GLP-1s and Skin: Rash, Itching, and Injection-Site Reactions Evidence [2026]

GLP-1 drugs like Ozempic, Wegovy, Mounjaro, and Zepbound are taken as a shot under the skin, so it makes sense that some of their side effects show up on the skin too. Most of these are mild and local: a red, itchy, or sore spot where the needle went in. This guide walks through what the actual trial data and post-market reports say about rashes, itching, and injection-site reactions, where the evidence is solid, and where it is thin.

What "skin reactions" actually means with GLP-1s

When people talk about GLP-1 skin problems, they are usually mixing together three different things. Keeping them separate matters because the evidence, the cause, and the fix are different for each.

Injection-site reactions (ISRs). These happen right at the spot where you injected. Think redness, a small raised bump, itching, mild swelling, bruising, or a sore feeling. They are local. They stay near the needle entry point and usually fade within a few days.

Generalized skin reactions. These show up away from the injection site, anywhere on the body. Examples are a widespread rash, hives (urticaria), or general itching (pruritus). These are less common and harder to pin on the drug, since a lot of things cause a rash.

True allergic and hypersensitivity reactions. These are the rare, more serious end. They include angioedema (swelling of the lips, face, throat) and anaphylaxis. These are medical emergencies and are not the same as a normal itchy injection spot.

The distinction is not academic. A red itchy bump at your injection site almost never means you are allergic to the drug. Swelling of your throat means you might be, and that is an emergency.

There is also a fourth category that gets lumped in but does not belong: skin changes from the weight loss itself. Loose skin, "Ozempic face," and a more drawn appearance are real, but they come from losing fat and volume, not from any direct skin toxicity of the drug. This guide is about reactions, not cosmetic changes from weight loss. If your concern is facial volume loss, that is a different topic with different evidence.

How common are injection-site reactions, really?

This is where the best evidence sits, and the honest answer is: more common than the drug labels alone suggest, but still not common, and rarely serious.

The strongest single piece of evidence is a 2026 meta-analysis by Taj and colleagues in Diabetes, Obesity and Metabolism. They pooled 14 randomized controlled trials covering 4,861 patients with 396 injection-site reaction events. People on a GLP-1 had a risk ratio of 3.55 (95% CI, 2.35-5.36) for injection-site reactions compared with the comparison groups. In plain terms, GLP-1 users were about three and a half times more likely to get a reaction at the shot site than people not on the drug.

That sounds dramatic. But the baseline rate is low, so even tripling a small number keeps the absolute risk modest. The same analysis found that generalized skin events away from the injection site were "infrequent and not significantly elevated." So the signal is real and specific to the injection site, not a sign these drugs broadly inflame the skin.

The individual drug labels report lower numbers than the meta-analysis, partly because trials count reactions conservatively and partly because the drugs differ:

Drug	Injection-site reaction rate	Placebo/comparator rate	Source
Semaglutide (Wegovy)	1.4%	1.0%	FDA label
Tirzepatide (Mounjaro/Zepbound)	3.2%	0.4%	FDA label
Dulaglutide (Trulicity)	0.5%	0.0%	FDA label

So the picture from controlled trials: roughly 1 to 3 patients in 100 report a local reaction, versus well under 1 in 100 on placebo. That is the most reliable framing. The "3.5x" risk ratio and the "1 to 3 percent" absolute rate are both true and describe the same thing from different angles.

Why the numbers jump around

You will see wildly different figures online, from "1.4%" to "10.4%." They are not all wrong; they measure different things. Older trials of some drugs reported higher rates. Pooled antibody-positive subgroups (more on that below) run higher. And passive reporting databases, which catch anything anyone reports, look different again. When you see a scary-sounding percentage, check whether it is from a controlled trial, a subgroup, or a voluntary report.

How strong is the evidence, honestly?

It is worth grading the evidence rather than treating every number as equal. The injection-site reaction signal is the strongest part of this picture. It comes from a meta-analysis of randomized controlled trials, which is the highest tier of evidence, and the finding is consistent and biologically sensible. You can trust that GLP-1s raise local reaction risk somewhat.

