.avif)
Facial changes after major weight loss predate GLP-1 drugs. New studies document measurable volume loss in GLP-1 users—but cannot yet tell us whether the medication itself contributes beyond the weight lost.
“Ozempic face” sounds like a new drug complication. It is a memorable phrase, easy to visualise and even easier to market.
But the face it describes is not new.
People who lost substantial amounts of weight after bariatric surgery were described as having hollow cheeks, deeper nasolabial folds, loose neck skin and an older-looking face years before semaglutide became a cultural phenomenon. What GLP-1 drugs changed was not the existence of facial deflation after weight loss. They changed the number of people able to achieve large weight losses without surgery.
That difference in scale matters.
In the pivotal STEP 1 trial, adults receiving weekly semaglutide 2.4mg lost an average of 14.9 percent of their body weight over 68 weeks, compared with 2.4 percent in the placebo group. Half of those receiving semaglutide lost at least 15 percent. Weight loss of that magnitude was no longer confined mainly to people undergoing bariatric surgery.
Ozempic did not invent facial volume loss after substantial weight loss. It made substantial weight loss possible for far more people—and gave a familiar consequence a new name.
The unresolved question is whether that is the whole explanation. Does the face change because the body has lost a considerable amount of fat, or might GLP-1 drugs also affect facial fat and skin biology directly?
The honest scientific answer is: we do not yet know.
First, “Ozempic face” is not a diagnosis
Ozempic is a brand of semaglutide authorised for type 2 diabetes. Semaglutide is also marketed as Wegovy for weight management, while tirzepatide—sold under brands including Mounjaro and Zepbound—acts on both GIP and GLP-1 receptors.
Yet “Ozempic face” is routinely used for facial changes associated with all of them. One brand name has become shorthand for several medications, different indications and a wide range of weight-loss experiences.
There is no accepted diagnostic definition. The term usually refers to some combination of:
- temple and cheek hollowing;
- flatter midface contours;
- more visible tear troughs;
- deeper folds around the nose and mouth;
- reduced lip fullness;
- jowling or loose skin along the jawline and neck.
These changes can create the appearance of accelerated ageing. That does not automatically mean that the drug has made the skin biologically age faster. A face can look older because its supporting volume has changed, even if no medication-specific injury to collagen, elastin or regenerative cells has occurred.
What substantial weight loss does to the face
Facial fat is not one uniform layer. It is organised into superficial and deep compartments that contribute to contour, tissue movement and structural support. When those compartments lose volume, the skin and retaining structures above them do not always retract at the same rate.
The result can resemble ageing: hollows become more visible, folds deepen and unsupported tissue appears to descend.
Age, baseline facial volume, the amount of weight lost, previous obesity, sun exposure, smoking and individual skin elasticity may all influence the final appearance. The rate of loss is widely assumed to matter too, but direct studies comparing rapid and gradual weight loss are surprisingly lacking.
Most importantly, this pattern was documented before the current GLP-1 era.
A prospective bariatric-surgery cohort found that the mean perceived facial age increased from 40.8 years before surgery to 43.7 years one year afterwards. Patients over 40 and those with greater weight loss were more susceptible to an older appearance. Other studies of patients after massive weight loss described midface deflation, deeper nasolabial folds and laxity in the lower face and neck.
These studies are not perfect: many were small, relied on photographs or assessed surgical patients seeking facial rejuvenation. But together they establish a crucial fact. The visual phenotype now called “Ozempic face” can occur after substantial weight loss without Ozempic.
What has actually been measured in GLP-1 users
In 2025, Sharma and colleagues published one of the first objective imaging studies of the phenomenon. They identified 20 patients who had CT or MRI imaging before and after receiving a GLP-1 receptor agonist.
During an average treatment period of approximately 321 days, participants lost an average of 11kg. Median total midfacial volume decreased by 9 percent. The superficial compartments lost 11 percent and the deep compartments 7 percent.
The headline finding was striking—but so was the relationship underneath it: superficial facial-volume loss increased with total body-weight loss. The authors estimated approximately 7 percent midfacial-volume loss for every 10kg of weight lost.
This confirms that measurable facial deflation occurs in GLP-1 users. It does not establish that GLP-1 therapy has a special toxic effect on the face. There was no comparison group of people who lost the same 10kg without medication.
An earlier conference poster analysed serial imaging from only five semaglutide users and reported very large reductions in superficial temporal and cheek fat. Those figures have since been repeated widely, but a five-person conference abstract should not be treated as a population estimate.
The Hany study: does the method of weight loss matter?
A 2024 study by Hany and colleagues offers an unusually useful comparison, although it did not examine GLP-1 drugs or facial skin directly.
The researchers analysed 80 skin biopsies from 77 patients undergoing body-contouring surgery after massive weight loss. Forty biopsies came from patients who had lost weight following bariatric surgery; 40 came from patients who had achieved massive weight loss without surgery. Samples were taken from the abdomen and breast.
Measured collagen content was virtually identical between the surgical and nonsurgical groups. Abdominal skin from the surgical group contained fewer elastic fibres, but this difference was not found in breast skin.
