Zepbound Side Effects: The Complete List and How Common Each One Is

Most side effect lists for Zepbound give you six bullet points and no numbers. That does not help when you are lying awake wondering whether what you are feeling is normal.

This page works differently. It walks through everything that appears in the FDA prescribing information for Zepbound, and puts the actual trial percentage next to each item. Where something is widely talked about but is not on the label, that is flagged as such, along with what kind of evidence sits behind it.

Two things to get straight first.

Zepbound and Mounjaro are the same drug. The active ingredient in both is tirzepatide, made by Eli Lilly. In the United States, Zepbound is the version approved for weight management and for obstructive sleep apnea, and Mounjaro is the version approved for type 2 diabetes and for reducing cardiovascular risk [1][11]. If you are reading British coverage, be aware that in the UK “Mounjaro” is also the weight-loss brand, so UK articles about Mounjaro for weight loss do not line up with the US Zepbound label.

Tirzepatide is also not the same drug as semaglutide (Ozempic, Wegovy). It works on two gut hormone receptors instead of one, it comes from a different company, and it has its own label. Numbers from one do not transfer to the other.

Nothing here is medical advice. It is a map of what the label and the published trials say, so you can have a better conversation with a healthcare provider.

What are the most common Zepbound side effects?

The Zepbound Highlights section names twelve reactions that occurred in at least 5% of people. That is a longer list than most GLP-1 drugs carry, and it includes several items, such as hair loss and hypersensitivity reactions, that do not appear on the Mounjaro label at all [1].

Here is the main adverse reaction table from the two 72-week placebo-controlled weight-reduction trials, covering 2,519 adults who received tirzepatide [1]:

Side effect	Placebo	5 mg	10 mg	15 mg
Nausea	8%	25%	29%	28%
Diarrhea	8%	19%	21%	23%
Constipation	5%	17%	14%	11%
Vomiting	2%	8%	11%	13%
Abdominal pain	5%	9%	9%	10%
Dyspepsia (indigestion)	4%	9%	9%	10%
Injection site reactions	2%	6%	8%	8%
Fatigue	3%	5%	6%	7%
Hypersensitivity reactions	3%	5%	5%	5%
Eructation (burping)	1%	4%	5%	5%
Hair loss	1%	5%	4%	5%
Gastroesophageal reflux (GERD)	2%	4%	4%	5%
Dizziness	2%	4%	5%	4%
Hypotension (low blood pressure)	0%	1%	1%	2%

A few things worth noticing in that table.

Constipation runs backwards. It is the one side effect that was more common at the lowest maintenance dose than the highest. Nobody has a settled explanation for that, and the label does not offer one.

More than half of people had at least one gut side effect. The label gives the combined figure: 56% at each of the three doses, versus 30% on placebo [1].

Most of it happens early. The label states plainly that the majority of nausea, vomiting and diarrhea occurred during dose escalation and decreased over time [1].

Some terms in the table are broader than they look. “Injection site reactions” covers bruising, redness, itching, pain and rash. “Fatigue” covers asthenia, lethargy and malaise. “Abdominal pain” covers discomfort, tenderness and upper and lower pain. “Hypotension” covers orthostatic hypotension and blood pressure decreases [1].

How many people stop Zepbound because of side effects?

In the two main trials, 4.8% of people on 5 mg, 6.3% on 10 mg and 6.7% on 15 mg stopped permanently because of an adverse reaction, against 3.4% on placebo. Gut side effects specifically accounted for 1.9%, 3.3% and 4.3%, versus 0.5% on placebo [1].

In a third trial that used a diet-and-exercise lead-in, 10% stopped versus 2% on placebo [1].

The label adds one detail that is easy to miss: most of the people who stopped did so during the first few months [1]. That is consistent with the pattern in the tables, where the gut symptoms cluster around dose escalation.

For a point of comparison, the only head-to-head randomized trial against semaglutide (SURMOUNT-5, 751 adults, 72 weeks) found 6.1% stopping on tirzepatide versus 8.0% on semaglutide for any adverse event, and 2.7% versus 5.6% for gut events specifically [3].

What are the serious warnings on the Zepbound label?

Zepbound carries a boxed warning plus nine other Warnings and Precautions. Here they are in label order, with what the evidence actually shows [1].

Boxed warning: risk of thyroid C-cell tumors

In a two-year rat study, tirzepatide caused dose-dependent thyroid C-cell adenomas and carcinomas. The label states it is unknown whether Zepbound causes these tumors, including medullary thyroid carcinoma, in humans, because the human relevance of the rodent finding has not been determined [1]. In a six-month study in a different animal model (rasH2 transgenic mice), tirzepatide was not tumorigenic [1].

The counseling instruction is not routine screening. The label says routine monitoring of serum calcitonin or thyroid ultrasound is of uncertain value, and tells patients to report a lump or swelling in the neck, hoarseness, trouble swallowing or shortness of breath [1].

In the four-year SURPASS-CVOT trial of 13,299 adults, medullary thyroid cancer was reported in 2 people who received tirzepatide over a median of four years; one of those tumors was positive for a RET proto-oncogene mutation, which is the inherited driver of that cancer [10].

