The US Food and Drug Administration is requiring a new warning about a rare, potentially sight-threatening eye condition across GLP-1 medicines. Its November 9, 2026 safety communication concerns non-arteritic anterior ischemic optic neuropathy, or NAION. The agency describes a potential increased risk, with important uncertainty about its size.
For readers using these medicines for weight management, two points deserve attention together: the trial events were uncommon, and sudden, painless changes in vision still require prompt professional contact. The new warning does not provide a reliable safety ranking between individual medicines.
Which medicines does the FDA action cover?
FDA is requiring application holders to add NAION information to the Warnings and Precautions section of labeling for all GLP-1 receptor agonists. Its product list includes semaglutide medicines such as Wegovy and Ozempic, tirzepatide medicines Mounjaro and Zepbound, and orforglipron, marketed as Foundayo. Tirzepatide acts at both GIP and GLP-1 receptors but is included in this action. Our semaglutide profile and tirzepatide profile provide background on those medicines.
This is a requirement to update safety labeling, not an announcement that the products are being withdrawn. It also does not establish that every revised product label or patient leaflet is already available. The distinction matters when an older leaflet does not yet mention the new communication.
What the small trial counts can tell us
In its pooled analysis of 126 randomized trials, FDA reviewed 134,253 participants and identified 15 probable NAION events. Ten occurred among GLP-1 recipients and five among placebo recipients; none occurred among people receiving the other glucose-lowering comparators.
After accounting for follow-up time, the reported incidence was 6.4 events per 100,000 person-years with GLP-1 treatment versus 3.8 with the combined comparator group. That comparison includes placebo and non-GLP-1 medicines. Calling it a placebo-only rate would misdescribe the analysis.
A person-year means one person followed for one year, or an equivalent amount of observation accumulated across people. These rates are therefore different from a percentage of users who will ever develop the condition. With only 15 probable events across the pooled trials, the figures deserve more caution than their decimal places might suggest. The communication does not supply a precise risk estimate for a particular weight-loss drug, dose or person.
The warning reflects FDA’s wider assessment of trials, observational studies, reports after marketing and animal data. FDA says the animal evidence did not identify a mechanism explaining how GLP-1 use might increase NAION risk. Absence of an identified mechanism does not settle the safety question.
Why the real-world comparisons differ
FDA also studied adults with type 2 diabetes who began treatment during 2019–2024 in its Sentinel commercial healthcare-data network. Compared with SGLT-2 inhibitors, GLP-1 treatment had a hazard ratio of 1.19, with a 95% confidence interval of 0.91–1.54. That interval includes no difference, and FDA reported no statistically significant increase for this comparison.
Against DPP-4 inhibitors, the hazard ratio was 1.43, with a 95% confidence interval of 1.02–2.01, indicating an increased risk in that analysis. Both comparisons belong in the story. Selecting only one could make the evidence appear either more reassuring or more definite than FDA’s full account.
These are diabetes-population findings, not a direct estimate for adults taking a medicine solely for obesity. FDA also notes that NAION risk factors overlap with conditions common among GLP-1 users, complicating interpretation of safety reports. Our guide to randomized and real-world evidence explains why study design and comparator choice affect the answer.
Europe’s earlier finding had a narrower scope
On June 6, 2025, EMA’s safety committee concluded that NAION is a very rare side effect of semaglutide and recommended updating its product information. That announcement named Ozempic, Rybelsus and Wegovy. It was not a class-wide assessment assigning the same risk to every GLP-1 medicine.
The new US action is broader in product scope. Neither announcement allows a reader to conclude that one medicine is a safer substitute for another on the basis of NAION risk alone. Different populations, comparisons and evidence sources must stay separate; the trial-reading guide offers a framework for checking those boundaries.
What symptoms and questions matter now?
FDA describes NAION as reduced blood flow to the front of the optic nerve. It can cause sudden, painless loss of vision in one eye, sometimes noticed on waking. Other listed symptoms include missing or blurred areas of vision, altered colors and reduced central sharpness. Patients should contact a health professional right away for these symptoms, according to the FDA’s patient advice.
For someone who has had NAION, FDA advises discussing with their health professional whether to continue or stop the GLP-1 medicine. That decision requires an individual assessment of benefits and risks. A headline or pooled event rate cannot make it.
Useful questions for a care discussion include whether a previous eye diagnosis was NAION, how the new warning relates to the exact medicine being taken, and whom to contact promptly if vision changes occur. This article checked the FDA and EMA announcements, not the underlying patient data, complete pooled-trial methods or every revised product label. Those limits leave the personal risk estimate unresolved while the need to respond promptly to symptoms remains clear.



