Tirzepatide Side Effects: 28% Nausea at 15 mg, 8% on Placebo

Tirzepatide is a dual GIP and GLP-1 receptor agonist (PMID 32519795). Mounjaro is the tirzepatide product whose label safety data come from adults with type 2 diabetes mellitus (Mounjaro label section 6.1). Zepbound's approved uses include chronic weight reduction and maintenance in adults with obesity or overweight with a related comorbidity, and moderate-to-severe obstructive sleep apnea with obesity (Zepbound label, Medication Guide).

PeptidesDirect does not sell tirzepatide; this article summarizes the published safety data. For the study reference see tirzepatide. This article is a report of published study results, not medical advice and not a dosing protocol.

The headline nausea figures are 28 % in the 15 mg arm versus 8 % on placebo (Zepbound label section 6.1). This report draws on the SURMOUNT obesity program, SURPASS diabetes program, US labels, EU Summary of Product Characteristics (SmPC), and posted registry tables. Personal accounts are outside its evidence base. Each result belongs to its named population, comparator and observation window.

Gastrointestinal Side Effects: The Main Label Table

Nausea, diarrhea, vomiting and constipation recur throughout the trial record. The Zepbound pooled safety window is 72 treatment weeks plus a 4-week off-drug follow-up (Zepbound label section 6.1). The label describes treatment for up to that duration. These are reported reaction percentages, with the placebo column retained beside the active arms.

The pattern differs by symptom. Vomiting increases across the active columns, whereas constipation decreases. An aggregate gastrointestinal row therefore does not describe the behavior of every individual symptom. Participants may also appear in several symptom rows, so adding those rows would count some people repeatedly.

Registry Counts and Their Longer Window

SURMOUNT-1 records the following counts from baseline to week 193, including extended observation (NCT04184622 posted results). These are participants with an event, not numbers of episodes.

Severe Events and Symptom-Specific Discontinuation

The open-access SURMOUNT-3 Table 3 covers the double-blind period and the 4-week safety follow-up (PMID 37840095). It separates events leading to study-drug discontinuation by term:

Severe gastrointestinal reactions in the Zepbound pool were 1.7 % in the 5 mg arm, 2.5 % in the 10 mg arm and 3.1 % in the 15 mg arm versus 1 % on placebo (Zepbound label section 5.2). Severity describes intensity; the serious-event classification is a separate endpoint.

How Many Stopped Because of Adverse Events?

Permanent treatment discontinuation, gastrointestinal discontinuation and withdrawal from a study period answer different questions. A participant can stop study medication while remaining in follow-up. That distinction explains why a participant-flow row can be much smaller than a drug-discontinuation total.

The Zepbound rows use the pooled window described above; the Mounjaro row belongs to its placebo-controlled adult diabetes pool. It measures gastrointestinal discontinuation specifically.

SURMOUNT-3 reported study-drug discontinuation because of adverse events in 30 participants (10.5 %) versus 6 (2.1 %) on placebo during the double-blind and safety follow-up periods (PMID 37840095). Its overall-study flow row records withdrawal for an adverse event in 4 of 287 participants versus 2 of 292 on placebo (NCT04657016 posted results).

SURMOUNT-1's Primary Treatment Period flow row lists 4 of 630 participants in the 5 mg arm, 8 of 636 in the 10 mg arm and 6 of 630 in the 15 mg arm versus 6 of 643 on placebo (NCT04184622 posted results). These flow counts retain that period definition rather than the longer adverse-event window.

Tirzepatide Versus Semaglutide: Direct Comparisons

SURMOUNT-5 was an open-label, active-controlled obesity trial. Its registry window runs from baseline up to 72 weeks (NCT05822830 posted results). MTD means maximum tolerated dose, identifying the trial arm. The table reports observed participants for each event.

Different rows favor different arms numerically. These counts do not establish a universal tolerability ranking, and open-label reporting differs from blinded ascertainment.