The drug-to-drug comparisons are weaker. Most of those rest on FAERS, a passive database that cannot tell you how many people took each drug, cannot confirm the drug caused the event, and over-counts whatever is prescribed most. Treat ranking tables of "which drug is worst for skin" as hypothesis-generating, not settled.

The rare, severe reactions, blistering eruptions and the like, sit on the weakest tier: isolated case reports. They prove the events can happen but say nothing reliable about how often. So the overall message is layered: solid evidence that local reactions are real and modest, soft evidence on which drug is worst, and barely-there evidence on the scary tail.

Does the specific drug matter?

Somewhat, and the differences track with how the molecule and its formulation behave.

Tirzepatide (Mounjaro, Zepbound) shows a higher trial rate of injection-site reactions than semaglutide, and there is a clear mechanistic reason. In the FDA's pooled trial data, hypersensitivity reactions occurred in 4.1% of tirzepatide patients who developed anti-tirzepatide antibodies versus 3.0% of those who did not. The split is even sharper for injection-site reactions specifically: they occurred in 4.6% of patients with anti-drug antibodies versus just 0.7% of those without. That is a strong association. The body making antibodies against the drug appears to drive a lot of the local reaction risk.

Real-world reporting tells a slightly different story, and it is worth not over-reading. A 2026 analysis by Fat and colleagues in the Journal of Drugs in Dermatology mined the FDA's adverse event database (FAERS) from 2018 to 2024. Of 129,330 total adverse events reported for GLP-1s, 5.75% were skin-related. By drug, the share of reports that were cutaneous was:

Drug	Share of reports that were skin-related (FAERS, 2018-2024)
Semaglutide	8.16%
Liraglutide	7.66%
Exenatide	5.74%
Dulaglutide	3.75%

It is tempting to read this as "semaglutide is the worst for skin." Don't. FAERS is a voluntary, passive database. Semaglutide has been prescribed to far more people than the others, so it generates more reports of everything. The authors of that study said plainly that their findings "should be used to signal awareness and not establish causation." Dulaglutide had the lowest share, which the authors noted could reflect prescription patterns rather than a true safety edge.

There is one genuinely reassuring number in that same dataset. Compared with DPP-4 inhibitors (a different diabetes drug class), GLP-1s had a proportional reporting ratio of 0.27 (95% CI, 0.257-0.284) for skin events, meaning GLP-1s were actually reported less often for skin problems than that comparison class.

The Taj meta-analysis also flagged that exenatide and dulaglutide stood out specifically for injection-site bleeding or bruising in real-world data, with disproportionately high reporting. That is plausibly a needle and formulation issue more than a deep biological one.

Why does the injection site react?

Several things stack up, and most are mechanical, not allergic.

The needle itself. Any subcutaneous injection nicks tiny blood vessels and disturbs tissue. That alone explains a lot of the bruising, minor bleeding, and short-lived redness. This is the most common cause and has nothing to do with the drug being "wrong" for you.

Immune recognition of the drug. GLP-1 medications are large peptide molecules. The immune system can make antibodies against them, and as the tirzepatide antibody data show, those antibodies line up with higher injection-site reaction rates. This is a delayed, local immune response, not a classic allergy.

Temperature and technique. Injecting cold medication straight from the fridge, reusing or dulling needles, injecting too shallow, or hitting the same spot repeatedly all raise the odds of a reaction. A review of GLP-1 dermatologic effects noted that biopsies of reactive sites can show perivascular eosinophilic infiltration, a sign of a localized inflammatory response rather than a whole-body allergy.

Compounded products. Compounded semaglutide and tirzepatide are not FDA-approved and vary in purity, salt form, and preservatives. Some skin reactions reported with compounded versions may trace back to the formulation rather than the active drug. If you switched to a compounded product and reactions started, the product is a reasonable suspect. For more on how compounded and brand-name products differ, see the comparison linked at the end.

