Skin sensitivity is in the trial data, it is now published, and you are not imagining it.
In TRIUMPH-1, the pivotal phase 3 trial published in the New England Journal of Medicine on 29 September 2026, abnormal skin sensation was reported by 12.5% of people on the 12 mg dose against 0.9% on placebo. That is roughly fourteen times the placebo rate, in 2,339 people.1
It tracks dose: 5.1%, 12.3% and 12.5% at 4, 9 and 12 mg. In the knee osteoarthritis trial it reached 20.9% at the top dose.2
The phase 2 trial, which is where this was first noticed, recorded it in 7% of participants against 1% on placebo and added the detail that explains why people doubt themselves: none of the events came with anything visible on the skin.3 Your skin looks completely normal and feels like sandpaper, or burns under a shirt seam, or cannot tolerate a shower at the temperature you have used for years.
Lilly describes these events as generally mild to moderate, mostly resolving during treatment, with most participants continuing the drug.2
Raised resting heart rate is in the same programme, and dry mouth and sulfur burps are class effects. All four are documented. The one nobody should ignore is the heart rate.
Is it allodynia?
Probably not in the strict sense. The distinction matters if you are going to describe this to a doctor.
Allodynia is pain from something that should not hurt at all: a bedsheet, a light breeze.
Hyperesthesia is ordinary sensation turned up too high. Touch registers as more intense than it should.
Dysesthesia is sensation that is unpleasant or wrong in quality: burning, tingling, crawling, sandpaper.
The phase 2 trial used hyperesthesia and skin sensitivity. The phase 3 trials use dysesthesia, which is the term now in the published record. What people describe on forums, sandpaper skin, burning, tingling, itching that has nothing to scratch, is usually dysesthesia or hyperesthesia rather than true allodynia, though the line blurs in practice.
Say "abnormal skin sensation" and describe what it actually feels like. That gets you further than a label a clinician may disagree with.
How common is it, and does the dose matter?
Both answered, and the dose relationship is less tidy than the headline figures suggest.
| Trial | Population | 4 mg | 9 mg | 12 mg | Placebo |
|---|---|---|---|---|---|
| TRIUMPH-4, knee osteoarthritis | Obesity, knee OA | not studied | 8.8% | 20.9% | 0.7% |
| TRIUMPH-1, 2,339 people, published | Obesity, no diabetes | 5.1% | 12.3% | 12.5% | 0.9% |
| TRIUMPH-3, cardiovascular | Severe obesity, CVD | not studied | 6.4% | 6.4% | not stated |
| TRIUMPH-2, published | Obesity with diabetes | 4.5% to 7.3% across doses | 0.7% | ||
| TRANSCEND-T2D-1, published | Type 2 diabetes | 4.5% | 2.3% | 4.4% | 0.0% |
| Phase 2, 338 people | Obesity | 7% across all doses | 1% |
TRIUMPH-1, TRIUMPH-2 and TRANSCEND-T2D-1 are published trials.1,8,9 TRIUMPH-3 and TRIUMPH-4 are company announcements.2
Three things follow.
It rises sharply from 4 mg to 9 mg in TRIUMPH-1, 5.1% to 12.3%, and from 9 mg to 12 mg in the knee trial, 8.8% to 20.9%. It is flat from 9 mg to 12 mg in TRIUMPH-1, 12.3% against 12.5%, and flat across both doses in TRIUMPH-3. So dose matters, and the curve is not a straight line.
The rate varies by population, and the pattern is striking once the trials are lined up. The two trials in people with type 2 diabetes report the lowest rates: 2.3% to 4.5% in TRANSCEND-T2D-1 and 4.5% to 7.3% in TRIUMPH-2. The trials in people without diabetes report the highest: 12.5% in TRIUMPH-1 and 20.9% in the knee trial. The cardiovascular trial sits between them.
There is a second explanation for the same ordering, and it cannot be separated from the first. The trials line up in exactly the same order by weight lost: 28.7% and 28.3% in the two trials without diabetes, 22.6% in the cardiovascular trial, 20.8% and 16.8% in the two diabetes trials. Weight loss on these drugs is consistently smaller in people with diabetes, so "has diabetes" and "lost less weight" move together and no amount of looking at these five trials will tell them apart.
Either way, the practical reading is the same: the people most likely to notice this are those losing the most weight, which is most of this site's readers. We are making this observation by lining up separate trials rather than analysing within one, and populations differ in more than their diabetes status.
And across the trials as a whole, somewhere between one in sixteen and one in five people at the higher doses reported it.
What has not changed since phase 2: none of these events were severe or serious, and most people continued treatment.
Does this happen on semaglutide and tirzepatide?
