Eye creams: what the evidence actually supports

The honest answer first. An eye cream can help with the look of dryness and fine lines. It cannot change the shape of your face. Most disappointment with eye creams comes from buying one to fix something a cream was never able to touch.

Why the eye area gets its own product

A review of eye cream ingredients describes the area around the eye as highly susceptible to damage from outside factors, and as one of the first places to show the signs of ageing (Hamie et al., 2024). That is the practical case for treating it differently. Eye products are generally made at lower strength and with fewer irritating ingredients than face products, and on this part of the face gentleness matters more than strength.

One thing we are going to skip. You will see a claim everywhere that eyelid skin is a specific number of times thinner than the rest of your face. We could not find that figure in the research we read, so we are not repeating it.

Dark circles are at least three separate things

One research team compared people with dark circles against people without, using multispectral imaging, videocapillaroscopy and ultrasound. In the group with dark circles they measured a higher melanin index, meaning more pigment in the skin, a higher haemoglobin index alongside early signs of more dilated or more numerous capillaries, meaning blood vessels showing through, and thinner skin under the eye (Mac-Mary et al., 2019).

Two caveats before you lean on that. The main measurements were made on Caucasian skin, with smaller exploratory work on African and Far East Asian skin, so it is not the last word for everyone. And four of the seven authors worked for Walgreens Boots Alliance, a cosmetics company. It looks like careful work, but it is not independent.

The useful point still stands. Two people can have under-eyes that look the same for different reasons, which is the likeliest explanation for why a product does something for your friend and nothing for you.

The same team also tracked dark circles over seven days and found they varied within a person, within a single day and between days (Mac-Mary et al., 2019). If yours look worse this morning, that is normal and measurable, not a product failing.

The fourth cause, which is not skin at all

Shadow. Try this at the mirror. Shine a phone light upwards at your face from below your chin. If the darkness largely disappears, you are most likely looking at a shadow cast by the hollow where the lower lid meets the cheek. It is a rough check rather than a diagnosis, but no cream changes the shape casting a shadow.

What an eye cream cannot do

This section matters more than the ingredient list.

• It cannot fill a hollow. A systematic review of 50 studies describes infraorbital hollowing as the result of structural volume loss, and places volume loss and poor lower lid support with doctors performing injectable or surgical procedures (Beer et al., 2026). A cosmetic is nowhere near that category.
• It cannot move a fat pad. In that same review, surgery is what is described for herniated fat. If your under-eye bag is a fat pad, no cream relocates it.
• It cannot remove a shadow. If the darkness comes from that hollow, you are looking at light and geometry, and there is nothing for a topical product to act on.
• It cannot erase pigment. That review places topical products with superficial discolouration only (Beer et al., 2026). At best a cosmetic softens how pigment looks. Anyone promising to erase it is overselling.
• It does not fix tiredness or poor sleep. It might make you feel better about how you look. It does not change what is causing it.

The ingredients, and what the research actually says

Caffeine

Caffeine is one of the ingredients that turns up in a 30-year review of topical treatments for the look of under-eye discolouration (Woolhiser et al., 2025). Read that review carefully before you get excited about anything in it. The authors screened 339 studies and included 22, and every single included study reported a statistically significant result, while only 7 of the 22 were randomised. When nothing in a field ever reports a miss, that is a reason for caution about the field rather than confidence in it. The authors' own advice is sensible: work out the cause first, then choose.

We went looking for good evidence that caffeine reduces the appearance of morning puffiness specifically, and did not find it in the reviews above. Caffeine is popular and it is well tolerated. That is a different thing from proven.

The percentage on the label also tells you less than you would think. Researchers compared four purpose-built 2% caffeine creams against three commercial eye creams, measuring release and passage across Strat-M, a synthetic membrane used to stand in for skin. All four purpose-built creams released more caffeine than the commercial products, and the one containing two penetration enhancers gave significantly more permeation than the commercial products (Reyes et al., 2023). That was laboratory equipment, not living skin and not a trial in people. It tells you formulation matters. It tells you nothing about what a given product will do on your face.

