How Common Is NAION on GLP-1 Drugs? FDA Numbers and Studies

If you saw the headline and wondered how likely this is, the short answer is that NAION is rare. In the pooled clinical trials the FDA reviewed, the rate was 6.4 cases per 100,000 person-years for people on a GLP-1 drug, compared with 3.8 per 100,000 for people on comparison treatments. Published estimates for adults over 50 in the general population run from 2.3 to 10.2 per 100,000 a year, and the FDA cites those figures as background.

Below you will find the FDA’s trial numbers, the 2024 study behind the headline, a 2024 study that found no increase in the general population, and the symptoms to watch for.

Quick facts, as of 2026-11-10

What happened with the FDA warning

On November 9, 2026, the FDA told patients and health care providers about a “potential increased risk” of non-arteritic anterior ischemic optic neuropathy (NAION) with GLP-1 RA medications. The FDA safety communication says it is requiring makers of all GLP-1 RA medications to add this information to the Warnings and Precautions section of their labels.

CBS News reported that FDA scientists began looking after a 2024 paper tied NAION to certain patients taking semaglutide. Reuters said the new labeling will appear on Mounjaro, Zepbound, Ozempic and Wegovy, among others.

The companies have spoken for themselves. Novo Nordisk told CBS News it continues to believe the data have not shown a cause-and-effect relationship, and that it remains confident in the benefit-risk profile of its GLP-1 medicines. Eli Lilly said it is carefully evaluating the update, that the communication applies to all of its GLP-1 RA medicines, and that it cites no findings specific to Lilly products.

What the warning means for the whole drug class

The FDA calls NAION a rare condition. It says NAION is caused by decreased blood flow to the front part of the optic nerve, and that the underlying cause is not completely understood. The agency also reviewed animal data and found no mechanism for how GLP-1 RAs could increase NAION risk.

Dr. Joseph Rizzo, a neuro-ophthalmologist at Massachusetts Eye and Ear who co-authored the 2024 paper, told CBS News he believes the evidence shows GLP-1 use raises the risk of NAION. He does not discourage patients from taking the drugs. In CBS’s quote, he said: “The medicines have really been profoundly beneficial to a large number of people.”

Symptoms that mean you should call a health care professional right away

The FDA describes NAION as sudden, painless vision loss in one eye, which some patients notice upon waking. Other signs are blurred or missing parts of vision, such as blind spots or tunnel vision, distorted colors, and reduced sharpness in the center of vision. The FDA says to contact a health care professional right away for any of these.

The American Academy of Ophthalmology patient page, published August 13, 2026, covers ischemic optic neuropathy in general. It describes brief darkening of vision that clears as something that can come before the condition, and it says to call an eye doctor or primary care doctor right away.

Can NAION be treated?

There is no known treatment that reverses NAION vision loss. ABC News reported that point on November 9, 2026. The AAO and NANOS AAO and NANOS consensus statement says there is no known effective treatment, and that one-third of patients may have mild spontaneous improvement, though vision loss is often permanent.

Rizzo told CBS News that the vision loss tends to be irreversible and can keep getting worse even after a patient stops the medication. The degree of loss in the first eye does not predict the degree in the second, so a mild loss in one eye could be followed by a severe loss in the other. The consensus statement says up to one-quarter of patients may develop NAION in the other eye, and the FDA says a NAION event in one eye raises the risk for the other.

The FDA’s trial numbers, explained

The FDA’s analysis pooled 126 randomized controlled trials with 134,253 patients: 78,217 on GLP-1 RAs and 56,036 on a comparator. Those trials added up to 288,879 person-years of follow-up. Fifteen probable NAION events occurred. Ten were on GLP-1 RAs, five were on placebo, and none were on other non-GLP-1 glucose-lowering drugs.

The FDA reports 6.4 NAION cases per 100,000 person-years on GLP-1 RAs, compared with 3.8 per 100,000 on comparators.

What “per 100,000 person-years” means in plain words

A person-year is one person followed for one year. The 288,879 person-years is the total time all trial participants were followed, added together. A rate per 100,000 person-years counts the cases against that total time. In plain terms, 6.4 means about six cases for every 100,000 years of patient time on a GLP-1 RA.

How does that compare with the general population?

The FDA cites published general-population incidence of 2.3 to 10.2 per 100,000 persons per year in adults older than 50. The AAO and NANOS consensus statement gives the same range. Both trial figures, 6.4 and 3.8, fall inside that range.

