New GLP-1 Side Effects: What Research Has Found Since 2024

What it is: Non-arteritic anterior ischemic optic neuropathy (NAION) is a rare eye condition in which blood flow to the optic nerve is suddenly reduced, causing rapid, often permanent, vision loss in one eye. It is sometimes described as a stroke of the optic nerve.

What the evidence shows: A July 2024 study in JAMA Ophthalmology by researchers at Mass Eye and Ear found that people with diabetes taking semaglutide were 4 times more likely to develop NAION than comparable patients not on the drug. Those using it for obesity were 7 times more likely. In June 2025, the European Medicines Agency completed its formal review and confirmed NAION as a very rare side effect of semaglutide, affecting up to 1 in 10,000 people. In December 2025, federal litigation in the US was centralised under MDL-3163 in Pennsylvania federal court.

The nuance: The absolute number of cases remains small — approximately 14 cases per 100,000 person-years among semaglutide users in the largest study. The EMA confirmed the signal while noting the medications’ substantial benefits remain. People with pre-existing vascular risk factors, sleep apnoea, or prior eye problems are at higher risk.

What to do: If you experience sudden or rapidly worsening vision in one eye while on semaglutide, contact your doctor immediately. This is a time-sensitive condition. People with existing optic nerve conditions or significant vascular risk factors should discuss this specifically with their prescriber before starting or continuing semaglutide.

What it is: Reductions in bone mineral density at the hip, femoral neck, and lumbar spine in people using GLP-1 medications, increasing osteoporosis and fracture risk over time.

What the evidence shows: A 2025 AAOS study of nearly 150,000 adults found approximately 30% increased relative risk of osteoporosis in GLP-1 users over five years. A February 2026 Weill Cornell study published in the Journal of Clinical Endocrinology and Metabolism found significant total hip bone loss in semaglutide and tirzepatide users without diabetes. A November 2025 TriNetX study of 130,000 patients found tirzepatide specifically was associated with a 44% higher risk of osteoporosis or fragility fracture compared to other GLP-1 medications. The amount of bone loss correlates directly with the amount of weight lost — more weight loss, more bone loss.

The nuance: The bone loss appears to be driven primarily by the rapid weight loss rather than a direct drug effect. This means it is largely preventable with the right protective strategies. A 2024 JAMA Network Open study found that combining exercise with GLP-1 therapy preserved bone density at the hip, spine, and forearm despite larger weight loss — whereas GLP-1 therapy without exercise led to measurable reductions.

What to do: Resistance training 2–3 times per week, adequate calcium (1,000–1,200mg daily), vitamin D (600–1,000 IU), and protein (minimum 1.2g/kg/day). A baseline DEXA scan is recommended for postmenopausal women, people over 65, and anyone with prior fractures or existing osteopenia.

What it is: A 2024 study from the University of Alberta found that semaglutide reduced the size of individual heart muscle cells and overall cardiac mass in both lean and obese mice — independently of weight loss. The same effect was observed in cultured human heart cells.

What the evidence shows: The Alberta study, published in JACC: Basic to Translational Science in October 2024, found an approximately 8% reduction in skeletal muscle in lean mice and measurable reductions in heart muscle cell size across both groups. Critically, no changes in heart function or wall thickness were observed. The researchers cautioned that their findings may not translate to humans and called for evaluation of cardiac structure in ongoing clinical studies.

The nuance: This finding needs to be placed in context. Multiple large human clinical trials — including the STEP-HFpEF programme and the SELECT trial — have found that semaglutide improves cardiac remodelling in people with heart failure and reduces major cardiovascular events by 20% in people with obesity. The animal finding is a potential concern worth monitoring in clinical studies — but it does not override the substantial cardiovascular benefit evidence in humans.

What to do: Resistance training protects both skeletal and cardiac muscle and is the most practical protective intervention available. People with existing cardiomyopathy or heart disease should discuss this specific finding with their cardiologist. For most people using GLP-1 therapy, this remains a monitoring signal, not a clinical action signal.

What it is: Reports to pharmacovigilance databases of depression, anxiety, and suicidal ideation in GLP-1 users, particularly with semaglutide and liraglutide, prompted formal reviews by the FDA and EMA beginning in 2023.

What the evidence shows: The picture here is genuinely complex and requires careful interpretation.

Findings suggesting a signal: A 2025 multinational VigiBase pharmacovigilance study of over 2 million reports found significant statistical signals for anxiety, depressed mood disorders, and suicidality with semaglutide specifically. A separate 2025 meta-analysis found a reporting odds ratio of 1.45 for suicidal ideation with semaglutide — a modest but statistically significant elevation. The risk appeared substantially higher in patients also prescribed antidepressants or benzodiazepines.

Findings against a causal link: The FDA reviewed the evidence in January 2024 and stated its preliminary evaluation found no evidence that GLP-1 medications cause suicidal thoughts or actions. The EMA similarly found no causal link after reviewing approximately 150 case reports. A large Nature Medicine 2024 cohort study of semaglutide users found a lower risk of suicidal ideation compared to non-users — a hazard ratio of 0.27. Multiple other large cohort studies also found no increased risk or reduced risk compared to alternative medications.

The nuance: The pharmacovigilance signals and the cohort study results point in opposite directions. This may reflect confounding — people with obesity and diabetes have higher baseline rates of depression and suicidal ideation than the general population, which can distort adverse event reporting systems. The randomised trial data, which controls for this confounding, has consistently shown no increased psychiatric risk.

What to do: The current clinical guidance is for healthcare providers to monitor patients for new or worsening depression, unusual mood changes, or suicidal thoughts. If you notice these changes after starting or increasing GLP-1 therapy, discuss them with your prescriber. The evidence does not support stopping GLP-1 medications due to psychiatric concerns for the general population without a specific clinical indication to do so.

What it is: A slightly elevated rate of gout (painful joint inflammation caused by uric acid crystal deposits) in people using GLP-1 medications, identified in the same 2025 AAOS study that found the bone density signal.

What the evidence shows: Among nearly 150,000 adults studied over five years, gout rates were 7.4% in GLP-1 users versus 6.6% in non-users — approximately a 12% increase in relative risk. The researchers believe this is linked to rapid weight loss, which temporarily elevates uric acid levels as tissues break down. This is a known phenomenon with rapid weight loss generally — it is not unique to GLP-1 medications.

The nuance: The absolute risk increase is modest. For most people, this is not a reason to avoid GLP-1 therapy. However, if you have a personal history of gout or elevated uric acid, mention this to your prescriber. Staying well hydrated, avoiding alcohol, and not restricting purine-rich foods (meat, seafood) excessively during treatment are practical steps to reduce flare risk.