The authors noted that collagen content in both groups appeared lower than normal-skin values reported elsewhere. However, the study did not include its own healthy control group. The two weight-loss groups also differed in age, starting BMI, current BMI, percentage of excess weight lost and time since intervention. The precise dietary methods used by the nonsurgical group were unknown.
So this study cannot tell us what happens in facial skin after semaglutide. What it does show is that altered skin structure after massive weight loss is not automatically evidence of a drug-specific effect—or even an effect unique to bariatric surgery.
Could GLP-1 drugs also act directly on facial tissue?
Possibly. This is where the question becomes scientifically interesting—and where certainty falls away.
Adipose tissue is not simply passive padding. Adipocytes and adipose-derived stem cells participate in signalling, inflammation, extracellular-matrix maintenance and tissue repair. GLP-1 receptors and related signalling pathways can influence adipocyte metabolism and differentiation under experimental conditions.
In cultured human adipocytes, GLP-1 has been reported to inhibit fat-cell differentiation and lipogenesis while stimulating lipolysis. Other laboratory studies found that liraglutide affected the proliferation and differentiation of human adipose-derived stem cells.
These findings make a direct medication effect biologically plausible. But cultured cells exposed to a selected concentration of native GLP-1 or liraglutide are not the same as facial tissue in a person using semaglutide for a year. Experimental outcomes also vary according to the compound, dose, cell source, metabolic environment and timing of exposure.
The ten-biopsy study
In 2026, Firsowicz and colleagues compared abdominal adipose-tissue samples from people taking GLP-1 receptor agonists with samples from untreated controls. Across ten samples, the GLP-1 group had approximately four times fewer counted adipose-derived stem cells: 11.0 versus 44.8 cells per mm². Fibroblast populations did not differ significantly.
This is an important early signal. It is also a very small, cross-sectional study of abdominal—not facial—fat. From such a study, we cannot determine whether the medication caused the difference, whether the groups differed beforehand, or how much weight loss, diabetes, obesity severity and other metabolic factors contributed.
It generates a hypothesis. It does not prove that GLP-1 drugs deplete regenerative cells in the face.
Evidence pointing in the other direction
The cellular story is not even consistently negative.
A 2024 study of patients with obesity and type 2 diabetes reported that six months of semaglutide restored aspects of adipose-progenitor function, including their ability to develop into metabolically active white and beige adipocytes. Other experimental research describes context-dependent anti-inflammatory, anti-apoptotic or metabolically beneficial effects on adipose and mesenchymal cells.
These findings are not proof that semaglutide protects facial tissue. They do demonstrate why a simple chain—GLP-1 drug, stem-cell depletion, accelerated facial ageing—is premature.
The drug can reduce total fat mass while improving some features of dysfunctional adipose tissue. Those processes are not mutually exclusive.
What would prove a medication-specific effect?
The missing study is relatively easy to describe and difficult to conduct.
Researchers would need to compare people of similar age, sex, baseline BMI, facial anatomy and metabolic health who lose approximately the same percentage of body weight over the same period through different routes:
- GLP-1 or dual-incretin medication;
- bariatric surgery;
- diet and lifestyle treatment without a GLP-1 drug.
Participants would need facial imaging before, during and after weight loss, ideally combined with objective assessments of skin thickness, elasticity and tissue composition. Stable-weight follow-up would help distinguish temporary deflation from persistent change. Facial tissue studies would be more relevant than extrapolating from abdominal biopsies.
No published study currently does all of this.
Until one does, the evidence cannot cleanly divide the visible result into “weight-loss effect” and “drug effect.”
So, is Ozempic actually the cause?
In one sense, yes. GLP-1 drugs enable substantial weight loss, and substantial weight loss can reduce facial fat. The medication therefore sits at the beginning of the causal chain for many patients.
But that is different from saying Ozempic uniquely damages the face.
The evidence currently supports three levels of conclusion:
That distinction matters because a marketable term can quietly turn association into causation. “Ozempic face” makes the drug sound like a facial disease. The available evidence more strongly supports a familiar consequence of substantial weight loss now appearing in a much larger and more visible population.
What patients should take away
Visible facial change does not mean that an effective metabolic treatment has harmed a patient overall. Nor should someone stop or alter a prescribed GLP-1 medication because of aesthetic concerns without discussing it with the clinician managing their treatment.
For anyone considering aesthetic treatment, the stage of weight loss matters. When weight and facial volume are still changing, repeated attempts to replace every new hollow may lead to overtreatment. A conservative assessment can separate volume loss, skin laxity and surface-quality concerns rather than treating “Ozempic face” as one uniform condition.
Most of all, patients deserve accurate language.
The facial changes are real. The association with GLP-1 treatment is real. But the claim that Ozempic itself uniquely accelerates facial ageing goes beyond what current evidence can prove.
Editorial note: “Ozempic face” is used in this article because it is the term patients commonly encounter. It is not a recognised diagnosis and should not be used to imply that semaglutide is proven to cause medication-specific facial ageing.