Liraglutide Side Effects: Incidence Rates Across Clinical Trials

Print or save as PDFOpens the printable side-effect diary and your print dialog. No sign-up.

Where these incidence figures come from

Liraglutide is sold as two products with separate FDA labels: Saxenda, a 3 mg once-daily subcutaneous injection approved 2014-12-23 for weight management, and Victoza, a 1.2 to 1.8 mg once-daily injection approved 2010-01-25 for type 2 diabetes [1][2]. Label rates below come from each product's own label.

The Saxenda safety database covers five double-blind, placebo-controlled adult trials: 3,384 people on liraglutide 3 mg and 1,941 on placebo, treated for up to 56 weeks, plus a 56-week trial in 125 adolescents [1]. The Victoza adverse reaction table reflects 1,673 adults with type 2 diabetes and a mean exposure of 37.3 weeks [2].

The Saxenda label states that adverse reaction rates from clinical trials of one drug cannot be directly compared with rates in studies of another drug and may not reflect practice [1].

Gastrointestinal side effects at 3 mg (Saxenda)

Rates in liraglutide 3 mg versus placebo from the pooled adult trials [1]:

• Nausea: 39.3% vs 13.8%
• Diarrhea: 20.9% vs 9.9%
• Constipation: 19.4% vs 8.5%
• Vomiting: 15.7% vs 3.9%
• Dyspepsia: 9.6% vs 2.7%
• Abdominal pain: 5.4% vs 3.1%
• Eructation: 4.5% vs 0.2%

About 68% of Saxenda-treated adults reported a gastrointestinal disorder versus 39% on placebo, and nausea declined as treatment continued [1].

In the same trials, 9.8% of liraglutide patients and 4.3% of placebo patients stopped treatment for adverse reactions. The most common reactions leading to discontinuation were nausea (2.9% vs 0.2%), vomiting (1.7% vs less than 0.1%) and diarrhea (1.4% vs 0%) [1].

Gastrointestinal side effects at 1.2 and 1.8 mg (Victoza)

The Victoza label's adult table lists these reactions, all more frequent than placebo [2]:

• Nausea: 5% placebo, 18% at 1.2 mg, 20% at 1.8 mg
• Diarrhea: 4%, 10%, 12%
• Vomiting: 2%, 6%, 9%
• Decreased appetite: 1%, 10%, 9%
• Dyspepsia: 1%, 4%, 7%
• Constipation: 1%, 5%, 5%
• Headache: 7%, 11%, 10%

Withdrawals because of gastrointestinal adverse reactions occurred in 4.3% of Victoza-treated patients and 0.5% of placebo patients, mostly during the first 2 to 3 months of the trials [2].

Severe gastrointestinal reactions and dehydration

Severe gastrointestinal reactions occurred in 4.4% of Victoza patients taking 1.2 mg and 4.2% taking 1.8 mg, versus 1.1% on placebo [2]. Both labels state that liraglutide is not recommended in patients with severe gastroparesis [1][2].

The Victoza label describes postmarketing reports of acute kidney injury, some requiring hemodialysis, mostly in patients who had gastrointestinal reactions causing dehydration such as nausea, vomiting or diarrhea. The label says to monitor renal function in patients reporting reactions that could lead to volume depletion, especially during dose initiation and escalation [2]. The Saxenda label carries the same acute kidney injury warning [1].

Hypoglycemia: what the labels measured

In Saxenda's trial in adults with type 2 diabetes and overweight or obesity, severe hypoglycemia requiring another person's assistance occurred in 3 of 422 liraglutide patients (0.7%, all taking a sulfonylurea) and none of 212 placebo patients. Among patients taking a sulfonylurea, glucose below 54 mg/dL occurred in 28.2% on Saxenda versus 12.7% on placebo, with sulfonylurea doses cut 50% at the start of the trial per protocol. Among those not taking a sulfonylurea, the figures were 7.1% versus 4.5% [1].

For Victoza, hypoglycemia requiring assistance occurred in 8 liraglutide-treated patients (7.5 events per 1,000 patient-years); 7 of the 8 were also taking a sulfonylurea. As an add-on to metformin plus glimepiride, 27.4% of liraglutide patients could self-treat hypoglycemia versus 16.7% on placebo, and 2.2% versus 0 could not self-treat [2]. The Victoza label says to consider reducing the dose of insulin secretagogues or insulin when starting treatment [2].

Heart rate

In the Saxenda adult trials, resting heart rate rose by a mean of 2 to 3 beats per minute versus placebo. More liraglutide patients had a rise above 10 bpm at two consecutive visits (34% vs 19%) and above 20 bpm (5% vs 2%); at least one reading above 100 bpm was recorded in 6% versus 4%, and tachycardia was reported as an adverse reaction in 0.6% versus 0.1% [1]. The Saxenda label says to monitor heart rate at regular intervals [1].

