Reports of sudden vision loss in people taking semaglutide have led to headlines about Ozempic and blindness. The condition behind them has a name, NAION, and regulators in Europe and the UK have now ruled on it.
This page explains what NAION is, what those regulators concluded, what the U.S. labels say, and what the studies measured. The drug itself is covered in the semaglutide guide.
What NAION is
NAION stands for non-arteritic anterior ischemic optic neuropathy. The UK regulator describes it as reduced blood flow to the front of the optic nerve, the cable that carries vision from the eye to the brain [2].
It “typically causes sudden, painless vision loss, usually in one eye,” often described as “a blurring or cloudiness of vision.” People with type 2 diabetes are already at higher risk, as are smokers and people with high blood pressure or cholesterol.
One of the cohort studies below calls NAION untreatable, and says it often causes “severe and irreversible visual loss in the affected eye” [6]. That is why a rare signal drew attention.
What European regulators concluded
In June 2025 the European Medicines Agency’s safety committee, known as PRAC, finished a review of semaglutide and NAION [1]. It covered non-clinical studies, clinical trials, post-marketing reports and the medical literature.
PRAC concluded that “NAION is a very rare side effect of semaglutide,” meaning it “may affect up to 1 in 10,000 people taking semaglutide.” The finding covers Ozempic, Rybelsus and Wegovy.
The committee said large studies in adults with type 2 diabetes suggest “an approximately two-fold increase in the risk.” That corresponds to about one additional case per 10,000 person-years of treatment.
It recommended adding NAION to the European product information. It also advised that “if NAION is confirmed, treatment with semaglutide should be stopped.”
What the UK regulator said
The UK’s medicines regulator, the MHRA, issued a Drug Safety Update on 5 February 2026 [2]. It states semaglutide “may be very rarely associated with NAION.”
From semaglutide’s first UK approval in 2018 to 1 August 2025, the UK reporting scheme received three reports suggestive of NAION. The regulator estimates roughly 10.2 million packs were dispensed over the past five years.
The MHRA notes that a report “does not mean that it was definitely linked to the medicine.” It also says its expert group noted “conflicting evidence in the literature” but agreed the product information should be updated.
What the U.S. labels say
The Ozempic label effective June 2026 does not contain the words NAION, optic neuropathy or ischemic optic [3]. The Wegovy label effective the same month does not contain them either [4].
The eye warning the Ozempic label does carry is different. Section 5.3 covers diabetic retinopathy, damage to the retina’s blood vessels from diabetes [3].
In a two-year trial in people with type 2 diabetes and high cardiovascular risk, retinopathy complications occurred in 3.0% on Ozempic and 1.8% on placebo. The gap was larger in people who already had retinopathy: 8.2% against 5.2%.
The label explains that “rapid improvement in glucose control has been associated with a temporary worsening of diabetic retinopathy.” It tells patients to contact their physician “if changes in vision are experienced.”
The rest of the label’s warnings are in semaglutide side effects, and its boxed warning is covered in Ozempic and thyroid cancer.
What the studies measured
A 2024 study drew on one U.S. academic eye hospital’s neuro-ophthalmology records [5]. It compared patients prescribed semaglutide with patients on other drugs for diabetes or weight.
Among 710 people with type 2 diabetes, the hazard ratio for NAION was 4.28 (95% CI 1.62 to 11.29). Among 979 with overweight or obesity it was 7.64 (95% CI 2.21 to 26.36).
These were people already referred to eye specialists, and the event counts were small: 17 against 6, and 20 against 3. The authors wrote that “future study is required to assess causality.”
A Danish study then followed all 424,152 people with type 2 diabetes in the country for five years [6]. NAION occurred at 0.228 per 1,000 person-years with semaglutide and 0.093 without.
After adjustment the hazard ratio was 2.19 (95% CI 1.54 to 3.12). The median time from first prescription to NAION was 22.2 months.
A 2026 U.S. study used insurance claims from three databases, matching 482,912 pairs of people with type 2 diabetes [7]. It compared starting a GLP-1 drug with starting an SGLT2 inhibitor, another diabetes drug class.
The GLP-1 group had a higher risk of presumed NAION (hazard ratio 1.85, 95% CI 1.51 to 2.27). The absolute difference was 0.29 extra cases per 1,000 person-years, and the authors called the absolute increase small.
| Study | Who | Comparison | Finding |
|---|---|---|---|
| Hathaway 2024 | Patients seen at one neuro-ophthalmology service | Semaglutide vs other diabetes or weight drugs | HR 4.28 (diabetes); HR 7.64 (overweight or obesity) |
| Grauslund 2024 | All 424,152 adults with type 2 diabetes in Denmark | Semaglutide vs no semaglutide | HR 2.19; 0.228 vs 0.093 per 1,000 person-years |
| Tesfaye 2026 | 482,912 matched pairs with type 2 diabetes, U.S. claims | Any GLP-1 drug vs SGLT2 inhibitor | HR 1.85; 0.29 extra cases per 1,000 person-years |
Reading the evidence together
The relative risks look large, and the absolute risks are small. That is how both regulators put it: roughly double the chance, and about one extra case for every 10,000 years of treatment.
All three studies are observational, so they show an association rather than proof that the drug causes NAION. The two larger ones studied people with type 2 diabetes, who already carry a higher background risk.
Risk also rises with age. The MHRA puts background NAION at 10.2 per 100,000 person-years from age 50 and 14.79 from age 60 [2]. Other considerations for older adults are in GLP-1 medications for seniors.