Yes, and both carry it on the label. The rates are the interesting part.
Tirzepatide. The Zepbound label has a dysesthesia subsection: 0.2% at 5 mg, 0.2% at 10 mg and 0.4% at 15 mg, against 0.1% on placebo, pooled from its two weight-management trials.6
Semaglutide. The Wegovy label lists dysesthesia among its most common adverse reactions, those occurring in 5% or more of patients. In the trial of Wegovy tablets, 4.9% of treated patients reported dysesthesia against none on placebo, with preferred terms including sensitive skin, hyperesthesia, paresthesia, allodynia and skin burning sensation.7
One line in that label matters for anyone reading this page: the incidence of dysesthesia with Wegovy increased with increasing dosage and drug levels in the blood.7 The same relationship appears in the retatrutide trials, in a different drug, written into an approved label.
Lined up, the three drugs separate sharply.
| Drug | Top labelled or trial dose | Rate | Placebo |
|---|---|---|---|
| Tirzepatide | 15 mg | 0.4% | 0.1% |
| Semaglutide, tablet form | 25 mg | 4.9% | 0% |
| Retatrutide | 12 mg | 12.5% to 20.9% | 0.7% to 0.9% |
So this is a class effect, labelled for both approved drugs, and an order of magnitude more common with retatrutide than with tirzepatide.
That ordering also follows the amount of weight lost, and follows how many receptors each drug hits, and the three cannot be separated from one another with the data available.
Why does my skin look completely normal?
Because the effect is in the nerves, not the skin. The single most useful line comes from the phase 2 paper: none of these events were associated with overt skin findings.3
There is no rash to point at, no redness, no dryness that explains it. The sensation is neurological rather than dermatological, which is why moisturiser does nothing and why a partner looking at your arm sees nothing wrong.
It is also why people get dismissed. A symptom with no visible sign, reported by someone taking an unapproved compound bought online, is easy to wave away. The trial data is the answer to that, and it is worth having the citation ready.
Why forums and clinicians disagree about this
Ask in a subreddit and the answer is yes, everyone gets it. Ask a doctor unfamiliar with the compound and you may be told it is not a known effect.
Both responses have a basis. The forum answer is right that it is real and common, and wrong that it is universal, since the trials put it between one in sixteen and one in five at the higher doses. The clinical answer reflects the fact that retatrutide has no label, no approval and no package insert to consult. Until a few days ago there was no published phase 3 paper either.
The thing that settles it is the published trial: Jastreboff and colleagues, New England Journal of Medicine, 29 September 2026. Bringing that citation is more useful than bringing a forum thread, and it is recent enough that a clinician may not have seen it.
Why would a weight-loss drug do this to your skin?
Nobody knows. That answer has three candidate explanations rather than none.
The drug itself, by some direct mechanism. Retatrutide acts on three receptors, including the glucagon receptor, which the other drugs in this class do not touch. Whether any of that reaches peripheral nerves is unstudied. The trial recorded the effect without explaining it.
Rapid metabolic change. In diabetes care there is a recognised phenomenon with a name: treatment-induced neuropathy of diabetes, historically called insulin neuritis. Correcting long-standing high blood glucose too quickly can produce an acute, painful small-fibre neuropathy within weeks, and the largest series of it was published in Brain in 2015.5 It is counterintuitive, it is real, and the standard advice that follows from it is to improve glucose control gradually.
Nobody has looked at whether rapid weight loss on an incretin does anything comparable, and retatrutide trial participants were not people correcting years of untreated hyperglycaemia. So this is an analogy rather than an explanation. It is worth knowing because it establishes that fast metabolic improvement can itself provoke nerve symptoms, which is not an obvious idea.
Eating much less. Thiamine and B12 both matter for nerve function, and a sharp sustained drop in food intake is a plausible route to being short of either. That is testable with a blood test, unlike the other two, which is a reason to mention it to a doctor.
The dose relationship is the main reason to suspect a direct drug effect, though it is not clean across trials, and more drug also means more weight lost. A nutrient or glucose effect would not track the dose that neatly. But that is reasoning from a pattern, not a finding, and the paper offers no mechanism at all.
Is it even retatrutide?
The question this site exists to ask, and it is not rhetorical here.
The trials tell you what pharmaceutical retatrutide did to thousands of people under supervision. They tell you nothing about the vial on your counter. Retatrutide is approved nowhere, so every vial in circulation came from the grey market.
In our September capture, 336 listings across 105 sellers carry it, and 205 of those listings across 58 sellers use a code or abbreviation rather than the name, GLP-3 and its variants among them.4 Our decoder guide covers what those codes mean and how often they are wrong.