Severe gastrointestinal adverse reactions (5.2)

Severe gut reactions were reported in 1.7%, 2.5% and 3.1% at 5, 10 and 15 mg, versus 1% on placebo. This section was updated in February 2026. The current wording states that Zepbound is not recommended in people with severe gastroparesis [1].

Acute kidney injury due to volume depletion (5.3)

This is really a dehydration warning. Acute kidney injury occurred in 0.5% on Zepbound versus 0.2% on placebo in the weight trials. The label says the majority of reported events happened in people who had nausea, vomiting or diarrhea leading to volume depletion, and the postmarketing section lists acute renal failure sometimes requiring hemodialysis [1].

No dose adjustment is recommended for kidney impairment, but the label says to monitor kidney function in people reporting severe reactions that could cause dehydration [1].

Acute gallbladder disease (5.4)

Gallstones occurred in 1.1% versus 1% on placebo, gallbladder inflammation in 0.7% versus 0.2%, and gallbladder removal in 0.2% versus 0%. The label says acute gallbladder events were associated with weight reduction [1]. This is one of the areas where randomized evidence for tirzepatide is consistently positive rather than null.

Acute pancreatitis (5.5)

The warning covers fatal and non-fatal hemorrhagic or necrotizing pancreatitis. In the weight-management trials, adjudication-confirmed acute pancreatitis occurred in 0.2% on Zepbound and 0.2% on placebo, so this is a warning that rests mainly on postmarketing reports and class experience rather than a trial signal [1].

The number was higher in the sleep apnea program: an exposure-adjusted rate of 0.84 per 100 patient-years on Zepbound versus 0 on placebo, on two adjudicated cases [4].

The symptom to know is severe, persistent abdominal pain, sometimes radiating to the back, with or without vomiting [1].

Hypersensitivity reactions (5.6)

Anaphylaxis and angioedema have been reported after approval. Severe hypersensitivity occurred in 0.1% on Zepbound and none on placebo. Reactions were roughly twice as common in people who developed antibodies to tirzepatide: 6.2% versus 3% [1].

Hypoglycemia (5.7)

On its own, tirzepatide lowers blood sugar in a glucose-dependent way and rarely causes lows. Risk climbs when it is combined with insulin or a sulfonylurea: in the trial that enrolled people with type 2 diabetes, blood glucose under 54 mg/dL occurred in 10.3% of those on a sulfonylurea versus 2.1% of those not on one [1].

Zepbound’s warning goes further than Mounjaro’s by noting that hypoglycemia has also been reported in adults without type 2 diabetes [1]. Any change to an insulin or sulfonylurea dose is a prescriber decision.

Diabetic retinopathy complications (5.8)

This section was updated in August 2026 on the basis of SURPASS-CVOT [1][2]. It applies to people who have type 2 diabetes and a history of retinopathy, and it reflects a long-standing observation that rapid improvement in blood sugar control can be followed by temporary worsening of retinopathy. The label also notes that Zepbound has not been studied in people with proliferative diabetic retinopathy or diabetic macular edema [1].

Pulmonary aspiration during general anesthesia or deep sedation (5.9)

There are postmarketing reports of people on GLP-1 drugs having food left in the stomach at the time of a procedure despite following fasting instructions, with aspiration in some cases. The label instructs patients to tell providers about the drug before any planned procedure. It explicitly says the available data are insufficient to determine whether holding the drug would reduce the risk [1].

The best randomized evidence is a 2026 trial at two US centers comparing holding versus continuing one dose before elective upper endoscopy. It stopped early at an interim analysis of 60 patients: clinically significant residual stomach contents occurred in 3.1% of the hold group versus 25.0% of the continue group, with no aspiration or intubation in either arm. Among those who had a clear liquid diet the day before, none had significant residual contents regardless of whether the drug was held [6]. A large observational study in people undergoing upper endoscopy found no increase in aspiration (risk ratio 0.98) but roughly double the rate of aborted procedures [7].

Never share a KwikPen between patients (5.10)

Added in January 2026. A KwikPen must never be shared even if the needle is changed, because of the risk of transmitting bloodborne pathogens [1].

One warning that is no longer there

Zepbound used to carry a Suicidal Behavior and Ideation warning. The label’s revision history records it as removed in February 2026 [1], following the FDA’s January 13, 2026 request that manufacturers take the warning off GLP-1 labels after reviewing placebo-controlled trial data and a large database cohort [5]. Mounjaro never carried the warning.

A removal is a regulatory conclusion drawn from accumulated evidence. It is not a statement that no individual ever experiences a mood change, and new or worsening mood symptoms are still worth raising with a healthcare provider.

Who should not take Zepbound at all?

The label lists two absolute contraindications [1]:

1. A personal or family history of medullary thyroid carcinoma, or Multiple Endocrine Neoplasia syndrome type 2.
2. A known serious hypersensitivity reaction to tirzepatide or to any of the ingredients in the product.

Two other groups are addressed separately. The label says weight reduction offers no benefit in pregnancy and may cause fetal harm, and directs that Zepbound be stopped when pregnancy is recognized [1]. And the Limitations of Use section says coadministration with other tirzepatide-containing products, or with any GLP-1 receptor agonist, is not recommended [1] — which matters if you are moving between a branded product and a compounded one.