Gallbladder disease

The Saxenda label reports cholelithiasis in 2.2% of liraglutide-treated adults versus 0.8% on placebo, and cholecystitis in 0.8% versus 0.4%. Most affected liraglutide patients needed cholecystectomy, and the excess occurred even after accounting for the degree of weight loss [1].

A UK population-based cohort of 71,369 adults starting diabetes drugs found hospitalized bile duct or gallbladder disease in 6.1 per 1,000 person-years among current GLP-1 analogue users versus 3.3 per 1,000 person-years among users of at least two oral antidiabetic drugs (adjusted hazard ratio 1.79, 95% CI 1.21 to 2.67) [3].

Pancreatitis

Acute pancreatitis, including fatal and non-fatal hemorrhagic or necrotizing cases, has been observed with GLP-1 receptor agonists including liraglutide. Both labels say to discontinue if pancreatitis is suspected; the Saxenda label adds that patients should be watched for persistent severe abdominal pain, sometimes radiating to the back, with or without nausea or vomiting [1][2].

A meta-analysis of 41 randomized trials covering 14,972 patients found no significant difference in pancreatitis risk between GLP-1 receptor agonists and comparators (Mantel-Haenszel odds ratio 1.01, 95% CI 0.37 to 2.76, P=0.99). The authors noted that the number of pancreatitis cases was very small and the confidence intervals wide [4].

Thyroid C-cell tumors and the boxed warning

Liraglutide causes dose- and duration-dependent thyroid C-cell tumors in rats and mice at clinically relevant exposures. Whether it causes thyroid C-cell tumors, including medullary thyroid carcinoma, in humans is unknown. Both labels contraindicate liraglutide in patients with a personal or family history of medullary thyroid carcinoma or MEN 2, and both state that routine calcitonin measurement or thyroid ultrasound is of uncertain value for early detection [1][2].

The Victoza label reports 7 cases of papillary thyroid carcinoma in liraglutide-treated patients and 1 in a comparator-treated patient in adult glycemic control trials (1.5 vs 0.5 cases per 1,000 patient-years), most smaller than 1 cm and found in surgical specimens after screening-prompted thyroidectomy [2]. Both labels state that postmarketing medullary thyroid carcinoma reports are insufficient to establish or exclude a causal relationship [1][2].

Injection sites, antibodies and hair loss

Injection site reactions occurred in 13.9% of Saxenda-treated adults versus 10.5% on placebo; the most common, each reported by 1% to 2.5%, were erythema, pruritus and rash at the site [1]. In the Victoza trials, injection site reactions such as rash and erythema were reported in about 2% of adults, and fewer than 0.2% stopped treatment for them [2].

Immunogenicity-related events, including urticaria, were more common with Victoza (0.8%) than with comparators (0.4%) [2]. Alopecia appears under postmarketing skin reactions in the Saxenda label with frequency not established; it is not in the trial adverse reaction table [1].

Mood and suicidal thoughts: the warning was removed in 2026

On 2026-01-13 the FDA asked the makers of Saxenda, Wegovy and Zepbound to remove suicidal behavior and ideation language from their labels, after reviewing a meta-analysis of 91 placebo-controlled GLP-1 receptor agonist trials (107,910 patients) plus a claims cohort of 2,243,138 new users and published studies. The FDA stated it found no increased risk of suicidal ideation or behavior with these drugs [5].

The current Saxenda label, effective 2026-02-25, has no suicidal behavior and ideation section, and the word does not appear in the label text. The Victoza label never included one [1][2]. What remains in the Saxenda label is depression reported as an adverse reaction in the adolescent trial: 4% with liraglutide versus 2.4% with placebo [1].

What the adolescent trial showed

In the 56-week Saxenda trial in 125 adolescents on liraglutide and 126 on placebo, nausea occurred in 42.4% versus 14.3%, vomiting in 34.4% versus 4%, diarrhea in 22.4% versus 14.3%, symptomatic hypoglycemia (glucose below 70 mg/dL) in 15.2% versus 4%, and depression in 4% versus 2.4%. Ten point four percent of liraglutide patients stopped for adverse reactions versus none on placebo [1][6].

For Victoza in patients aged 10 and older with type 2 diabetes, 21.2% had hypoglycemia with glucose below 54 mg/dL with or without symptoms (335 events per 1,000 patient-years), and no severe episodes occurred in the liraglutide group [2].

Liraglutide versus semaglutide in one head-to-head trial

SUSTAIN 10 randomized 577 adults with type 2 diabetes to once-weekly semaglutide 1.0 mg or once-daily liraglutide 1.2 mg for 30 weeks. Gastrointestinal disorders, the most common adverse events, occurred in 43.9% with semaglutide and 38.3% with liraglutide, and adverse events leading to premature discontinuation in 11.4% versus 6.6% [7].