What the FDA’s Sentinel cohort found

The FDA’s Sentinel cohort followed adults with type 2 diabetes who started a GLP-1 RA or another diabetes drug between 2019 and 2024. NAION incidence was 26 to 32 per 100,000 person-years across the groups. It was 30 to 32 for GLP-1 RAs, 31 for SGLT-2 inhibitors and 26 for DPP-4 inhibitors.

Compared with SGLT-2 inhibitors, the hazard ratio was 1.19 (95% CI 0.91 to 1.54). The FDA says that is not a statistically significant increase. Compared with DPP-4 inhibitors, the hazard ratio was 1.43 (95% CI 1.02 to 2.01), which the FDA reports as an increased risk.

A hazard ratio compares how quickly events happened in two groups. The 95% CI in brackets is the range where the true figure most likely falls. When that range includes 1, the difference is not statistically significant.

The 2024 JAMA Ophthalmology study behind the headline

Hathaway and colleagues published their study online in July 2024 in JAMA Ophthalmology, issue dated August 1, 2024. The Hathaway 2024 JAMA Ophthalmology abstract describes a retrospective matched cohort. It used a registry of patients seen by neuro-ophthalmologists at one academic institution, from December 1, 2017 to November 30, 2023. It included 16,827 patients with no history of NAION and compared those prescribed semaglutide with those prescribed non-GLP-1 RA drugs, in two groups.

In the type 2 diabetes group, 710 patients were included: 194 on semaglutide and 516 on other antidiabetes drugs. There were 17 NAION events on semaglutide and 6 on the other drugs. The 36-month cumulative incidence was 8.9% (95% CI 4.5 to 13.1%) on semaglutide, versus 1.8% (95% CI 0 to 3.5%). The hazard ratio was 4.28 (95% CI 1.62 to 11.29).

In the overweight and obesity group, 979 patients were included: 361 on semaglutide and 618 on other weight-loss drugs. There were 20 events on semaglutide and 3 on the others. The 36-month cumulative incidence was 6.7% (95% CI 3.6 to 9.7%) versus 0.8% (95% CI 0 to 1.8%), with a hazard ratio of 7.64 (95% CI 2.21 to 26.36).

The authors wrote that their findings suggest an association between semaglutide and NAION. Because the study was observational, they said future study is needed to assess causality.

Why its rates look so much higher

The study drew on patients seen at one institution’s neuro-ophthalmology practice. The authors of the later 2024 study in Ophthalmology described the earlier study’s patients as referred to a single major medical center in a city with a high degree of medical sophistication. They wrote that differences in population characteristics and drug prescribing between single institutions and global databases may contribute to varying results. The 8.9% and 6.7% figures are cumulative percentages over 36 months, a different measure from the FDA’s rate per 100,000 person-years.

The 2024 study of the general population that found no increase

Chou and colleagues published a population-based real-world study in Ophthalmology, online ahead of print on November 2, 2024. It used the TriNetX global network, with data from about 200 million people across 21 countries. It included adults with type 2 diabetes, with obesity, or with both, and compared semaglutide with non-GLP-1 RA drugs. The groups were 37,245 with type 2 diabetes only, 138,391 with obesity only, and 64,989 with both.

The authors reported no increase in NAION risk with semaglutide in the general population, consistent across the groups at one, two and three years of follow-up. They wrote that avoiding semaglutide based solely on NAION concerns “may not be warranted,” because its potential benefits for blood glucose control and cardiovascular health likely outweigh its potential risks.

The 2025 JAMA Network Open study on semaglutide and tirzepatide

Wang and colleagues published in JAMA Network Open in 2025 (volume 8, issue 8). They used a nationwide, multicenter database of electronic health records covering more than 118 million US patients, and a target trial emulation design. The study included adults with type 2 diabetes and no prior eye diagnoses, with data from December 2017 to January 2023.

It compared semaglutide or tirzepatide with other antidiabetic medications in 159,398 patients, 79,699 in each group. NAION was reported in 35 patients (0.04%) on semaglutide or tirzepatide and 19 (0.02%) on the other drugs. Other optic nerve disorders were reported in 93 (0.12%) and 54 patients.

Which GLP-1 is causing eye issues, and does tirzepatide increase the risk?

The FDA’s labeling change covers the whole GLP-1 RA class, and tirzepatide is included. The FDA notes that tirzepatide is a GIP and GLP-1 receptor agonist, and Reuters named Mounjaro and Zepbound among the products that will carry the new information. FDA’s wording is “potential increased risk,” not a finding that the drug causes NAION.

