. It does mean the real-world effect is likely smaller than the paper reads.

Where cosmetic camouflage begins

Nothing structural happens to skin in fifteen minutes. The usual account is that water content rises and the surface becomes optically smoother, so fine lines catch less shadow. That is the standard explanation, not something the study demonstrated.

Light-diffusing pigments and concealer work the same way. They change the photograph, not the skin, which is fine unless it is sold as remodelling. No study here measured what happens once you stop.

Hollows are a different job

A systematic review of 42 papers on tear trough filler found that retrospective descriptive work was the most common grade of evidence, that no preference between needle and cannula could be made, and that standardising the procedure was not currently feasible (Rao, 2022).

We sell these products, so read this bit twice

None of the studies above tested a product we stock. Formulation and concentration change how an ingredient behaves, so those results do not carry across to a tube on our shelf.

Medik8 built its reputation on vitamin A and its eye range reflects that, alongside peptide options. Both families appear above, which is a reason to look at them rather than to rank them. One of the four studies tested five ingredients at once, so it cannot tell you which did the work.

What to actually do

  • If the concern is a shadow rather than texture, get the subtype assessed before you spend.
  • Judge an eye product at eight to twelve weeks. That is our house rule, not a figure from these trials.
  • Build vitamin A slowly around the eye, two or three nights a week to start.
  • Keep the shade, sunglasses and daily sunscreen habit. No paper here compared sun protection with a cream, so that is our view rather than a result.
  • Be sceptical of any before-and-after taken at a different angle, in different light, or fifteen minutes apart.

The gains measured here are real and modest. Whether that is worth the price is a judgement we would rather leave with you.

Frequently asked questions

Do eye creams actually do anything, or is it all marketing?
They do something measurable, but narrower than the advertising suggests. Studies around the eye most consistently report changes in hydration, water loss and surface texture, with wrinkle measurements shifting over weeks rather than days. In one manufacturer-run study of 33 women, hydration and surface profile improved within fifteen minutes, which tells you how much of the instant effect is water rather than structural change.
Why do my dark circles look worse than my friend's when we sleep the same amount?
Because under-eye darkness is not one thing. Reviews split it into pigmented, vascular, structural and mixed types, and note the cause is often hard to pin down in a given person. A randomised study of 181 people also found that one night of total sleep deprivation made no measurable difference to periorbital darkness.
Vitamin A or peptides for the eye area?
Both families have published work behind them, and neither has been compared head to head in a way that would let us pick for you. The retinaldehyde evidence is a three-month trial in 40 Korean women with no placebo arm; the peptide evidence is a 56-day trial of a single engineered molecule run by the company that makes it. Neither tested a product we sell.
Will an eye product get rid of a hollow under my eye?
No. If the shadow comes from volume loss or the shape of the bone underneath, a topical is working on the wrong layer, and the review evidence points to fillers or fat grafting for that cause. Worth knowing that a separate review of 42 tear trough papers found the evidence there is mostly low-level retrospective work with no agreed technique.
How long should I use an eye product before deciding it is not working?
Eight to twelve weeks of consistent use is a reasonable window, and that is our rule of thumb rather than a figure from a trial. The longest studies here ran 12 weeks and three months, so nobody assessed anyone beyond that. Anything you notice on day one or two is hydration, and no study here followed what happens after people stop.
Back to blog

Keep reading