The order of these causes matters for how you respond. Needle trauma and technique problems cover the large majority of what people experience, and they are fixable without changing drugs. Antibody-driven reactions are less common and usually settle as the body adjusts. Formulation issues are specific to a product. True drug allergy is rare. So before you assume your body is rejecting the medication, it is worth ruling out the boring, mechanical explanations first, because that is where most reactions actually live.

What about lipohypertrophy and lumps?

People who inject in the same spot over and over can develop lipohypertrophy: firm, rubbery lumps of thickened fat under the skin. This is well known from decades of insulin injection and is driven by repeated trauma to one area, not by an allergy or a drug defect. It matters for two reasons. The lumps look and feel alarming, and injecting into them can change how the drug absorbs. The fix is the same as for most local reactions: rotate sites systematically so no single area takes repeated hits. Lumps usually soften over months once you stop injecting into them.

Timeline: when reactions show up and how long they last

Knowing the usual timing helps you tell a normal reaction from one worth worrying about.

Most local reactions appear within minutes to hours of the injection and peak in the first day. Simple redness and stinging often fade within a few hours. An itchy bump or patch of redness typically clears within three to five days. Bruising follows the usual bruise timeline, darkening then yellowing over a week or two.

Reactions that are delayed, that keep getting worse after the first 48 hours, or that spread well beyond the injection site break the normal pattern and deserve attention. A reaction that appears for the first time many weeks into treatment, after dozens of uneventful injections, is also worth mentioning to your prescriber, since that timing can point to a developing immune response or a change in product or technique.

There is a reassuring trend baked into the trial data: for most people, local reactions become less frequent over time, not more. Early injections are when reactions are most likely, and the skin tends to settle as treatment continues. That is the opposite of what you would expect from a true escalating allergy, and it is one more reason not to quit over an early itchy spot.

Timing	What it usually means
Minutes to a few hours, local	Normal needle and formulation response
Itch or redness clearing in 3-5 days	Typical injection-site reaction
Worsening after 48 hours or spreading	Worth a call to your prescriber
First appearance weeks into treatment	Mention it; could be immune or product-related
Throat/face swelling, breathing trouble, anytime	Emergency, stop the drug

How to tell the reaction types apart

When a reaction happens, the most useful question is "where is it, and is it spreading?" Location and spread sort the minor from the serious better than any single symptom.

A reaction confined to a circle around the needle entry point, even if it itches or looks angry, is almost always a benign injection-site reaction. A rash or hives appearing on parts of the body you did not inject suggests a more systemic response and moves you up a tier of concern. Any swelling of the lips, tongue, face, or throat, or any trouble breathing, is the top tier and means stop and get emergency help, regardless of how mild it started.

Itching alone, without spread, is reassuring. Itching plus a spreading rash plus any swelling is not. The pattern, not the itch itself, is what tells you which bucket you are in.

The rare and serious end: what to actually watch for

Most skin reactions are a nuisance. A small number are not. Here is the honest grading.

Hives and widespread rash (uncommon). Urticaria, generalized itching, and rash do happen and show up in case reports and the FAERS data. A 2024 case report documented a clear cutaneous hypersensitivity reaction to dulaglutide that resolved after stopping the drug. These are real but uncommon, and the evidence is mostly case-level rather than from large trials.

Angioedema and anaphylaxis (rare, serious). Both the Wegovy and Mounjaro labels carry warnings that serious hypersensitivity reactions, including anaphylaxis and angioedema, have been reported. These are genuine emergencies. Swelling of the lips, tongue, face, or throat, trouble breathing, or a sudden full-body rash with dizziness means stop the drug and get emergency care. The drug should not be restarted.

Very rare blistering and severe drug eruptions (weak, case-level evidence). Scattered case reports tie GLP-1s to bullous pemphigoid, acute generalized exanthematous pustulosis (AGEP), and other severe eruptions. These reports exist, but they are isolated, the evidence that the drug caused them is weak, and they are far too rare to estimate a rate. They are worth knowing about and not worth losing sleep over.

A reasonable rule of thumb: local and itchy is almost always minor; spreading, blistering, or swelling away from the injection site deserves a doctor's eyes; throat or breathing involvement is a 911 situation.

