A reader sent me four photos of her under-eyes taken in different lights and asked which eye cream she should try next. She had been through eight in eighteen months. Caffeine, peptides, retinol, vitamin K, vitamin C, an “Asian” eye gel that promised pigmentation correction, a 110 dollar serum with a glass dropper, and a drugstore basic that her dermatologist had recommended as a placebo control.
The photos showed a shadow that deepened in overhead light and almost disappeared in light from below. Her under-eye skin was thin but otherwise unremarkable. No discolouration when stretched. No swelling. Just shadow.
I told her none of the eye creams were going to work. The shadow was structural, not pigmentary, and no topical was going to fill a depression that lives below the dermis.
This is the conversation I find myself having most often about the eye area, and it took me a humbling year to learn that the answer for many readers is not a product. It is sometimes a procedure, and it is sometimes nothing at all.
What you are actually looking at
There are three things people describe as “dark circles” and they have different mechanisms. Telling them apart is the question that determines whether any topical has a chance.
The first is pigmentary. Melanin deposition in the dermis or post-inflammatory hyperpigmentation around the orbital rim. This is the most common form in Fitzpatrick IV-VI skin. The pigment is visible from any angle, does not change with lighting, and does not vanish when you stretch the skin gently outward.
The second is vascular. The thin under-eye skin allows the underlying venous plexus to show through, producing a blue or purple appearance. The colour deepens when blood pooling increases, which is why poor sleep, allergies, and dehydration intensify it. Stretching the skin laterally usually makes the colour fade because you are flattening the vascular pattern.
The third is structural. The tear trough is an anatomical depression that exists in everyone but deepens with age as fat pads shift, the orbicularis retaining ligament tethers, and the cheek descends. The dark appearance is not pigment or vasculature. It is a shadow cast into a depression by overhead light. It vanishes in front-lit photos and deepens in selfies taken from below.
The reader who sent me four photos was almost entirely category three. Her shadow was a shadow.
What the studies actually show
Vrcek and colleagues published a thorough review on periorbital hyperpigmentation in 2016 that I still send people (Vrcek et al., J Clin Aesthet Dermatol 2016, PMID: 26962392). They proposed a structural classification that maps the visible problem onto its actual mechanism: pigmentary, vascular, structural, or mixed.
The treatment implications follow the classification.
For pigmentary cases, topical actives have a real role. Hydroquinone, azelaic acid, retinoids, vitamin C, and tranexamic acid have data for pigment lightening, though the periorbital skin is more sensitive than the cheek and the irritation threshold is lower. Roh and colleagues studied retinol around the eye and found measurable lightening over 24 weeks at concentrations under 0.1 percent (Roh et al., Dermatol Surg 2009, PMID: 19663871).
For vascular cases, topicals do almost nothing in any rigorous study I have found. Caffeine has temporary vasoconstrictive effects that might reduce the visible blue tint for an hour or two. Vitamin K creams, despite the marketing, do not have human data showing meaningful effects on vascular under-eye appearance. The cleanest intervention is sleep, sodium restriction, and antihistamines if the underlying driver is allergic.
For structural cases, no topical has produced clinically meaningful change in the literature I have read. The depression is a depression. You cannot fill it from the outside. The 110 dollar serum did not work because nothing in that category was going to work.
The honest treatment options for structural tear troughs are hyaluronic acid filler, fat grafting, blepharoplasty, or accepting the shadow. The filler studies, when read carefully, are not as encouraging as the before-and-afters suggest. Goldberg and colleagues found high patient satisfaction but also reported a non-trivial rate of Tyndall effect (blue cast from superficial filler) and prolonged oedema in the tear trough region (Goldberg et al., Dermatol Surg 2014, PMID: 24447298). The procedure is also harder than most dermal filler work and requires a practitioner with specific training.
The mirror test
Before spending money on anything, run the three-light test.
Stand in front of a mirror in even, front-lit light. A ring light at eye level works. Look at your under-eyes. The shadow should be at its minimum here.
Tilt your head down so overhead light hits the orbital rim. The shadow should deepen. This is your structural component.
Stretch the skin laterally with your fingers, pulling outward toward your temples. If the colour fades, your component is mostly vascular. If it stays the same, you have pigmentary or structural causes.
Take a flash photo. Pigmentary shadows persist in flash. Vascular and structural shadows often vanish.
I have run this test with about a dozen people in person and the results are consistent. Most adults under 35 with dark circle complaints have a mixed vascular-structural picture with a small pigmentary component. Most adults over 45 have a more structural-dominant picture. Pigmentary-dominant cases are more common in higher Fitzpatrick types and in people with allergic histories.
