Can Ozempic or Mounjaro Cause Liver Damage? Liver Enzymes Evidence Review

Ozempic, Wegovy, Mounjaro and Zepbound do not damage the liver in most people who take them. In the trials, liver enzymes went down on average, not up, and the drop tracked weight loss.^[13][14] None of the four labels carries a liver warning, and none asks for a dose change in liver disease.^[1][3] Drug-induced liver injury does happen, but rarely: a 2026 review counted 31 published case reports across the whole class, some severe, most improving once the drug was stopped.^[12] When liver tests rise on one of these drugs, the commoner culprit is a gallstone, a known effect of the drug and of fast weight loss.^[2]

The short answer

What happens to liver enzymes in the trials

ALT (alanine aminotransferase) is the enzyme a routine liver panel uses to flag liver cell damage. Many people starting a GLP-1 already have a high ALT, usually from fatty liver. In the semaglutide weight-management trial that a 2019 analysis examined, 52% of participants began with an elevated ALT. In that group, ALT fell 6% to 21% more than on placebo by the end of treatment, and it returned to normal in 25% to 46% of people, against 18% on placebo.^[13]

The same pattern holds for liraglutide. A patient-level pooling of six phase 3 diabetes trials found an abnormal ALT in about half of 4,442 patients; on the 1.8 mg dose it dropped 8.2 IU/L against 5.0 on placebo. After adjusting for weight lost, the difference was no longer significant.^[14] The semaglutide analysis reached the same conclusion.^[13] In plain terms: these drugs lower liver enzymes mostly by shrinking the fat in the liver, through weight loss.

Tirzepatide data point the same way. In a 26-week phase 2 diabetes trial, ALT fell from baseline in every group, and the two highest tirzepatide doses lowered it more than dulaglutide did. That was a post hoc analysis.^[15] In the Wegovy MASH trial, average ALT and AST were trending lower on semaglutide than on placebo from week 12 to week 72.^[2]

Two rows stand out because they go the other way. In the adolescent Wegovy trial, four teenagers on the drug had ALT at five times the upper limit of normal or more, and none on placebo did. The label adds that gallstones and other factors were involved in some of them.^[2] On Saxenda, five people out of more than 3,000 had ALT ten times the limit or higher, two of them over 20 and 40 times. The label says the cause was not worked up in most and the link to the drug is uncertain.^[5] These are small numbers. They are also the reason a raised liver test on these drugs deserves a look rather than a shrug. For the full set of teen blood results, see what changed in the teenagers’ bloodwork.

What the labels say about the liver

Ozempic, Wegovy, Mounjaro and Zepbound list no liver injury warning. The warnings they do carry that involve the liver area are about the gallbladder. The peptide drugs in this class have little or no liver metabolism,^[10] which helps explain why liver disease barely changes how much drug reaches the blood.

“No dose adjustment of OZEMPIC is recommended for patients with hepatic impairment. In a study in subjects with different degrees of hepatic impairment, no clinically relevant change in semaglutide pharmacokinetics (PK) was observed.”^[1]

— OZEMPIC prescribing information, section 8.7

That study gave one 0.5 mg injection to 44 people, from normal liver function to severe impairment, and found blood levels within a few percent of each other.^[17] Tirzepatide went through the same test with a single 5 mg dose in 32 people, with the same result.^[18] Both Mounjaro and Zepbound say no dosage adjustment is recommended for hepatic impairment.^[3][4] One caution: those studies used a single dose, in small groups.

Two labels in the class are more guarded. Saxenda, a daily liraglutide injection, says experience in hepatic impairment is limited and advises caution, and its postmarketing section lists hepatitis, cholestasis, raised bilirubin and raised liver enzymes as reported events.^[5] Foundayo, the oral pill, is a small molecule cleared by the liver; in severe hepatic impairment its exposure rose 4.6-fold, and the label does not recommend it for that group.^[6] The pill’s own trials did not show liver harm: in a sponsor-run pooling of seven phase 3 trials, liver enzymes fell on average and no case met the standard test for serious drug liver injury.^[16] More on the pill is in our Foundayo review.

What LiverTox says about Ozempic and Mounjaro

LiverTox is the National Institutes of Health reference on drug-induced liver injury. Its class overview states that GLP-1 drugs have little or no liver metabolism and have not been convincingly implicated in clinically apparent liver injury.^[10] The drug-level entries, revised in 2025, are more specific:

• Semaglutide: “There have been rare published reports of hepatotoxicity attributed to semaglutide therapy.”^[7]
• Tirzepatide: linked to “minimal instances of elevations in serum aminotransferase levels during therapy and rare episodes of clinically apparent liver injury.”^[8]
• Liraglutide: the entry, last revised in 2019, said no published reports existed at that time.^[9] Several have appeared since (below).
• Orforglipron: enzyme elevations no higher than on placebo, and no case of liver injury with symptoms or jaundice so far.^[11]

“Rare” is the operative word, and it is also all anyone can honestly say. Case reports cannot measure how often something happens, only that it can.