NAION cases have been reported for all currently marketed GLP-1 RA products. The 2024 Hathaway study looked only at semaglutide. The 2025 Wang study grouped semaglutide and tirzepatide together, so its figures cover the two drugs combined.

Why are there so many GLP-1 medications?

The FDA’s table lists 12 GLP-1 RA products approved between 2005 and 2026. The FDA lists these uses for the class: type 2 diabetes; reducing excess body weight and maintaining weight reduction; reducing the risk of major cardiovascular events in certain adults; reducing the risk of worsening kidney disease in adults with type 2 diabetes and chronic kidney disease; moderate to severe obstructive sleep apnea in adults with obesity; and MASH with moderate-to-advanced liver scarring.

For Byetta, the FDA notes that its application was voluntarily withdrawn, not for safety or effectiveness reasons, per an August 4, 2025 Federal Register notice. Only generic exenatide is marketed.

Where news reports rounded the numbers

Several outlets reported the same FDA analysis with different rounding. Use the FDA’s figures when you quote them, and read a hazard ratio as a ratio, not a count of people. For more on that difference, see our guide to relative and absolute risk in a trial.

Common mix-ups

• Per person-years vs. percentages. The 6.4 and 3.8 figures are rates per 100,000 person-years. The 8.9% and 6.7% figures from the 2024 study are cumulative percentages over 36 months.
• A ratio is not a count. “Four times as likely” comes from a hazard ratio of 4.28, which compares how quickly events happened between two groups.
• Trial rates vs. everyday risk. The 6.4 figure comes from pooled trials. The 2.3 to 10.2 range comes from the general population of adults over 50.
• “Potential” is the FDA’s word. Novo Nordisk told CBS News the data do not establish a cause-and-effect link; Lilly said it is evaluating the update.
• Semaglutide studies vs. the whole class. The 2024 Hathaway study looked only at semaglutide. The FDA label change covers all GLP-1 RA medications.
• Headline vs. numbers. The 2025 Wang study reported 35 NAION diagnoses (0.04%) on semaglutide or tirzepatide and 19 (0.02%) on other antidiabetic drugs.

Who it affects

• Anyone taking a GLP-1 RA medicine. The FDA’s labeling change applies to all of them.
• Adults over 50. The AAO says people 50 and older are more likely to develop ischemic optic neuropathy in general.
• People with the FDA’s listed risk factors. These include smoking, diabetes, high blood pressure, sleep apnea, anemia and kidney disease. The AAO’s list adds high cholesterol, heart disease, glaucoma, migraine and temporal arteritis.
• People who have already had NAION. The FDA says to talk with the health care professional about whether to continue or stop the GLP-1 RA.

Key dates

• December 1, 2017 to November 30, 2023: the registry window for the Hathaway study.
• July 2024: Hathaway et al. published online in JAMA Ophthalmology; issue dated August 1, 2024.
• November 2, 2024: Chou et al. published online ahead of print in Ophthalmology.
• 2025: Wang et al. published in JAMA Network Open, volume 8, issue 8.
• 2025: according to Reuters, the European Medicines Agency’s safety committee said NAION may affect up to 1 in 10,000 people taking semaglutide for at least one year.
• August 4, 2025: Federal Register notice on the Byetta application, as the FDA describes it.
• August 13, 2026: publication date of the AAO patient page on ischemic optic neuropathy.
• November 9, 2026: FDA safety communication, reported the same day by CBS News, Reuters, ABC News and NBC News.

Questions to bring to your prescriber or eye doctor

• Does the new NAION warning change anything about the GLP-1 medicine I take? Your prescriber or pharmacist can walk you through the updated label.
• Do any of my conditions match the FDA’s risk factor list?
• What should I do if I notice sudden, painless vision changes in one eye? Ask your eye doctor to explain the symptoms in person.
• If I have had NAION, should I keep taking my GLP-1 medicine? The FDA says that is a conversation to have with your health care professional.

If you have a problem with a medicine, you can report it to FDA MedWatch online or by calling 1-800-332-1088. For more on how these medicines fit into weight and metabolic care, see our weight loss and metabolic health guide.

One study to watch is a registered ClinicalTrials.gov registration NCT07846956. It plans to review NAION in adults with obesity who do not have diabetes. Its status is not yet recruiting, with 28 participants planned. A registration describes a planned study. It is not a result.

Every number on this page, with its source

By the Clinical Research Explained Editorial Team

This article is for general information purposes only and does not constitute medical advice. Consult your doctor or qualified healthcare provider before making changes to your health routine.