The implication is uncomfortable but useful. If your test reveals structural dominance, you have three real choices: filler, surgery, or acceptance. You are not in the topical category and you are not going to be talked into it by a serum.
The contrarian part
Tear trough filler has become the default recommendation for under-eye shadows in cosmetic medicine, and I have watched friends and readers spend 800 to 1500 dollars on it with results ranging from satisfying to disastrous.
The problems are mechanical. The tear trough region has very thin skin, a constrained anatomical space, and a hyaluronic acid filler that is hydrophilic. Filler placed too superficially produces a visible blue cast through the skin (the Tyndall effect mentioned earlier). Filler placed in an area with poor lymphatic drainage produces sustained swelling that can last months. Filler that migrates produces lumps that are visible under the eye for the life of the product, which with the most common Restylane and Juvederm formulations is one to two years.
The reversal option exists. Hyaluronidase dissolves HA filler. But hyaluronidase carries its own complications, including allergic reactions and over-dissolution of native HA, and the recovery from a bad filler outcome can take six to twelve months.
I am not arguing against filler. I am arguing that the decision to do it should be made with a practitioner who has the specific training, after a careful look at the anatomy, and with realistic expectations about what one round of filler will and will not do. If you cannot find a practitioner who has done at least a few hundred tear trough procedures, the right answer might be to wait.
The acceptance option is also real. The shadow that you see in the bathroom mirror is amplified by overhead lighting. It is not what other people see at conversational distance under normal indoor lighting. The cognitive distortion of staring at your own face in a magnifying mirror under unflattering light is part of the problem, and stepping back from that mirror is a free intervention with a measurable effect on perceived severity.
What I would tell my past self
Run the three-light test before you spend any money on the eye area. The diagnosis determines the entire treatment plan.
If the test points to pigmentary, the cheap interventions are azelaic acid, low-dose retinol, vitamin C, and sunscreen. The eye creams are mostly the same actives in smaller bottles at higher prices. You can use a regular face product carefully, kept off the lid margin, and get the same effect for less money.
If the test points to vascular, save your money. Sleep, hydration, allergy management, and a cool compress in the morning do more than any eye cream. Caffeine eye creams give a temporary boost that fades within hours.
If the test points to structural, decide whether you can accept the shadow. If you cannot, talk to a practitioner with specific tear-trough training and ask about Restylane Eyelight or fat grafting. The decision is not topical-versus-procedure. It is procedure-versus-acceptance.
Stop reading eye cream reviews. The reviewers are reporting placebo, packaging, and texture preferences. They are not measuring the variable you actually care about.
Frequently asked
Does retinol thin under-eye skin?
The opposite. Properly used, low-concentration retinoids thicken the dermis modestly over months and can improve fine lines. The risk is irritation, not thinning. Start at very low concentration around the eye and build slowly.
Are eye creams ever worth buying?
Sometimes. For pigmentary cases, a well-formulated eye product with retinol, vitamin C, or azelaic acid can be useful, especially if you have a sensitivity that prevents using stronger face products near the eye. The 100 dollar versions are not better than the 30 dollar ones.
Will filler last longer in the tear trough than in the lips?
Generally yes. The literature suggests tear trough filler can last 12 to 24 months in many patients, compared to 6 to 12 months in the lips, because the area moves less and metabolises slower.
Why do my dark circles look worse on some days?
Sleep, sodium intake, allergy load, hydration status, and the angle of available light. None of these change the underlying anatomy. They change how visible the existing pattern is.
Should I use a peptide cream around my eyes?
Peptide data is weaker than retinol or vitamin C data. The cleanest study I have read on Matrixyl (Robinson et al., Int J Cosmet Sci 2005) showed measurable but small improvements in fine lines over months. If you want to use one, fine. Do not expect it to outperform the cheaper actives.
Related Elelaf tools
Eye cream decision | Dark circle decoder | Retinol strength | Tretinoin decoder
Sources
- Vrcek I, Ozgur O, Nakra T. Infraorbital dark circles: a review of the pathogenesis, evaluation and treatment. J Clin Aesthet Dermatol 2016;9(1):65-72. PMID: 26962392.
- Roh MR, Chung KY. Infraorbital dark circles: definition, causes, and treatment options. Dermatol Surg 2009;35(8):1163-1171. PMID: 19663871.
- Goldberg RA, et al. Filler-associated complications in the tear trough region. Dermatol Surg 2014;40(2):209-216. PMID: 24447298.
- Ranu H, et al. Periorbital hyperpigmentation in Asians: an epidemiologic study and a proposed classification. Dermatol Surg 2011;37(9):1297-1303. PMID: 21492307.