Peptides

Peptides are comfortable to use and well tolerated, which counts for something on skin that reacts easily. What they change is another matter. A meta-analysis of 19 randomised controlled trials and 1,341 participants found improvements in hydration and brightness, particularly with oral formulations, and only a modest pooled effect on wrinkles that was largely driven by peptides taken by mouth rather than applied to skin (Nukaly et al., 2026). Effects on elasticity and skin density were inconsistent. None of it was specific to the eye area. Peptide marketing runs a long way ahead of peptide data.

So a product like Medik8 Eyelift Peptides is a fair choice if you want something hydrating and comfortable that you can use daily without your eyes complaining. It is not a fair choice if you are buying it to remove lines.

Retinoids

Eye-area retinoid products are usually made at lower strength, or in a slower-release form, on purpose. One research team put the reason plainly: retinoids can irritate, and that irritation is what limits their use around the eye (Kaufman et al., 2022). Their own 12-week study of an eye-specific retinoid reported improvement in the appearance of lines, texture, under-eye darkness, puffiness and dryness. Then read the design. It was open label with no control group, only 16 people used the retinoid cream on its own, every result was measured against each person's own starting point, and the product tested is not one sold here. A study with no control group cannot separate the active from the base cream carrying it, or from twelve weeks simply passing.

Be realistic about retinol more broadly. A systematic review of nine randomised, double-blind, vehicle-controlled trials of over-the-counter retinol found that four showed no statistically significant difference from the plain base cream. The other five suggested at most a mild effect on fine lines, and all five carried major methodological flaws, with the reviewer concluding that the positive trials serve advertising better than decisions (Spierings et al., 2021). Those trials were on facial skin, not the eye area. Nobody should buy a retinol eye cream expecting a documented result.

If you still want to try a retinoid near your eyes, do it the cautious way. Choose the lowest strength in an eye-specific range, such as Crystal Retinal Ceramide Eye 3, use it at night, start with a couple of nights a week, and keep it away from the lash line and the inner corner. Use sunscreen every morning while you are using it. Do not substitute your face-strength retinoid. Stop if it stings. If you are pregnant or breastfeeding, ask your doctor or pharmacist before using any retinoid product.

Everything else on the label

Vitamins C and E, niacinamide, ceramides and hyaluronic acid all appear in the eye cream literature, with reported benefits for hydration and for the look of lines and discolouration (Hamie et al., 2024). The review authors then name their own limitation, and it is a large one. There is a notable lack of clinical trials testing eye cream formulations on the skin around the eye. Most of the evidence comes from studies on facial skin generally, then assumed to carry across to the eye area. For most products on the shelf, that assumption has not been tested.

The boring parts, which are the best supported

• Sun protection and sunglasses. Damage from outside factors is what the eye area is most exposed to (Hamie et al., 2024), and squinting in bright light adds movement to skin that already moves all day. Under a New Zealand sun that is relevant most of the year. A dedicated eye product such as Advanced Day Eye Protect SPF 30 is easier to apply close to the lash line than a face sunscreen, though it is a 15ml eye product and not a replacement for sunscreen on the rest of your face. Sunglasses and a hat do a share of the work for free.
• Gentleness and consistency. Pat product on with a ring finger rather than dragging it. Keep it off the lash line and out of the eye itself. A simple moisturiser used every day is a reasonable baseline for most people, and one product used properly beats five used impatiently.
• One new thing at a time. Patch test a new eye product on the skin beside your eye for a few days before you use it closer in. If you are adding an active, add it on its own, so that if something reacts you know what caused it.
• Sleep and salt. These move morning puffiness more than anything in a jar.

When it stops being a skincare question

If darkness or puffiness around your eyes has changed noticeably, is on one side only, or came on suddenly, see a doctor rather than buying a cream. There are medical causes for swelling and for changes around the eyes, and a cosmetic is not a substitute for having them looked at. Persistent redness, itching, flaking or swelling of the eyelid is also a matter for a doctor or pharmacist, and adding more product is the wrong move. Stop any product that stings, and rinse with water if it gets into your eye. If you wear contact lenses and a product makes your eyes uncomfortable, stop using it.

Nothing described here treats, prevents or cures anything. All of it is about the appearance of the eye area. If you would like help working out which cause you are actually dealing with, come and see us in Parnell. Sometimes the honest answer is that a cream is not the right tool, and we would rather say so.