GLP-1s and Vision Loss: What the NAION and Blindness Studies Show
Headlines about Ozempic and "blindness" have been circulating since mid-2024. In November 2026 the story moved from research papers to drug labels: the U.S. Food and Drug Administration announced that every approved GLP-1 receptor agonist must carry a warning about a rare eye condition called NAION.
This guide explains what NAION is, what the major studies found, how big the risk looks in absolute terms, what is known for each drug (semaglutide, tirzepatide, liraglutide, retatrutide, and others), and the separate, older issue of diabetic retinopathy. It is educational context, not medical advice, and it is not a reason to start, stop, or change any medicine on your own.
If you notice sudden vision loss or a sudden change in vision in one eye, treat it as urgent and get medical care right away.
The Short Version
- What changed: On November 9, 2026, FDA required all approved GLP-1 receptor agonists, including Ozempic, Wegovy, Rybelsus, Mounjaro, Zepbound, Victoza, Saxenda, Trulicity, and Foundayo, to add NAION to the Warnings and Precautions section of their labels. (FDA Drug Safety Communication)
- How rare: In FDA's pooled analysis of 126 randomized trials, NAION occurred at 6.4 cases per 100,000 person-years on GLP-1 medicines versus 3.8 per 100,000 person-years on comparators. Europe's regulator classifies NAION as a "very rare" side effect of semaglutide, meaning up to 1 in 10,000 people.
- How certain: Most of the evidence is observational, and results are mixed. The 2026 joint statement from the North American Neuro-Ophthalmology Society and the American Academy of Ophthalmology calls the semaglutide link "possible," likely around a 2-fold relative increase if real, with causation not established.
- Who was studied: Most data come from adults with type 2 diabetes, who already carry higher baseline NAION risk. Data in people using GLP-1s only for weight loss are thinner and less consistent.
- What it is not: NAION is different from diabetic retinopathy, which is a separate, longer-known label warning about temporary retinal worsening when blood sugar drops quickly in people who already have retinopathy.
What Is NAION?
Non-arteritic anterior ischemic optic neuropathy (NAION) happens when blood flow to the front of the optic nerve, the optic disc, drops and the nerve is damaged. It is sometimes described informally as an "eye stroke."
According to FDA, typical symptoms include:
- Sudden, painless vision loss in one eye, often noticed on waking
- Blurred or missing areas of vision, such as blind spots or tunnel vision
- Colors looking washed out or distorted
- Reduced central sharpness, such as blurry or hazy spots
The 2026 ophthalmology consensus statement notes that vision loss from NAION is often permanent, that about one-third of patients may see mild spontaneous improvement, that there is no proven effective treatment, and that up to one-quarter of patients may later develop NAION in the other eye.
NAION was rare and well described long before GLP-1 medicines became popular. FDA cites a background incidence of about 2.3 to 10.2 cases per 100,000 adults over age 50 per year. Known risk factors include diabetes, high blood pressure, cardiovascular disease, metabolic syndrome, obstructive sleep apnea, smoking, anemia, kidney disease, and a small, "crowded" optic disc. Many of those risk factors are exactly why people are prescribed GLP-1 medicines, which is one reason this signal has been hard to untangle.
Timeline: How the Signal Developed
July 2024: The first study. A retrospective study using records from a single academic neuro-ophthalmology service reviewed 16,827 patients. Among 710 patients with type 2 diabetes, those prescribed semaglutide had a higher NAION risk than those on other diabetes medicines (hazard ratio 4.28). Among 979 patients with overweight or obesity, the hazard ratio was 7.64. The authors called it an association that needed further study. Because the patients had all been referred to a neuro-ophthalmology clinic, the absolute rates in this study (8.9% versus 1.8% cumulative incidence in the diabetes group) are far higher than anything seen in the general population, and experts have cautioned that the relative risks are likely overestimated. (Hathaway et al., JAMA Ophthalmology 2024)
2025: Large database studies, with mixed results.
- A network study across 14 databases covering 37.1 million adults with type 2 diabetes, including 810,390 new semaglutide users, found a NAION incidence of 14.5 per 100,000 person-years on semaglutide. Most comparisons against other diabetes drugs were not statistically different; one comparison (versus empagliflozin, using a strict NAION definition) showed a hazard ratio of 2.27, and a self-controlled analysis found a modest increase (incidence rate ratio 1.32). The authors described a "modest increase" in risk. (Cai et al., JAMA Ophthalmology 2025)
- A Danish and Norwegian registry study of 61,377 semaglutide users with type 2 diabetes compared them with people starting SGLT-2 inhibitors. The pooled hazard ratio was 2.81, which translated to about 1.4 additional NAION cases per 10,000 person-years. The authors concluded the absolute risk remains low. (Simonsen et al., Diabetes, Obesity and Metabolism 2025)
- A U.S. health-record study grouped semaglutide and tirzepatide together in 79,699 people with type 2 diabetes and compared them with matched patients on other diabetes drugs. Over two years, NAION occurred in 35 people (0.04%) versus 19 people (0.02%), a hazard ratio of 1.76. The study found no association with papilledema, optic neuritis, or optic atrophy. (Wang et al., JAMA Network Open 2025)
- A separate study in the same database looked at all GLP-1 type medicines in type 2 diabetes. The NAION estimate (hazard ratio 1.26) was not statistically significant, and all-cause blindness was coded less often among GLP-1 users than among matched controls (hazard ratio 0.77). (Ramsey et al., JAMA Network Open 2025)
- Some studies found no increase, and one military health system analysis even reported a lower risk among semaglutide users with diabetes. The 2026 consensus statement summarizes these conflicting results in detail.
