GLP-1 and NAION Vision: What the Evidence Actually Says About Eye Risk

What NAION actually is

NAION stands for non-arteritic anterior ischemic optic neuropathy. It is injury to the front of the optic nerve from a drop in its blood supply. It usually affects one eye, comes on suddenly, and is typically painless: people notice a smudge, a missing patch of visual field, or a general dimming rather than pain or redness.

People who develop NAION tend to be older adults with vascular conditions. The 2024 single-center matched cohort study accounted for sex, age, high blood pressure, type 2 diabetes, obstructive sleep apnea, obesity, high cholesterol and coronary artery disease [3]. Many of these are also common among people prescribed GLP-1 drugs.

What Europe concluded about semaglutide

In June 2025 the European Medicines Agency's safety committee, PRAC, finished a review of NAION in people taking semaglutide. It looked at non-clinical studies, clinical trials, post-marketing surveillance and the medical literature, and concluded that NAION is a very rare side effect of semaglutide, very rare meaning it may affect up to 1 in 10,000 people taking the drug [1].

On the size of the risk: several large epidemiological studies suggested that semaglutide exposure in adults with type 2 diabetes is associated with approximately a two-fold increase in NAION compared with people not taking the medicine, which corresponds to about one additional case per 10,000 person-years of treatment [1]. Data from clinical trials also pointed to a slightly higher risk in people taking semaglutide than in people taking placebo [1]. The committee recommended updating the product information for semaglutide medicines to list NAION at a frequency of very rare [1].

In the United States the position is different. NAION is not listed in the US semaglutide labels. What the Wegovy label does list for the eye is a warning about diabetic retinopathy complications in patients with type 2 diabetes [2].

The 2024 study, in its own terms

The human dataset most often discussed is a retrospective matched cohort study by Hathaway and colleagues, published in JAMA Ophthalmology in 2024 [3]. It used a centralized data registry of patients evaluated by neuro-ophthalmologists at one academic institution between December 2017 and November 2023. From 16,827 patients with no history of NAION, the authors compared people prescribed semaglutide with people prescribed non-GLP-1 drugs in two groups: 710 adults with type 2 diabetes (194 on semaglutide, 516 on other diabetes drugs) and 979 adults with overweight or obesity (361 on semaglutide, 618 on other weight-loss drugs) [3].

In the diabetes group, 17 NAION events occurred among patients prescribed semaglutide versus 6 among those prescribed other diabetes drugs. Cumulative incidence of NAION over 36 months was 8.9% in the semaglutide group and 1.8% in the comparison group [3].

Two features of the design matter for reading those numbers. First, it is a single-center study, and the patients came from a registry of people already seen by neuro-ophthalmologists, so the incidence figures describe that referral population rather than the general population of people taking semaglutide [3]. Second, it is retrospective, so matching could account only for the measured risk factors listed above [3].

Does the same apply to tirzepatide and the other GLP-1 drugs?

The European review covered the semaglutide medicines Ozempic, Rybelsus and Wegovy [1], and the 2024 cohort compared semaglutide with non-GLP-1 drugs [3]. Neither looked at tirzepatide, which the FDA label describes as a dual GIP and GLP-1 receptor agonist rather than a pure GLP-1 agonist [4], or at liraglutide and the other products in the class.

The eye warning that is on the labels

The Wegovy label carries a warning about diabetic retinopathy complications in patients with type 2 diabetes, which have been reported in trials with semaglutide, and it says patients with a history of diabetic retinopathy should be monitored [2]. In the SUSTAIN 6 cardiovascular outcomes trial in type 2 diabetes, retinopathy complications occurred in 3% of semaglutide patients versus 1.8% on placebo [2].

The same type of warning appears in newer labels in the class. The Foundayo (orforglipron) label states that diabetic retinopathy complications have not been studied in patients with diabetic retinopathy and/or macular edema requiring acute treatment, and that patients with a history of diabetic retinopathy should be monitored for progression [5].

What the EMA advises if vision changes

The EMA's statement is specific: if patients experience a sudden loss of vision or rapidly worsening eyesight during treatment with semaglutide, they should contact their doctor without delay, and if NAION is confirmed, treatment with semaglutide should be stopped [1]. A sudden visual field defect, a dark patch, or loss of vision in one eye is the pattern described. There is no equivalent NAION instruction in the US semaglutide labels, where NAION is not listed [2].

Putting the two numbers side by side

The EMA figure and the cohort figure are not comparable. One is a population-level estimate drawn from large epidemiological studies in adults with type 2 diabetes: about one extra case per 10,000 person-years, with a frequency category of very rare [1]. The other is a 36-month cumulative incidence from a single academic center whose patients had been evaluated by neuro-ophthalmologists: 8.9% versus 1.8% [3].

Neither figure gives the risk for any one person. What is established is that NAION is rare and that a major regulator has accepted the association with semaglutide at the very rare frequency [1].