Eye bags, honestly.

Fluid under the eye is a plumbing problem: the tissue fills when the body holds water and when the head is level with the heart for hours, and drains once you are upright. In the only controlled study of its kind, faces photographed after 31 hours awake were rated as having more swollen eyes, more hanging eyelids and darker circles than the same faces after normal sleep, and the eye signs were among the largest effects (Sundelin 2013). A 60-woman study found poor sleepers had more signs of intrinsic aging and slower barrier recovery (Oyetakin-White 2015). Salt and alcohol retain water and dilate vessels; there is no trial of a low-salt evening on eye bags, and there does not need to be.

The protocol is dull and works: seven to eight hours, the head raised on a second pillow (or the bed head lifted), a salty or alcoholic dinner treated as a known cost, and water rather than a diuretic. If the morning bag is gone by noon, this is the treatment; if it is not, it was never fluid.

When the nasal lining swells with hay fever or a dust or pet allergy, the veins that drain the lower lid back up, producing a dark, puffy under-eye — the allergic shiner — often with itching, watering and a crease under the eye from rubbing (Cleveland Clinic). The treatment is the allergy's: a non-sedating antihistamine (cetirizine, loratadine, fexofenadine), a steroid nasal spray in season, saline rinses, and the hard part, not rubbing, which thickens and darkens the skin over years. Eyelid eczema and contact allergy to a mascara or cream do the same and answer to stopping the product. Cosmetic treatment of a puffy eye whose cause is an untreated allergy is money spent on the wrong organ.

Thin, crepey lower-lid skin exaggerates every bag, and the one topical that thickens it is a retinoid: the randomised tretinoin trials and their meta-analysis show more dermal collagen, fewer fine wrinkles and better texture over six to twelve months (meta-analysis). A retinoid eye cream improved rated lines by 33% and puffiness by 55% at 12 weeks in an uncontrolled study — read the puffiness number as skin firming, not fat loss (JDD, 2022). Apply a pea-sized amount to the orbital bone, not the lid margin, two or three nights a week at first; the lower lid irritates faster than any other skin. Our wrinkles guide covers the retinoid ladder.

The thin lower-lid skin photoages fastest and is where most people stop their sunscreen. The Nambour trial's 24% reduction in measured skin aging with daily use applies here as everywhere (Hughes 2013), and pigmented dark circles darken with every unprotected summer. Sunglasses with UV-blocking lenses protect the skin and stop the squinting that etches crow's feet. A mineral sunscreen stings less near the eye. None of this shrinks a bag; all of it decides how the skin over one looks at 55.

Cold narrows blood vessels and slows fluid leaking into tissue, so a cold compress, a chilled spoon or a roller from the fridge visibly reduces a fluid bag for an hour or two. A review of edema management in aesthetic practice notes that ice has not been proven to reduce swelling in trials yet is recommended by most practitioners for exactly this purpose (JCAD review). Tea bags add caffeine and tannins to the cold; cucumber adds water. Use it for the morning of the wedding; expect nothing tomorrow.

Caffeine constricts vessels and pulls water out of tissue when it gets through the skin, which under the eye it partly does. A four-week placebo-controlled study of a 3% caffeine gel measured about a 17% reduction in under-eye fluid and a small elasticity gain; an eye cream with caffeine, a xyloside and blueberry extract improved rated dark circles and puffiness over 12 weeks in a manufacturer trial (JAAD, 2009); a 2024 "multicorrective" eye cream did the same without a control group (2024 study); a 2025 study of a dark-circle formulation reported objective improvement (2025 study). Real, small, temporary, and confined to the fluid type. A caffeine product on a fat bag does nothing at any price.

An eye cream is a face cream in a smaller jar, and the evidence for "eye" formulations specifically is thin: a 2024 review of popular eye-cream ingredients finds retinoids, vitamin C, niacinamide and caffeine supported by small studies, mostly manufacturer-run, and peptides and "de-puffing" complexes supported by almost nothing (IJWD, 2024). The instant "de-puffing" is caffeine and hydration — see those rows. The one study that looked at eye appearance and habits found dark circles tracked with poor sleep (60-woman study). Spend on a retinoid, sunscreen and sleep; keep the €150 cream for the pleasure of it. Our 30s guide covers the eye-cream question in full.

Gentle outward strokes from the inner corner along the orbital rim move pooled fluid toward the lymph nodes in front of the ear, and a cold roller does the same with vasoconstriction thrown in. The effect on a fluid bag is real for an hour or two and there is no trial of it around the eye; the general facial-massage trials measure millimetres of temporary contour change. The hazard is friction: rubbing the lower lid, whether with a roller or a knuckle, is a documented route to pigmented dark circles and thickened skin. Feather-light, cold, outward, and never on a fat bag, which it cannot move.

