Your patient is 46, seven months into Mounjaro, several dress sizes down and miserable about her face.
You can see the mid-face has lost projection. What she talks about, though, is her skin. It is drier. It sits differently. The texture has changed in a way she cannot name, and you cannot quite see it under clinic lights.
Filler answers part of that. The rest tends to happen in a layer most of us were never taught to look for.
Facial imaging in five patients on semaglutide found average reductions of 41.8 per cent in the temporal fat pad and 69.9 per cent in the cheek. Whole-body scanning of a separate group showed a 9.2 per cent reduction in fat mass.
It is worth being clear that five patients is a series rather than a study, so the precise multiple is not one to quote. The gap is still hard to ignore.
Your patients worked this out before the literature did. By 2024, searches for “Ozempic face” were running close to searches for Ozempic side effects generally.
None of this is unique to the drugs. After bariatric surgery took mean BMI from 38 to 27, midface volume loss was recorded in 86 per cent of patients. Rapid weight loss has always done this. The medications have made it ordinary.
Every facial fat compartment you treat has a superficial and a deep half. Almost all our training concerns the deep one.
The superficial half is dermal white adipose tissue, known in humans as skin-associated adipose tissue. It went unnoticed for years partly because in mice it is divided from deeper fat by a muscle layer people do not have, so nobody looked for a boundary that was not there.
It rewards your attention. Its cells sit between adipocyte and fibroblast and shift between the two, and it expresses more collagen than the fat below it. When Staphylococcus aureus invades, the layer expands and its maturing cells produce antimicrobial peptides, making it part of the skin’s immune defence rather than a passive store. And it answers to calories fast: three weeks of restriction in mice thinned both it and the epidermis above.
This is not a fuel depot with a good address. It sits exactly where you work, and it is rarely idle.
Facial volume is thought to reduce through this superficial layer first, and through the deep compartments afterwards. Laxity follows mechanically: the stiffness of skin and fat together depends largely on the fibrous scaffold around each adipocyte, which rapid weight loss thins.
Read that back against your consultation. The dryness, the changed texture.
Those were not preliminaries to the real complaint. They were the first sign, months ahead of the hollowing that brought her in.
A 2025 review in Dermatology and Therapy describes the same pattern from the other end: volume loss most pronounced in the upper cheeks, lips and chin, deepening the nasolabial folds and marionette lines, alongside barrier compromise and dryness as fatty acid availability falls. The dryness is not incidental to the volume loss. It is part of the same event.
Caveolae are flask-shaped pits in the cell membrane, fifty to a hundred nanometres across, where a cell gathers its receptors. A protein called Caveolin-1 holds them in shape. What follows is a proposed mechanism, not a proven one.
Writing in the Journal of Cosmetic Dermatology this year, Dr Ilja Kruglikov proposes that these medications reach facial tissue through that route, and that it could be narrowed deliberately. “One encouraging possibility for such a preventive intervention,” he writes, “can be the local modulation of CAV1 in the facial adipose tissue during the GLP-1RA treatment course.”
Put plainly: not repairing a face afterwards, but making it a quieter target while the drug works elsewhere.
He proposes three ways to do it: drugs, heat and very high frequency ultrasound. Higher frequencies load the cell mechanically and loosen the cytoskeleton caveolae anchor to, an effect he describes as stronger when several arrive together. That is what LDM® Triple does, running up to three of its four frequencies (1, 3, 10 and 19 MHz), switching every one to ten milliseconds, faster than tissue can respond, so it reads them as one signal.
That paper is a mechanistic argument, and closes by asking for the work to be done. He is also the managing partner of the manufacturer, which he declares. It is a well-reasoned hypothesis with a clear route to testing. No published study has yet measured it in a patient on these drugs.
The most relevant clinical evidence is a 2024 case series in Clinical, Cosmetic and Investigational Dermatology: twenty women, eight to ten sessions, with sonoelastography measuring the tissue rather than a photograph judging it. Dermis elasticity improved from 2.96 to 4.08 and skin-associated adipose tissue from 2.91 to 3.91, both significant, no adverse effects. Twenty women without a control arm gives you direction rather than size, but the improvement is real.
What would settle it is a controlled study measuring facial dermal fat in patients on GLP-1, treated and untreated. Worth saying to patients, and worth watching for.
Ask the question. Around 1.6 million adults in Britain used weight loss medication in the year to early 2025, with another 3.3 million considering it. It belongs on your history form.
Start earlier than feels natural. If the prevention argument holds, the window is while they are on treatment. Month seven is when they tend to arrive. Month one is when it counts.
Hold the cadence. Frequency rather than intensity moves this: twice weekly, three days apart. Results keep building afterwards, so book the review at three to four months.
The interesting question is no longer how to refill a face after the event. It is whether we can help it keep more of itself in the first place.
References