GLP-1 Medications and Blindness: What the Research Actually Shows
By: Mehdia Amini, MD
If you have seen headlines suggesting that Ozempic or other GLP-1 medications can cause blindness, there is some science behind the concern - but the actual story is far more nuanced.
There are two separate eye issues that tend to get mixed together: a rare condition called NAION, which affects the optic nerve, and diabetic retinopathy, which affects the retina.
Neither should be interpreted as meaning that GLP-1 medications commonly cause blindness.
First: What is NAION?
NAION stands for non-arteritic anterior ischemic optic neuropathy. It occurs when blood flow to part of the optic nerve is reduced, which can result in sudden, painless vision loss, usually in one eye.
It is potentially serious, but it is also very rare.
One of the most important pieces of context is that many of the established risk factors for NAION substantially overlap with the population most likely to be prescribed GLP-1 medications. These include:
Type 2 diabetes
Obesity
High blood pressure
High cholesterol
Obstructive sleep apnea
Cardiovascular disease
Older age
This overlap makes the research difficult to interpret. A patient taking semaglutide who develops NAION may already have several independent reasons for being at higher risk of the condition.
In other words, some cases of NAION would be expected to occur in this population even if semaglutide had never been prescribed.
That does not mean a medication effect can be ruled out. It means researchers have to separate the potential effect of the drug from the underlying vascular and metabolic risk of the patients receiving it.
What does the research actually show?
One of the largest studies to date, published in 2025, analyzed data from 37.1 million adults with type 2 diabetes, including more than 810,000 people who started semaglutide (Cai et al., JAMA Ophthalmology, 2025).
Among semaglutide users, NAION occurred at a rate of approximately 14.5 cases per 100,000 person-years - an extremely rare event.
Importantly, the researchers compared semaglutide users with other patients who also had type 2 diabetes and were starting other diabetes medications. When semaglutide was compared with commonly used medications such as empagliflozin, sitagliptin, and glipizide, most comparisons did not show a statistically significant increase in NAION.
That distinction matters. NAION occurred among semaglutide users, but the available data did not consistently show that it occurred significantly more often than it did among comparable patients with diabetes receiving other treatments.
The study did identify a possible signal in some analyses, so the question is still being investigated. But the absolute risk remains extremely low, and the current evidence does not establish that semaglutide directly causes NAION.
Earlier research also raised concern. A 2024 study by Hathaway and colleagues reported more NAION among semaglutide-treated patients than among matched patients receiving other medications. However, the study came from a tertiary neuro-ophthalmology referral population, where eye disease is much more common than in the general population. It therefore raised an important safety signal, but it could not determine whether semaglutide itself was the cause.
Taken together, the research suggests there may be an association worth continuing to study. But the fact that NAION risk factors so strongly overlap with the metabolic conditions for which GLP-1 medications are prescribed is critical when interpreting these findings.
What about diabetic retinopathy?
This is a completely different issue.
The concern comes largely from the SUSTAIN-6 trial, where diabetic retinopathy complications occurred somewhat more frequently in patients receiving semaglutide than placebo.
However, most of the patients who developed complications already had diabetic retinopathy, often had poorly controlled diabetes, and experienced a large reduction in A1c after starting treatment.
This is not a phenomenon unique to semaglutide. For decades, we have known that when blood glucose improves very rapidly, existing diabetic retinopathy can temporarily worsen before it improves.
This phenomenon was demonstrated in the landmark Diabetes Control and Complications Trial. Patients receiving intensive glucose treatment experienced more early worsening of retinopathy during the first several months. But with continued improvement in glucose control, retinal outcomes improved over time and the long-term progression of diabetic retinopathy was substantially reduced.
That long-term perspective is extremely important.
When retinopathy worsens shortly after starting semaglutide, the concern is generally not that semaglutide is directly toxic to the retina.
Rather, in someone with poorly controlled diabetes and pre-existing retinopathy, a rapid improvement in glucose can temporarily destabilize retinal disease.
Over time, however, improved glycemic control is one of the most important ways we protect the eyes and reduce the progression of diabetic retinopathy.
Who should be monitored more closely?
Severe vision complications remain uncommon for the average person taking a GLP-1 medication.
Closer ophthalmologic monitoring may be appropriate in patients with:
Advanced diabetic retinopathy
A very high starting A1c
Longstanding poorly controlled diabetes
Significant obstructive sleep apnea
Multiple cardiovascular risk factors
Patients with significant diabetic retinopathy should remain up to date with their eye examinations, particularly when starting a medication expected to produce a large and rapid improvement in glucose control.
Anyone who develops sudden vision loss, a new dark or missing area of vision, or rapidly worsening vision should seek urgent ophthalmologic evaluation.
In Conclusion
Current evidence suggests a possible association between semaglutide and NAION, but causation has not been established. Observational studies can identify correlation, but they cannot prove that the medication itself caused the event - particularly when many of the patients receiving GLP-1 medications already have independent risk factors for NAION, including diabetes, obesity, hypertension, dyslipidemia, sleep apnea and cardiovascular disease. NAION remains extremely rare, and larger studies have not consistently demonstrated a statistically significant increase compared with other diabetes medications.
Diabetic retinopathy is a separate issue. In patients with pre-existing retinopathy and very poor glycemic control, a rapid drop in A1c can cause temporary early worsening. However, this does not appear to reflect direct retinal toxicity from semaglutide. With sustained improvement in glucose control, the long-term course generally moves in the opposite direction: retinopathy progression decreases, retinal outcomes improve, and the risk of vision-threatening diabetic eye disease is reduced over time.
Schedule with Dr. Mehdia Amini to learn more about GLP-1s or to answer any other endocrinology questions you have using the button below.
References
Cai CX, Hribar M, Baxter S, et al. Semaglutide and Nonarteritic Anterior Ischemic Optic Neuropathy. JAMA Ophthalmology. 2025.
Hathaway JT, et al. Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients Prescribed Semaglutide. JAMA Ophthalmology. 2024.
Marso SP, et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes. New England Journal of Medicine. 2016.
Vilsbøll T, et al. Semaglutide, reduction in glycated haemoglobin and the risk of diabetic retinopathy. Diabetes, Obesity and Metabolism. 2018.
Diabetes Control and Complications Trial Research Group. Early worsening of diabetic retinopathy in the Diabetes Control and Complications Trial. Archives of Ophthalmology. 1998.