So two readers with identical symptoms can be in completely different situations: one experiencing a documented effect of the compound, the other reacting to something else in the vial. Nothing in the symptom itself distinguishes them.
What the trials establish is that this symptom is not evidence of a bad vial. It happens with the pharmaceutical product, at known rates, across six controlled trials.
Does retatrutide raise your resting heart rate?
Yes, dose-dependently, peaking at 24 weeks in the phase 2 trial and declining after that.3 A rise of a few beats is an expected class effect and is on two approved labels; one that climbs substantially, or comes with palpitations, breathlessness, chest discomfort or faintness, is a reason to be seen rather than reassured.
Our guide on resting heart rate on these drugs has the figures from each label, what the trials found, and how to read what your watch is telling you.
Why the dry mouth and the sulfur burps?
These two are the least interesting on your list, because both have ordinary explanations.
Dry mouth usually tracks fluid balance. Eating much less means drinking much less, since food carries a substantial share of daily water, and these drugs slow everything down. Our dehydration guide covers the path from there to something that matters.
Sulfur burps are eructation, which appears in the tirzepatide label's adverse reactions table at 4 to 5% by dose against 1% on placebo.6 Food sitting longer in the stomach is the plausible mechanism. It is unpleasant, not dangerous, and it eases at a steady dose.
Neither is a reason to stop. Both are reasons to drink more than feels necessary.
When to stop and be seen
Most of this list is tolerable. These are not.
Skin sensation that becomes genuinely painful, spreads, or comes with weakness or numbness, which points at something other than this. A resting heart rate that has risen substantially, or any palpitations, chest discomfort, breathlessness or faintness. Severe abdominal pain, especially going through to the back. Vomiting you cannot stop, or passing very little urine.
And the one people skip: if you are having surgery or any procedure with sedation, say what you are taking, whatever it is and wherever it came from.
Common questions
Is skin sensitivity a known side effect of retatrutide?
Yes, and it is now in a published phase 3 trial. TRIUMPH-1 reported it in 12.5% of people at 12 mg against 0.9% on placebo, in 2,339 participants. The knee osteoarthritis trial reported 20.9% at the same dose. None of these events were severe or serious and most participants continued treatment.
My doctor says this is not a known side effect. What do I say?
That it appears in the phase 3 trial published in the New England Journal of Medicine on 29 September 2026, at 12.5% against 0.9% on placebo in 2,339 people. Retatrutide has no label for anyone to consult, and the paper is days old, so a clinician may not have seen it.
Is it allodynia?
Usually not in the strict sense. Allodynia means pain from something that should not hurt at all. What most people describe is hyperesthesia, ordinary sensation turned up too high, or dysesthesia, sensation that feels wrong in quality. Describe the feeling rather than naming it.
Why does my skin look completely normal?
Because the effect is neurological rather than dermatological. The phase 2 paper specifically notes that none of these events came with visible skin findings, which is why creams do nothing and why it is easy to be dismissed.
Does this happen on semaglutide and tirzepatide too?
Yes, and both carry it on the label. Tirzepatide's label reports dysesthesia at 0.2% to 0.4% by dose against 0.1% on placebo. Semaglutide's lists it among adverse reactions occurring in 5% or more of patients, with 4.9% reported in the tablet trial against none on placebo. Retatrutide's rates are several times higher than either.
Does it happen less if you have type 2 diabetes?
Possibly. The two trials in people with type 2 diabetes report the lowest rates, 2.3% to 4.5% and 4.5% to 7.3%, while the trials in people without diabetes report 12.5% and 20.9%. That comes from lining up five separate trials rather than from an analysis within one, so treat it as a pattern worth knowing rather than an established finding.
Does it go away?
Lilly reports that the majority of these events resolved during treatment and most participants continued the drug. The published data does not give a typical duration. Dose appears to matter, though not uniformly across trials.
Does retatrutide raise your heart rate?
Yes, dose-dependently, up to 24 weeks in the phase 2 trial, declining after that. A few beats is expected. A substantial rise, or palpitations, breathlessness or chest discomfort, is a reason to stop and be seen.
Why do I have sulfur burps?
Eructation is a labelled effect of this drug class, at 4 to 5% on tirzepatide against 1% on placebo, and the plausible mechanism is food moving more slowly through the stomach. It eases once the dose stops moving.
Why is my mouth so dry?
Usually fluid balance. Eating much less means drinking much less, because food carries a substantial share of daily water intake.
Does this mean my vial is fake?
No. The symptom appears in the trial of pharmaceutical retatrutide at a known rate, so it is not evidence either way about what you are holding. In our capture, 336 listings across 105 sellers carry retatrutide and most use a code rather than the name, which is a separate problem from this one.
