Question 1

What are the side effects of Saxenda and liraglutide?

Accepted Answer

The gut carries most of the record, and the cleanest figures come from a trial that ran liraglutide against placebo for 48 weeks in people with fatty liver inflammation. In that trial, gastrointestinal disorders were reported in 21 (81%) of 23 patients in the liraglutide group and 17 (65%) of 22 patients in the placebo group. Split by symptom, the same trial counted diarrhoea in ten patients on liraglutide against five on placebo, constipation in seven against none, and loss of appetite in eight against two. Look at the placebo column before the top line. Two in three people on a dummy injection also reported stomach or bowel trouble, so what liraglutide added is the gap between the arms rather than the headline figure. That trial used 1.8 mg a day, and a 2011 review of the diabetes programme names the most common side-effect reported as mild, transient nausea. Across the class, a 2025 review of trials in adults without diabetes found that the majority of adverse events were gastrointestinal-related, at 47% to 84% on the active injections against 13% to 63% on placebo.

Question 2

Is Saxenda a high-risk medication?

Accepted Answer

High risk is not a category any of these trials used, so the useful answer is what the serious end of the record actually holds. A 2025 review of placebo-controlled trials in adults without diabetes, liraglutide among them, found that serious adverse events ran at 0% to 10% on the active injections against 0% to 12% on placebo. Stopping because of side effects was the commoner outcome, at 0% to 26% of participants on the active arms against 0% to 9% on placebo. In the SCALE Insulin trial, which gave liraglutide 3.0 mg for 56 weeks to people already injecting insulin, the investigators reported that no new safety or tolerability issues were observed. The placebo column matters again here. Serious events turned up on dummy injections at a similar rate, and that comparison is exactly what a placebo arm exists to make. So the trial record describes a medicine whose common harms are frequent and mostly digestive, and whose serious harms did not separate from placebo in those trials. That is a description of the people who were enrolled rather than a rating of the risk to any one reader, and the people who were not enrolled are set out further down.

Question 3

Do liraglutide side effects ease over time?

Accepted Answer

In the 48-week liver trial, most adverse events were grade 1 (mild) to grade 2 (moderate) in severity and transient, which means they passed rather than lasting the whole course in those patients. A 2025 review of real-world use says much the same about the class outside trials, that these side effects are often manageable and decrease over time. The same review is the reason not to read that too comfortably, because it also names nausea and digestive problems as the main reasons patients stop taking these treatments. Both statements are true at once. Side effects that fade for the people who stay on an injection can still be the reason other people left before they faded, and an average taken at the end of a trial only counts the people who stayed. Neither source gives a week by which the nausea on liraglutide settles, so this page does not invent one. How the amount is stepped up in the published work is laid out on the Liraglutide dosage page, which is where the timing question properly belongs.

Question 4

Can liraglutide cause low blood sugar?

Accepted Answer

Low blood sugar, or hypoglycaemia, matters most for anyone who is also on insulin, and liraglutide has a trial built around exactly that group of people. SCALE Insulin enrolled people with overweight or obesity and type 2 diabetes treated with basal insulin and ≤2 oral antidiabetic drugs, then gave them liraglutide 3.0 mg or placebo. Over 56 weeks, more hypoglycemic events were observed with placebo than liraglutide 3.0 mg, and that was despite lower doses of basal insulin in the liraglutide arm. A 2011 review offers the reason, reporting that liraglutide does not impair the body's own glucagon response to a falling sugar level in patients with type 2 diabetes. Glucagon is the hormone that pushes sugar back up, so that response is the body's own brake. One detail changes what the trial result means. The insulin came down as the trial went on, so the finding describes liraglutide inside a managed protocol rather than liraglutide added on top of an unchanged insulin dose, and those are different situations with different risks.

Question 5

Is liraglutide hard on the liver?