The diabetes comparison, SURPASS-2, was also open-label, with a baseline-to-safety-follow-up window of 44 weeks (NCT03987919 posted results). Its semaglutide comparator was a different trial arm:

The Diabetes Program: Background Therapy Matters

The Mounjaro placebo-controlled pool includes SURPASS-1 and SURPASS-5, with mean exposure of 36.6 weeks (Mounjaro label section 6.1). Its reaction table remains separate from the obesity pool:

For hypoglycemia, the label's glucose threshold is below 54 mg/dL (Mounjaro label section 6.1). Monotherapy and basal-insulin observations cover 40 treatment weeks plus 4 weeks of treatment-free safety follow-up; the sulfonylurea observations cover up to 104 weeks (Mounjaro label section 6.1).

Severe hypoglycemia with basal insulin was 0 % in the 5 mg arm, 2 % in the 10 mg arm and 1 % in the 15 mg arm versus 0 % on placebo (Mounjaro label section 6.1). A glucose-threshold event and an event requiring assistance are distinct endpoints.

SURPASS-CVOT adds a longer active comparison, up to 259 weeks, in participants with established atherosclerotic cardiovascular disease (NCT04255433 posted results).

Injection-Site Reactions

The often-quoted pair of 3.2 % versus 0.4 % on placebo belongs to the Mounjaro diabetes pool (Mounjaro label section 6.1). Among treated adults in its controlled-trial antibody analysis, reactions occurred in 4.6 % with anti-tirzepatide antibodies versus 0.7 % without them (Mounjaro label section 6.1). Both antibody groups received tirzepatide.

The Zepbound antibody comparison similarly reports 11.3 % in antibody-positive participants versus 1 % in antibody-negative participants (Zepbound label section 6.1). These subgroup comparisons do not isolate the reason an individual reaction occurred.

SURMOUNT-3 records injection-site reaction in 32 of 287 participants versus 3 of 292 on placebo, baseline to week 76 (NCT04657016 posted results). SURMOUNT-5 records 32 of 374 versus 1 of 376 on semaglutide, baseline up to 72 weeks (NCT05822830 posted results). The comparators and observation periods remain attached to each result.

Hair Loss: Label Rates and Alopecia Rows

Hair loss was reported in 5 % in the 5 mg arm, 4 % in the 10 mg arm and 5 % in the 15 mg arm versus 1 % on placebo (Zepbound label section 6.1). The sex-specific figures were 7.1 % versus 1.3 % on placebo among female participants, and 0.5 % versus 0 % among male participants (Zepbound label section 6.1).

SURMOUNT-1's alopecia row, baseline to week 193, lists 33 of 630 participants in the 5 mg arm, 32 of 636 in the 10 mg arm and 37 of 630 in the 15 mg arm versus 7 of 643 on placebo (NCT04184622 posted results). SURMOUNT-3 lists 20 of 287 versus 4 of 292 on placebo, baseline to week 76 (NCT04657016 posted results).

The EU SmPC categorizes hair loss as common, mainly applying to participants with overweight or obesity (EU SmPC section 4.8). That frequency category does not identify the mechanism of shedding. Our hair-loss article provides class background; the rates here remain tirzepatide-specific.

Heart Rate and Blood Pressure

The Zepbound label reports a mean heart-rate increase of 1 to 3 beats per minute versus no increase on placebo (Zepbound label section 6.1). A meta-analysis reports a pooled mean difference versus placebo of 2.05 bpm, with a 95 % confidence interval of 0.96 to 3.13 (PMID 41582189). Its dose-level network estimate for the 5 mg arm was 0.52 bpm, with a 95 % confidence interval of -2.71 to 3.78 (PMID 41582189). These estimates describe different analyses, rather than interchangeable trial-arm measurements.

In the Mounjaro pool, sinus tachycardia accompanied by a heart-rate rise of at least 15 bpm occurred in 4.6 % in the 5 mg arm, 5.9 % in the 10 mg arm and 10 % in the 15 mg arm versus 4.3 % on placebo (Mounjaro label section 6.1).

Hypotension was reported in 1 % in the 5 mg arm, 1 % in the 10 mg arm and 2 % in the 15 mg arm versus 0 % on placebo (Zepbound label section 6.1). This reaction category includes decreased blood pressure and orthostatic hypotension; it is not a mean blood-pressure change.