Could it be a vitamin deficiency?
Possible and worth testing, since thiamine and B12 both matter for nerve function and intake drops sharply on these drugs. It is the only one of the plausible explanations you can check with a blood test, which is a reason to raise it.
Sources
- Jastreboff AM, Kaplan LM, Davies MJ, et al. Retatrutide, a Triple Hormone Receptor Agonist, for Treatment of Obesity. New England Journal of Medicine, published online 29 September 2026. doi:10.1056/NEJMoa2604169. NCT05929066.The pivotal phase 3 trial: 2,339 adults with obesity or overweight and at least one weight-related condition, without diabetes, over 80 weeks. Dysesthesia reported in 5.1%, 12.3% and 12.5% of participants at 4, 9 and 12 mg against 0.9% on placebo.
- Eli Lilly and Company. TRIUMPH-1 topline results, 21 May 2026, and subsequent programme announcements for TRIUMPH-3 and TRIUMPH-4.Company releases. Source for the statement that events of dysesthesia were generally mild to moderate, that the majority resolved during treatment, and that most participants continued taking retatrutide. Also the source for the TRIUMPH-4 figures of 8.8% at 9 mg and 20.9% at 12 mg against 0.7% on placebo, and the TRIUMPH-3 figure of 6.4% in both retatrutide arms. TRIUMPH-3 and TRIUMPH-4 remain topline announcements rather than published trials.
- Jastreboff AM, Kaplan LM, Frías JP, et al. Triple-Hormone-Receptor Agonist Retatrutide for Obesity: A Phase 2 Trial. New England Journal of Medicine 2023;389:514-526. doi:10.1056/NEJMoa2301972.The phase 2 trial, 338 adults, 48 weeks. Cutaneous hyperesthesia and skin sensitivity reported in 7% of retatrutide participants against 1% on placebo; none were severe or serious, none were associated with overt skin findings, and none led to discontinuation. Also the source for the heart rate pattern, which increased dose-dependently to 24 weeks and declined thereafter.
- Peptide Decoding vendor price capture, 29 September 2026. peptidedecoding.com/prices.Our own data, counting one priced product at one seller as a listing. Retatrutide appears in 336 listings across 105 sellers, of which 205 listings across 58 sellers use a code or abbreviation rather than the compound name.
- Gibbons CH, Freeman R. Treatment-induced neuropathy of diabetes: an acute, iatrogenic complication of diabetes. Brain 2015;138(1):43-52. doi:10.1093/brain/awu307.Cited as an established precedent that rapid metabolic improvement can provoke nerve symptoms, not as an explanation of the retatrutide finding, which has not been studied.
- Zepbound (tirzepatide) prescribing information, Adverse Reactions. pi.lilly.com.The approved label. Eructation appears in the adverse reactions table at 4 to 5% by dose against 1% on placebo.
- Wegovy (semaglutide) prescribing information. accessdata.fda.gov.The approved label. Dysesthesia is listed among the most common adverse reactions, those at 5% or higher. A dedicated subsection reports dysesthesia adverse reactions in 4.9% of patients treated with Wegovy tablets and none on placebo, with preferred terms including sensitive skin, hyperesthesia, paresthesia, allodynia and skin burning sensation. The label states that the incidence of dysesthesia with Wegovy increased with increasing dosage and drug levels in the blood.
- Retatrutide in adults with obesity and type 2 diabetes (TRIUMPH-2). The Lancet, published online 29 September 2026, presented at the 62nd Annual Meeting of the European Association for the Study of Diabetes.Phase 3 trial. Dysesthesia reported in 4.5% to 7.3% of retatrutide-treated participants against 0.7% on placebo, most events mild to moderate. Weight loss up to 20.8%. Figures read via trade reporting of the publication and presentation.
- Bajaj HS, et al. Efficacy and safety of retatrutide in adults with type 2 diabetes and inadequate glycaemic control with diet and exercise alone (TRANSCEND-T2D-1). The Lancet, 2026. NCT06354660.Phase 3 trial, 537 adults, 40 weeks. Dysesthesia occurred in 4.5%, 2.3% and 4.4% of participants at 4, 9 and 12 mg against 0.0% on placebo, generally mild with a majority resolving during treatment. The published paper reports dysesthesia in 15 participants, 4% overall. A1C reduction up to 2.0% and weight loss up to 16.8%.
Citing this page. Peptide Decoding. Retatrutide Skin Sensitivity: Is It Normal? Figures from TRIUMPH-1, NEJM 2026, and the Wegovy and Zepbound labels. https://peptidedecoding.com/guides/retatrutide-skin-sensitivity