Accepted Answer

The liver record points in two directions, and the difference between them is the kind of evidence each one is. On one side sits a single case report of a 52-year-old woman given liraglutide as part of her weight-reduction program, who then developed an idiosyncratic drug-related liver injury. Idiosyncratic means unpredictable and particular to one person. On the other side is a randomised trial in people who already had fatty liver inflammation, where scarring progressed in two of 23 patients on liraglutide against eight of 22 on placebo. Scarring here means fibrosis, the stiffening that follows long inflammation in the liver. One case shows that a liver injury can follow liraglutide in a particular person, and the trial shows the average direction across 45 biopsied patients. Neither cancels the other. A single case says nothing about how often something happens, while a trial of that size is too small to catch a reaction that strikes one person in thousands, which is why both belong on the page.

Question 6

Does liraglutide carry a thyroid or pancreas signal?

Accepted Answer

Thyroid cancer is the loudest signal in the FDA's own reporting system, and a 2025 search of it across five medicines in this class, liraglutide included, reported that medullary and papillary thyroid carcinoma had the highest signals and were significant in virtually all medications. That system collects reports rather than rates. A signal means a pairing was filed more often than expected, which is a reason to look and not a measured risk. Observational studies point the other way, and a 2025 review of real-world use found no clear increase in risks of severe events like pancreatitis or pancreatic cancer, thyroid disorders in the people those studies followed. Pancreatitis and pancreatic cancer were also named safety outcomes in a 2019 pooling of heart-outcome trials in patients with type 2 diabetes, and LEADER, the liraglutide trial, is one of the seven it included. So the reporting system raises the thyroid question and the observational record does not confirm it. That is neither a settled risk nor a dismissed one. How the same database reads across every molecule in the class is taken apart on the GLP-1 safety page.

Question 7

Does liraglutide affect mood or cause suicidal thoughts?

Accepted Answer

Reports of suicidal thoughts in people on these medicines led the European Medicines Agency to investigate, and a 2024 analysis went through the European report database to see what had been filed. Across semaglutide, liraglutide and tirzepatide, it found 372 reports of mental health problems, which came to just over 1% of all the reports filed for the three. Depression was the most common of those, then anxiety, then thoughts of suicide. The fatal outcomes occurred primarily among men, eight of the nine, and eight of those nine deaths were filed against liraglutide. Against that, the 2025 real-world review found no clear rise in the risk of depression and self-harm in the studies it read. Liraglutide also came first, with semaglutide made later, and an older medicine piles up more reports just by being on sale for longer. A database can show what was filed but not what caused it, which is why the authors wrote that the severity and fatal outcomes of some of these reports warrant further investigation.

Question 8

What does liraglutide do to bone and muscle?

Accepted Answer

Bone has a randomised answer. In a Copenhagen trial, 195 adults who had already lost weight on a low-calorie diet spent a year on exercise, liraglutide 3.0 mg, both together, or placebo, and their bone density was scanned at the hip, spine and forearm. Compared with the exercise group, bone mineral density decreased in participants on liraglutide alone at both the hip and the spine. In the combination group, exercise plus liraglutide, bone density was unchanged compared with the placebo group at the hip. Muscle reads more kindly. A 2025 network meta-analysis found liraglutide was the only GLP-1RA to achieve significant weight reduction without significantly reducing lean mass across the trials it pooled. Lean mass means all body tissue other than fat, which includes muscle. The two findings are not in tension. One is a scan of bone in a single trial, the other a pooled reading of lean mass, and they describe different tissues rather than one verdict on what liraglutide does to the body.

Question 9

Is liraglutide hard on the heart or kidneys?