Gallbladder, Pancreatitis and Pancreatic Enzymes

In the Zepbound pool, adjudication-confirmed acute pancreatitis was reported in 0.2 % versus 0.2 % on placebo (Zepbound label section 6.1). Cholelithiasis was 1.1 % versus 1 %, and cholecystitis was 0.7 % versus 0.2 % on placebo (Zepbound label section 6.1). These are separate outcomes.

Across Mounjaro clinical studies, adjudication confirmed 14 acute-pancreatitis events in 13 treated adults versus 3 events in 3 comparator-treated adults (Mounjaro label section 6.1). SURMOUNT-OSA reported 2 adjudicated cases in the tirzepatide arm of trial 2 during the planned 52-week treatment period plus 4-week safety follow-up; that passage prints no comparator count (PMID 38912654).

Mounjaro's mean pancreatic amylase increase was 33 % to 38 % versus 4 % on placebo; mean lipase increased 31 % to 42 % versus no change on placebo (Mounjaro label section 6.1). Laboratory concentration changes are not pancreatitis diagnoses. Adjudicated diagnoses, laboratory measurements and reported event terms therefore remain separate in a safety report.

Dysesthesia and Other Sensory Terms

Dysesthesia describes an unpleasant altered sensation. The Zepbound pool reports 0.2 % in the 5 mg arm, 0.2 % in the 10 mg arm and 0.4 % in the 15 mg arm versus 0.1 % on placebo (Zepbound label section 6.1). Mounjaro reports 0.4 % in each active arm versus 0 % on placebo (Mounjaro label section 6.1).

SURMOUNT-J's dysesthesia and paresthesia rows each list 1 of 73 participants in the 10 mg arm and 1 of 77 in the 15 mg arm versus 0 of 75 on placebo, baseline through safety follow-up up to 76 weeks (NCT04844918 posted results). Its reporting threshold was 0 % (NCT04844918 posted results). Separate term rows cannot establish whether the same participant experienced both sensations.

Mood and Suicidality: The Population Limit

The pooled psychiatric analysis excluded lifetime suicide attempt and active severe psychiatric illness. Its findings consequently apply to the enrolled population without known major psychopathology (PMID 41537305).

From randomization through the 4-week off-treatment visit after week 72, suicidal ideation was reported by 17 of 2793 tirzepatide participants versus 7 of 1246 placebo participants, both printed as 0.6 % (PMID 41537305). Suicidal behavior was reported by two tirzepatide participants versus none assigned to placebo (PMID 41537305).

Among participants with baseline PHQ-9 scores below 15, a maximum score of at least 15 occurred in 33 participants (1.2 %) versus 29 (2.3 %) on placebo (PMID 41537305). Questionnaire thresholds, ideation and behavior are separate measures. The Zepbound prescribing information of September 2026 contains no warning section on suicidal ideation (Zepbound US prescribing information, DailyMed setid 487cd7e7-434c-4925-99fa-aa80b1cc776b). This label fact does not broaden the psychiatric analysis beyond its exclusions.

Serious Adverse Events and Deaths by Trial

The registry group lines below retain their printed denominators. The columns give participants with serious adverse events and recorded deaths, respectively.

The cardiovascular trial's death denominators differ from its safety denominators. Preserving that distinction matters more than making every column look uniform.

Real-World Studies and Pharmacovigilance

Crisafulli and colleagues examined a composite of acute pancreatitis, biliary disease, bowel obstruction, gastroparesis and severe constipation. Tirzepatide's hazard ratio was 0.96 versus dulaglutide, with a 95 % confidence interval of 0.77 to 1.20, and 1.07 versus semaglutide, with a 95 % confidence interval of 0.90 to 1.26 (PMID 41183330). The matched-cohort design addresses routine practice while retaining possible residual confounding.

A Japanese single-center series reported approximately 1.3 % gastrointestinal discontinuation among 219 patients, without a comparator (PMID 41462404). That uncontrolled result cannot establish why discontinuation differed from a randomized trial.