June 2025: Europe acts on semaglutide. The European Medicines Agency's safety committee concluded that NAION is a "very rare" side effect of semaglutide (Ozempic, Rybelsus, Wegovy), affecting up to 1 in 10,000 people. It estimated roughly a 2-fold relative increase in adults with type 2 diabetes, corresponding to about one additional case per 10,000 person-years of treatment, and recommended that semaglutide be stopped in the EU if NAION is confirmed. (EMA PRAC, June 2025)
Early 2026: More data in type 2 diabetes. A study across three U.S. insurance databases compared people with type 2 diabetes starting a GLP-1 medicine with people starting an SGLT-2 inhibitor. Over a median of 6.7 months on treatment, the hazard ratio for presumed NAION was 1.85, with an absolute difference of 0.29 cases per 1,000 person-years. (Tesfaye et al., Diabetes, Obesity and Metabolism 2026)
2026: Ophthalmology societies weigh in. The North American Neuro-Ophthalmology Society and the American Academy of Ophthalmology published a joint consensus statement in Ophthalmology. Their conclusions:
- The evidence suggests a possible association between semaglutide and NAION. If real, the risk increase is likely smaller than first reported, in the range of about 2-fold.
- Causation has not been established, and new evidence may change the picture.
- Even a several-fold relative increase is only a small absolute risk difference because NAION is rare.
- Other GLP-1 medicines, especially tirzepatide, have been studied less, and the risk should not be assumed to be specific to semaglutide.
- They did not support an automatic rule to stop semaglutide after NAION, recommending shared decision-making instead, because stopping can carry real health risks for some patients.
- People with a past NAION or vision loss in one eye should be told about the possible risk.
(DeParis et al., Ophthalmology 2026)
November 9, 2026: FDA labels the whole class. FDA reviewed adverse event reports, animal data, published studies, its own trial meta-analysis, and a study in its Sentinel real-world data network. Key findings from the FDA communication:
- Trial meta-analysis: 126 randomized trials, 134,253 patients, and about 289,000 person-years. FDA identified 15 probable NAION events: 10 on GLP-1 medicines and 5 on placebo. That is 6.4 versus 3.8 cases per 100,000 person-years.
- Sentinel study (adults with type 2 diabetes, 2019 to 2024): NAION rates were 30 to 32 per 100,000 person-years on GLP-1 medicines, 31 on SGLT-2 inhibitors, and 26 on DPP-4 inhibitors. There was no statistically significant increase versus SGLT-2 inhibitors (hazard ratio 1.19), and an increase versus DPP-4 inhibitors (hazard ratio 1.43).
- Animal data: FDA did not identify a mechanism by which GLP-1 medicines would cause NAION.
- Conclusion: NAION has been reported with all marketed GLP-1 medicines, and FDA concluded there is a potential increased risk of rare NAION events. Labels must now say so.
FDA advises patients to contact a health care professional right away for NAION symptoms, and anyone who has had NAION to talk with their clinician about whether to continue their GLP-1 medicine. For clinicians, FDA advises prompt ophthalmology referral if NAION is suspected and considering stopping the medicine unless the benefits clearly outweigh the risk.
Putting the Numbers in Perspective
Relative risk and absolute risk tell very different stories here.
- A hazard ratio of 2 sounds dramatic. But when the starting risk is a few cases per 100,000 people per year, doubling it still means the large majority of people on these medicines never develop NAION.
- FDA's trial numbers work out to about 2.6 extra cases per 100,000 person-years. EMA's estimate for semaglutide in type 2 diabetes is about 1 extra case per 10,000 person-years. The Danish and Norwegian study estimated about 1.4 extra cases per 10,000 person-years.
- Those are still real events with potentially permanent consequences, which is why regulators chose to label it.
The main reasons experts remain cautious about causation:
- Confounding by indication: diabetes, obesity, sleep apnea, and high blood pressure raise NAION risk and are also why people take GLP-1s.
- Coding limits: there is no specific diagnosis code for NAION, so many database studies use a broader "ischemic optic neuropathy" code.
- Surveillance bias: people on newer medicines, or after widely publicized studies, may get more eye exams or more reports filed.
- Inconsistent results: comparable studies sometimes find an increase, no change, or even a decrease.
One proposed explanation, still unproven, is that rapid drops in blood sugar may stress a vulnerable optic nerve, similar to how rapid glucose improvement can temporarily worsen diabetic retinopathy.