Accepted Answer

This is the one safety question where the record runs the other way, because it comes from trials built to count harm rather than to measure weight. A 2019 meta-analysis of seven heart-outcome trials in patients with type 2 diabetes, LEADER among them, found that the class reduced major adverse cardiovascular events by 12% in those patients. Its authors concluded that these medicines have beneficial effects on cardiovascular, mortality, and kidney outcomes in patients with type 2 diabetes. Two limits travel with that. It is a class result with one liraglutide trial inside it, and it describes people who had diabetes rather than people using liraglutide for weight alone. The weight trials add a possible reason. In adults at high cardiovascular risk, liraglutide 3.0 mg lowered the fat packed around the organs, and those investigators suggest that visceral fat reduction may be one mechanism to explain the benefits seen on cardiovascular outcomes in earlier trials. May is the operative word in that sentence, since a suggested mechanism is not a measured one.

Question 10

Who was left out of the liraglutide trials?

Accepted Answer

Pregnancy is the plainest gap, and a 2023 JAMA review describes weight-loss medicines as recommended for nonpregnant patients only. Age is the next. The Copenhagen trial enrolled adults aged 18 to 65 years with a body-mass index of 32 to 43 and no diabetes, which leaves out older adults and anyone lighter or heavier than that band. Who actually enrolled narrows it further, and the visceral fat trial ran with 92% female participants at a mean age of about 50. Children are the partial exception, since liraglutide has trials as young as six, but the investigators in that trial wrote that its safety and efficacy have not been established in children before they began. None of this makes liraglutide unsuitable for the people left out. It means the published rates were measured on somebody else, which is a different statement from a finding of harm, and a reader outside those bands is reading figures that do not describe them.

Question 11

What does liraglutide interact with?

Accepted Answer

Insulin is the documented one. In SCALE Insulin, people on liraglutide 3.0 mg ended the trial with less need for insulin versus placebo, which is a dose change somebody has to manage. A fixed pairing exists as well, since Xultophy combines insulin degludec and liraglutide in one pen at 100 units/mL and 3.6 mg/mL, and a completed phase 3 trial compared two different titration algorithms for it. Metformin is the other common partner. In the trial in young people with type 2 diabetes, all the patients received metformin during the trial, so its result describes the pair rather than liraglutide alone. The quieter interaction is with timing. A 2011 review reports that liraglutide delays the rate of gastric emptying in patients, meaning food leaves the stomach more slowly. None of the studies covered here measured what that slower emptying does to tablets taken alongside it, so that remains a question rather than an answer. It is a gap in what was tested rather than evidence that nothing happens.

Question 12

What do the liraglutide recalls say about risk?

Accepted Answer

Three of the four recall listings were sample pens filed on one day in March 2021, for Saxenda, Victoza and Xultophy, and all three were pulled over temperature abuse. The notice states the hazard plainly, that storage below freezing could cause a lack of efficacy and damage to the cartridge and pen-injectors. The fourth, a 2026 generic pen made for Lupin, was withdrawn over the presence of particulate matter in the cartridge. Read what those reasons describe. Neither is about the molecule; both are about a container that left its specification, and the first one describes a pen that may work less well than it should rather than one that does harm. Each was caught because the pen carried a batch, a firm and a filing. A vial outside that supply has no register entry to land on, which is a different risk rather than a smaller one, and it is why the record above belongs to the pen rather than to anything else carrying the same chemical name.

Question 13

How long does the liraglutide safety record run?

Accepted Answer

The longest randomised follow-up among these sources is two years, made of a year on liraglutide 3.0 mg in Copenhagen and then outcome assessments one year after treatment termination, at week 104. Across the class, the 2025 review of trials in adults without diabetes found that treatment ranged from 16 to 104 weeks, with a median of 43. The heart-outcome trials were built to count a narrow set of events, cardiovascular death, stroke, or myocardial infarction, rather than everything else a person on daily injections for years might want to know. The real-world review lists what is still missing, asking for studies on the long-term effects of these therapies on outcomes including mental health, cancer, and rare conditions like eye diseases. So two years is the edge of the randomised record for weight use, and anything past it is observation rather than measurement. That does not mean harm appears later. It means the controlled record among these sources stops there, and a reader on year three is past the edge of what was measured.

Liraglutide side effects and safety data