FAERS measures reporting patterns. Nausea had a reporting odds ratio of 4.01 versus all other drugs in FAERS, with a 95 % confidence interval of 3.85 to 4.19 (PMID 39141075). Pancreatitis had a reporting odds ratio of 3.63 versus all other drugs in FAERS, with a 95 % confidence interval of 3.15 to 4.19 (PMID 39141075). Reporting odds are not incidence: the database lacks an exposed-patient denominator.

Longer Follow-Up and Randomized Withdrawal

The three-year SURMOUNT-1 extension concerns participants with obesity and prediabetes: 176 treatment weeks followed by a 17-week off-treatment period (PMID 39536238). Its abstract describes gastrointestinal events as mostly mild to moderate (PMID 39536238). The detailed registry counts above cover a broader population and cannot be assigned to this subgroup abstract.

For SURMOUNT-4, the randomized-withdrawal registry window is week 36 to week 92 (NCT04660643 posted results). Nausea occurred in 27 of 335 participants continuing tirzepatide versus 9 of 335 switched to placebo; serious adverse events were 10 of 335 in each arm (NCT04660643 posted results). These participants had already completed the open-label lead-in. The withdrawal comparison consequently describes a selected population with prior exposure.

Who the Programs Enrolled and Excluded

The obesity program includes adults with obesity or overweight and related comorbidity. SURMOUNT-1 excluded diabetes, while SURMOUNT-3 randomized participants after an intensive lifestyle lead-in (NCT04184622 posted results; NCT04657016 posted results). SURMOUNT-4 randomized lead-in completers (NCT04660643 posted results). The psychiatric analysis additionally excluded lifetime suicide attempt and active severe psychiatric illness (PMID 41537305).

The diabetes program spans different background treatments and cardiovascular histories. SURPASS-2 enrolled adults on metformin, whereas SURPASS-CVOT required established atherosclerotic cardiovascular disease (NCT03987919 posted results; NCT04255433 posted results). The Mounjaro label separately reports monotherapy, basal-insulin and sulfonylurea observations (Mounjaro label section 6.1). Eligibility and background therapy are part of the meaning of every reported rate.

Mechanism: Retatrutide and Semaglutide

Tirzepatide combines GIP and GLP-1 receptor activity (PMID 32519795). Our retatrutide side-effects article and semaglutide science article provide pharmacological background on those compounds. Receptor targets alone cannot establish the relative frequency of a particular adverse event.

Open Questions and Reporting Gaps

Several primary papers (SURMOUNT-1, -2, -4, -5, SUMMIT, SURPASS-2, -3, -4 and -6) were read at abstract level for this article because their full texts are paywalled. Registry tables supply arm-level counts, but they answer the registry's question over its stated window. The reviewed sources leave per-arm heart-rate changes and detailed symptom rates in the extended prediabetes analysis unresolved; the heart-rate evidence presented here comes from labels and meta-analysis.

Reporting thresholds also constrain interpretation. SURMOUNT-1 posts non-serious terms at a 5 % threshold (NCT04184622 posted results). An omitted term cannot be read as zero events. Likewise, the reviewed cardiovascular-trial flow table leaves adverse-event discontinuation unresolved, and the Mounjaro pool supplies gastrointestinal discontinuation rather than a complete all-reaction discontinuation estimate. These are limits of the reviewed documents, not evidence that an outcome never occurred.

Conclusions for Research

Gastrointestinal reactions dominate the reported tolerability findings, with symptom-specific patterns across trial arms. Discontinuation requires its own endpoint definition, particularly when study withdrawal and stopping medication are reported separately. Direct semaglutide comparisons describe the enrolled populations and comparators, while pharmacovigilance provides reporting signals. Reliable interpretation preserves the source, denominator and observation window of every result.

This article is for informational purposes only for scientific research.