What lab changes does Zepbound cause?

Two, both on the label [1]:

• Amylase rose by an average of 20% to 25% from baseline, versus 2.1% on placebo.
• Lipase rose by an average of 28% to 35%, versus 5.8% on placebo.

The label states that the clinical significance of these elevations is unknown in the absence of other signs and symptoms of pancreatitis. In other words, a raised lipase on a routine panel is expected on this drug and is not by itself a diagnosis.

The label also reports that most people develop antibodies to tirzepatide: 64.5% in the weight trials. Those antibodies had no identified effect on how well the drug worked, but people who had them reported more hypersensitivity reactions (6.2% versus 3%) and far more injection site reactions (11.3% versus 1%) [1].

Does Zepbound cause muscle loss?

Partly, and the label addresses it obliquely. The Clinical Pharmacology section states that “tirzepatide lowers body weight with greater fat mass loss than lean mass loss” [1]. Muscle loss is not listed as an adverse reaction.

The study behind that sentence is a substudy of SURMOUNT-1 in 160 participants who had body scans at baseline and week 72. Body weight fell 21.3% on tirzepatide versus 5.3% on placebo. Fat mass fell 33.9% versus 8.2%; lean mass fell 10.9% versus 2.6%. In absolute terms that is about 15.9 kg of fat and 5.6 kg of lean mass [9].

So roughly three quarters of the weight lost was fat and one quarter was lean mass — and that same split showed up in the placebo group, which is the usual ratio for weight loss by any means. The substudy is small, body scanning is an imperfect tool, and there are no head-to-head body composition data against semaglutide.

What side effects are reported but not on the Zepbound label?

This is the section worth reading carefully, because the evidence here is weaker and the internet treats it as though it were not.

NAION (sudden optic nerve vision loss). Not on the Zepbound label. As of September 14, 2026, a full-text search of the current prescribing information returns no occurrence of “NAION” or “optic neuropathy” [1]. In its October–December 2024 quarterly report of potential signals, the FDA listed NAION for the whole GLP-1 class, naming Mounjaro and Zepbound, and the entry said the agency was evaluating the need for regulatory action [8]. Reporting-database studies do find a tirzepatide signal, but a much weaker one than semaglutide’s. European regulators added NAION to semaglutide product information in 2025; they did not add it to tirzepatide’s.

Starvation and euglycemic ketoacidosis. Not on the label. It shows up repeatedly as an unlabeled signal in analyses of the FDA’s spontaneous reporting database, and there is a growing set of published case reports, almost all in people without diabetes who stopped eating because of nausea. These are case reports, not incidence data.

Gastroparesis. Not listed as an adverse reaction. The label instead says Zepbound is not recommended in severe gastroparesis, and describes delayed gastric emptying in the pharmacology section as largest after the first dose and diminishing over time [1]. Whether it can become persistent is the central contested question in ongoing litigation, and no court has ruled on it.

Muscle atrophy, sleep disorder, menstrual changes, food cravings. All flagged as reporting signals in published analyses of spontaneous report databases. Spontaneous reports measure how often something gets reported, not how often it happens, and cannot establish cause.

The short version

Zepbound’s common side effects are overwhelmingly digestive, front-loaded into the dose-escalation weeks, and the reason about one in sixteen people stops the drug. The serious warnings are mostly low-frequency: gallbladder disease has the clearest randomized signal, pancreatitis occurred at the placebo rate in the weight trials, and the thyroid boxed warning rests on rat data whose human relevance the label says is unknown.

The Zepbound label is more detailed than the Mounjaro one. Hair loss, fatigue, dizziness and hypotension carry percentages on Zepbound and appear nowhere on Mounjaro with a frequency. Injection site reactions, hypersensitivity reactions, burping and reflux are on both labels with percentages, but Zepbound puts them in the adverse reaction table while Mounjaro reports them in narrative text at lower rates. That is a difference in how the two trial programs recorded things, not necessarily a difference in the drug.

If something you are experiencing is not in this table, that does not mean it is not real. It means it is not in the label’s dataset, and it is worth raising with a healthcare provider rather than searching for.

Sources

1. ZEPBOUND (tirzepatide) prescribing information, revised 8/2026 — DailyMed
2. FDA approved Zepbound label, supplement s041 (2026)
3. Tirzepatide versus semaglutide in obesity (SURMOUNT-5), New England Journal of Medicine
4. Tirzepatide for obstructive sleep apnea (SURMOUNT-OSA)
5. FDA request for removal of the suicidal behavior and ideation warning, January 13, 2026
6. OCULUS randomized trial of holding versus continuing GLP-1 agonists before endoscopy, JAMA Internal Medicine
7. GLP-1 receptor agonists and pulmonary aspiration during endoscopy, BMJ 2024
8. FDA quarterly report of potential signals of serious risks, October–December 2024
9. SURMOUNT-1 body composition substudy
10. SURPASS-CVOT, New England Journal of Medicine
11. FDA approves first medication for obstructive sleep apnea