How to reduce injection-site reactions

The practical fixes are simple and well-supported by injection-technique guidance, even if no large trial has tested each one head to head. Treat these as low-risk, common-sense steps.

Problem	What to do
Cold-injection sting and redness	Let the pen sit at room temperature for ~20-30 minutes before injecting
Repeated reactions at one spot	Rotate sites: abdomen, thigh, upper arm; stay at least an inch from the last spot
Bruising and bleeding	Don't rub the site after; press gently; avoid injecting near visible veins
Itching after the shot	A cool compress or over-the-counter hydrocortisone or an oral antihistamine usually settles it
Stinging on injection	Inject slowly; make sure the needle is fully under the skin, not too shallow
Reactions started after switching products	If you moved to a compounded version, raise it with your prescriber

Do not pre-medicate routinely with antihistamines without talking to your prescriber, and do not stop a working medication over a mild local reaction. Most reactions improve as treatment continues. If they don't, switching to a different GLP-1 is a reasonable conversation, since the reaction is often specific to one molecule or formulation.

Who should be more cautious

Most people can manage skin reactions with simple technique changes. A few groups warrant a closer look.

Anyone with a prior anaphylactic or angioedema reaction to a GLP-1 should not restart that drug, and often should avoid the class. People with a history of severe drug allergies, mast cell disorders, or chronic urticaria should flag that before starting. And anyone using a compounded product who develops new skin reactions has a clear reason to question the product's quality.

For the vast majority, though, an itchy red spot is a minor, manageable, expected part of injecting a peptide under the skin, and not a reason to quit a drug that is helping.

Frequently Asked Questions

Is an itchy red bump at my injection site an allergic reaction?

Almost never. Local itching, redness, and a small bump are the most common type of injection-site reaction and reflect needle trauma and a mild local immune response, not a true allergy. True allergic reactions are systemic: hives spreading across the body, facial or throat swelling, or trouble breathing. If the reaction stays at the injection site and fades within a few days, it is a normal local reaction. If it spreads or involves swelling away from the site, contact a doctor.

Which GLP-1 is least likely to cause skin reactions?

In controlled trials, semaglutide (Wegovy) reported the lowest injection-site reaction rate at about 1.4%, versus roughly 3.2% for tirzepatide. But the real-world data is muddier: in the FAERS database, semaglutide had the highest share of skin-related reports, likely because it is prescribed to far more people. The most honest answer is that all GLP-1s carry a low risk, the differences are small, and an individual's reaction is often specific to one product rather than predictable from the class.

Why do I bruise so much at the injection site?

Subcutaneous injections nick tiny blood vessels, which causes bruising and minor bleeding. This is mechanical, not a drug allergy. Exenatide and dulaglutide showed disproportionately high reporting for injection-site bleeding in real-world data, which points to needle and formulation factors. To reduce it, avoid rubbing the site after injecting, press gently instead, steer clear of visible veins, and rotate injection sites so no single spot gets repeatedly traumatized.

Should I stop my GLP-1 if I get a rash?

Not for a mild, local rash. Most injection-site reactions improve as treatment continues and respond to simple measures like a cool compress, an antihistamine, or rotating sites. You should stop and seek care urgently if you develop swelling of the lips, tongue, face, or throat, trouble breathing, or a sudden widespread rash with dizziness, since those signal a serious hypersensitivity reaction. For a stubborn but non-emergency rash, talk to your prescriber about switching to a different GLP-1.

Can compounded semaglutide or tirzepatide cause more skin reactions?

Possibly. Compounded products are not FDA-approved and vary in purity, salt form, and added ingredients, any of which could trigger skin reactions that the brand-name version would not. There is no large trial proving compounded versions are worse, so this is reasoning from how the products differ rather than from hard outcome data. If your reactions started after switching to a compounded product, the formulation is a reasonable suspect, and it is worth raising with your prescriber.

Citations

This article is for general information only and is not medical advice. Talk to a licensed clinician about your own situation, and seek emergency care for any signs of a serious allergic reaction.