The backstage trick works for the same reason and the same hour as a cold spoon: phenylephrine constricts vessels, and when it wears off they dilate past baseline, so the puffiness returns worse. Many formulations also contain hydrocortisone, which on the thinnest skin of the body causes thinning, visible vessels and, absorbed through it, cortisol effects; the mineral-oil base clogs the lid glands. No study supports the practice and dermatologists advise against it (GoodRx review). Cold and caffeine do the same job without the rebound.

Filler treats the hollow, and for the hollow it now has the evidence: in a randomised, evaluator-blinded, no-treatment-controlled multicentre trial of a soft hyaluronic-acid gel made for the under-eye, 87.4% of treated patients were responders at three months, 63.5% at twelve, and 80.3% after a single retreatment at eighteen — the basis of the first approval specifically for infraorbital hollows (pivotal trial). A meta-analysis of 31 studies in 2,556 patients found a pooled satisfaction rate of 91% (2023 meta-analysis).

The same meta-analysis is the safety warning: swelling in 19.2%, bruising in 18.4%, contour lumps in 5.3%, and the blue-grey Tyndall tint in about 1%; malar edema — persistent puffiness on the cheekbone — follows in up to 11% in periocular reviews and can appear years later (periocular filler review; delayed complications). Gel placed over a real fat bag makes the bag bigger. This is a procedure for an injector who does eyes weekly, with a soft low-water gel, a cannula, deep placement on bone, tiny volumes, and hyaluronidase in the room. Our filler guide covers the zone in detail.

Hyaluronic-acid gel under the eye persists for years and attracts water, which is why a proportion of "eye bags" in clinics are old filler. Hyaluronidase dissolves it: in a series of 20 patients with eyelid edema, a single injection resolved the swelling effectively and rapidly with no relevant adverse effects (20-patient series), and reviews set out doses and the fact that denser gels need more, and repeated, treatment (review). It dissolves the filler you liked along with the filler you did not, and a small number of people react to the enzyme, so it is a reset rather than a fine adjustment. If your bags began after an aesthetic treatment, this drawer is the first question to ask.

The "jelly roll" is the pretarsal orbicularis muscle bunching under the lashes when you smile, and a tiny dose of toxin — one to two units — flattens it and widens the eye, as the original study showed (Flynn & Carruthers). The same muscle pumps lymph out of the lower lid and holds the lid against the eye, and the dose-finding study found that at 8 units every subject developed bothersome lower-lid edema and incomplete closure (dose-finding study). Over-treatment produces exactly the puffy, slack lower lid people came in to fix, and it lasts until the toxin wears off. Useful in a very light hand for a muscle roll in a young face; contraindicated for a bag of fat or fluid, which it worsens.

Injecting dilute tetracycline or doxycycline into a festoon inflames and scars its fluid pockets shut, the same principle as sclerosing a vein. It is a niche tool for edema-dominant festoons with roughly a 50% success rate in the published experience, uncomfortable for 10–20 minutes, and repeated once or twice (Cleveland Clinic; 2019 systematic review). For severe festoons, one surgical group combines serial tetracycline injections with a pinch excision of skin and a canthopexy to support the lid (preliminary series). Off-label, oculoplastic-surgeon territory, and the least invasive thing that has ever changed a festoon.

Platelet-rich plasma injected or needled into the lower lid delivers growth factors that thicken thin skin and, in the vascular and pigmented dark-circle types, reduce the tint. In a controlled, blinded randomised trial, PRP, stromal vascular fraction and a Q-switched Nd:YAG laser each improved dark circles and wrinkles after nanofat grafting (randomised trial), and a single-session study reported fair-to-good improvement in 80% at three months with about a 47% fall in measured melanin (study). No trial compares it with placebo, and it does nothing for fat or fluid. A reasonable add-on for thin, dark lower-lid skin; the regenerative guide covers the systems.

Fat harvested from the abdomen and placed in the tear trough restores volume permanently — and in a comparison of lower blepharoplasty with fat repositioning against onlay segmental fat grafting, grafting produced the better long-term result, confirmed in patients who had one technique on each eye (ASJ, 2021). Nanofat, an emulsified version, is injected superficially to thicken and brighten thin lower-lid skin, and a series combining it with transconjunctival blepharoplasty in young patients with bags and hollows reports high satisfaction (series). Lumps and irregularities under thin skin are the risk, retention is unpredictable, and it cannot be dissolved — which is why it is a surgeon's tool, usually combined with removing or moving the fat above.

Deoxycholic acid destroys any cell membrane it touches; it is authorised for subcutaneous fat under the chin (EU authorisation) and for nothing near the eye. An eye bag is not subcutaneous fat: it is orbital fat behind a membrane, next to the muscle that closes the eye and a millimetre of skin, and an injection that reached it would also reach those. Clinics offering "fat-dissolving" for eye bags are offering an unlicensed injection in the most dangerous place on the face. The treatment for orbital fat is surgery, which removes or moves it under direct vision.