The case reports: what drug-induced liver injury on a GLP-1 looks like

A September 2026 review in Clinical Therapeutics gathered 31 published case reports of suspected liver injury on a GLP-1 drug or tirzepatide. Patients were 17 to 79, more often women. Most had symptoms, chiefly abdominal pain, nausea and vomiting, which look a lot like the ordinary stomach side effects of these drugs. Injury began between two weeks and six months after starting, most often as damage to liver cells rather than to bile flow, and most cases resolved after the drug was stopped. Some cases followed a dose increase or a switch from one GLP-1 to another, and a few presented as liver failure.^[12] The author judges the true incidence rare but unknown and probably undercounted, partly because these drugs have a reputation for helping the liver, so a clinician may not suspect them.^[12]

The restart case matters most. A liver injury that clears when a drug stops and comes back when it is restarted is the strongest evidence a single patient can give, and here the causality score reached “highly probable.”^[26] A separate 2026 report counted its own liraglutide case as the ninth published.^[27] Two other details recur across the reports: underlying liver disease, as with the hepatitis B patient,^[21] and rapid weight loss during dose escalation.^[23] The review lists those, along with female sex and switching drugs, as possible risk markers, not proven ones.^[12] Inference, labeled as such: no mechanism has been established, and the case authors lean toward an idiosyncratic, person-specific reaction.^[23]

Adverse-event reports: a signal, not a rate

FDA’s adverse event database (FAERS) collects voluntary reports from patients, clinicians and companies. As of its July 2026 update, it held 44 reports coding “drug-induced liver injury” among reports naming semaglutide, 63 for tirzepatide and 73 for liraglutide.^[29] A report means someone suspected a link. It does not confirm the drug caused the injury, and the counts cannot be turned into a risk because nobody knows how many people took each drug or how many events went unreported.

Studies that mine FAERS for disproportion, events reported more often with one drug than with others, point mainly at the gallbladder. One analysis of 68,351 incretin-drug reports from 2006 to 2023 found that GLP-1 drugs had weaker hepatobiliary associations than DPP-4 inhibitors, chiefly gallbladder and bile duct disease.^[28] That is a reporting signal, not an incidence. It fits the trial and label data rather than overturning them.

The likelier cause of a raised liver test: gallstones

A gallstone that slips into the bile duct can push up liver tests fast, especially bilirubin and the cholestatic enzymes, and it causes the same yellow eyes and dark urine that liver injury does. The Wegovy label notes that some ALT rises in its trials came alongside gallstones.^[2] Gallstones are a listed effect of every drug in this class. Wegovy’s adult weight trials recorded gallstones in 1.6% on the drug against 0.7% on placebo.^[2] Zepbound’s label ties its gallbladder events to weight reduction.^[4] Wegovy’s label goes one step further:

“Substantial or rapid weight loss can increase the risk of cholelithiasis; however, the incidence of acute gallbladder disease was greater in WEGOVY-treated patients than in placebo-treated patients, even after accounting for the degree of weight loss.”^[2]

— WEGOVY prescribing information, section 5.3

That is why a high ALT on a GLP-1 usually gets an ultrasound before anyone blames the drug itself. The adolescent Wegovy cases above came with gallstones in some patients,^[2] and the Saxenda label names gallstones as a confounder too.^[5] The full gallbladder numbers, and how they compare with pancreatitis, are in GLP-1 pancreatitis and gallbladder risk.

Other reasons liver enzymes rise on a GLP-1

• Fatty liver that was already there. Half of trial participants started with a high ALT.^[13][14] A first liver panel drawn after starting the drug may simply be the first one anyone checked.
• Existing liver disease. Untreated hepatitis B was the backdrop to one of the most severe semaglutide cases.^[21] Hepatitis B or C, alcohol, and autoimmune liver disease all need ruling out.
• Other medicines and supplements. Any of them can be the cause instead, so whoever reads the blood test needs the full list, including anything bought for weight loss.
• Compounded products. Pharmacy-made versions of these drugs lack FDA approval and were never part of the trials above. The liver data on this page describe the brand products. What is known about compounded vials is in what is in a compounded GLP-1 vial.

When a liver problem on a GLP-1 needs a doctor today

Without symptoms, a mildly raised ALT on a GLP-1 is common and usually not a reason to stop. A large jump, a rise in bilirubin, or a rise soon after a dose increase is a reason to call the prescriber before the next injection. If the drug is the cause, stopping it is the treatment, and in most published cases the liver recovered.^[12] Restarting the same drug after a confirmed injury is what produced the recurrence in the liraglutide case.^[26] Stomach pain on these drugs has its own triage guide: GLP-1 stomach pain: when to worry.

If you already have liver disease

Liver disease does not change the semaglutide or tirzepatide dose on the label.^[1][3] For fatty liver with inflammation and scarring, semaglutide is an approved treatment, and that evidence, including where it stops working in cirrhosis, is set out in GLP-1 drugs and fatty liver disease. Drugs that add glucagon activity to target the liver more directly are still experimental; see pemvidutide. Saxenda advises caution in hepatic impairment and Foundayo is not recommended in severe impairment.^[5][6] Anyone with cirrhosis or chronic viral hepatitis should have the liver checked before starting and followed during treatment, a step the authors of the hepatitis B case recommend.^[21]