Frequently Asked Questions

---

Sources and further reading:

- Urva S et al. The novel dual glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1 (GLP-1) receptor agonist tirzepatide transiently delays gastric emptying similarly to selective long-acting GLP-1 receptor agonists. Diabetes Obes Metab, 2020. PMID 32519795. PubMed: https://pubmed.ncbi.nlm.nih.gov/32519795/

- Wadden TA et al. Tirzepatide after intensive lifestyle intervention in adults with overweight or obesity: the SURMOUNT-3 phase 3 trial. Nat Med, 2023. PMID 37840095. PubMed: https://pubmed.ncbi.nlm.nih.gov/37840095/

- Zhang Y et al. Effect of glucagon-like peptide-1 receptor agonists on heart rate in non-diabetic individuals with overweight or obesity: a systematic review and pairwise and network meta-analysis of randomized controlled trials. Eur J Med Res, 2026. PMID 41582189. PubMed: https://pubmed.ncbi.nlm.nih.gov/41582189/

- Malhotra A et al. Tirzepatide for the Treatment of Obstructive Sleep Apnea and Obesity. N Engl J Med, 2024. PMID 38912654. PubMed: https://pubmed.ncbi.nlm.nih.gov/38912654/

- Wadden TA et al. Psychiatric Safety of Tirzepatide in People With Obesity and No Known Major Psychopathology: A Post Hoc Analysis of SURMOUNT. Obesity (Silver Spring), 2026. PMID 41537305. PubMed: https://pubmed.ncbi.nlm.nih.gov/41537305/

- Crisafulli S et al. Comparative Gastrointestinal Safety of Dulaglutide, Semaglutide, and Tirzepatide in Adults With Type 2 Diabetes. Ann Intern Med, 2026. PMID 41183330. PubMed: https://pubmed.ncbi.nlm.nih.gov/41183330/

- Hiraide T et al. Low discontinuation rate of tirzepatide treatment in Japanese patients with diabetes mellitus; importance of traditional Japanese diet. J Health Popul Nutr, 2025. PMID 41462404. PubMed: https://pubmed.ncbi.nlm.nih.gov/41462404/

- Caruso I et al. The real-world safety profile of tirzepatide: pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS) database. J Endocrinol Invest, 2024. PMID 39141075. PubMed: https://pubmed.ncbi.nlm.nih.gov/39141075/

- Jastreboff AM et al. Tirzepatide for Obesity Treatment and Diabetes Prevention. N Engl J Med, 2025. PMID 39536238. PubMed: https://pubmed.ncbi.nlm.nih.gov/39536238/

- ClinicalTrials.gov. SURMOUNT-1, NCT04184622, posted results: https://clinicaltrials.gov/study/NCT04184622

- ClinicalTrials.gov. SURMOUNT-3, NCT04657016, posted results: https://clinicaltrials.gov/study/NCT04657016

- ClinicalTrials.gov. SURMOUNT-4, randomized-withdrawal period, NCT04660643, posted results: https://clinicaltrials.gov/study/NCT04660643

- ClinicalTrials.gov. SURMOUNT-5, NCT05822830, posted results: https://clinicaltrials.gov/study/NCT05822830

- ClinicalTrials.gov. SURMOUNT-J, NCT04844918, posted results: https://clinicaltrials.gov/study/NCT04844918

- ClinicalTrials.gov. SURPASS-2, NCT03987919, posted results: https://clinicaltrials.gov/study/NCT03987919

- ClinicalTrials.gov. SURPASS-CVOT, NCT04255433, posted results: https://clinicaltrials.gov/study/NCT04255433

- Zepbound (tirzepatide), US prescribing information and Medication Guide, Eli Lilly and Company, label version September 2026. DailyMed setid 487cd7e7-434c-4925-99fa-aa80b1cc776b: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=487cd7e7-434c-4925-99fa-aa80b1cc776b

- Mounjaro (tirzepatide), US prescribing information, Eli Lilly and Company, label version September 2026. DailyMed: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=d2d7da5d-ad07-4228-955f-cf7e355c8cc0

- Mounjaro (tirzepatide), EU Summary of Product Characteristics, section 4.8 Undesirable effects, European Medicines Agency: https://www.ema.europa.eu/en/documents/product-information/mounjaro-epar-product-information_en.pdf