Semaglutide NAION risk is small in absolute terms, and causation is not proven. Several large observational studies link semaglutide to a modest increase in non-arteritic anterior ischemic optic neuropathy (NAION), a sudden, usually painless loss of vision in one eye. In one large cohort, the gap worked out to about one extra case per 10,000 person-years. Sudden vision loss in one eye needs same-day, in-person eye care.
Key takeaways
- NAION is reduced blood flow to the front of the optic nerve, and it causes sudden, usually painless vision loss in one eye.
- The small 2024 study that first raised the signal reported large relative increases (JAMA Ophthalmol, 2024), and a much larger 2025 study found a smaller increase than first reported (JAMA Ophthalmol, 2025).
- In a Danish cohort, unadjusted NAION rates were 2.19 versus 1.18 cases per 10,000 person-years with semaglutide versus SGLT-2 inhibitors (another class of diabetes medication), roughly one extra case per 10,000 person-years (Diabetes Obes Metab, 2025).
- The evidence is observational and mixed, and an association is not proof that semaglutide causes NAION.
- Sudden vision loss in one eye while on a GLP-1 means urgent, in-person eye or emergency care the same day.
What is NAION, and how is it different from other GLP-1 vision problems?
NAION is a stroke-like event in the optic nerve. Blood flow to the front of the nerve drops, and vision in one eye is lost suddenly, usually without pain.
The front of the optic nerve, where it enters the back of the eye, is called the optic disc. In NAION, that area is starved of blood. The underlying cause is not completely understood. People often describe the change as sudden blurring, cloudiness or a missing section of vision in one eye. It usually affects one eye at a time.
It matters because the damage can last. Vision lost to NAION often does not fully return.
A GLP-1 receptor agonist (GLP-1 RA) is a medication that mimics glucagon-like peptide-1, a gut hormone involved in appetite and blood sugar. Semaglutide and tirzepatide are both in this class. NAION differs from two more familiar vision complaints in people taking these medications:
- Fluctuating blurriness. Shifting blood sugar can change how the lens focuses. This tends to come and go, often in both eyes. It is worth raising with a clinician, but it is a different pattern. Our guide to GLP-1 side effects and when to escalate covers the common ones.
- Diabetic retinopathy. This is damage to the blood vessels of the retina, the light-sensing layer at the back of the eye. It is a separate condition with its own monitoring.
The red flag is sudden loss of vision in one eye, including partial loss.
How rare is semaglutide NAION? Relative versus absolute risk
NAION on semaglutide is rare. In the Danish cohort, where the unadjusted rate was close to double, that meant about one extra case per 10,000 person-years.
Relative risk tells you how much higher a risk is compared with another group. Absolute risk tells you how often the event actually happens. Doubling a very small number still gives a small number. A person-year is one person followed for one year, so 10,000 person-years could be 10,000 people followed for a year.
| Study | Population | NAION rate with semaglutide | Comparison |
|---|---|---|---|
| Danish–Norwegian cohort, 2025 | Adults with type 2 diabetes starting treatment | 2.19 per 10,000 person-years (Denmark, unadjusted) | 1.18 with SGLT-2 inhibitors |
| OHDSI network study, 2025 | 37.1 million people with type 2 diabetes | 14.5 or 8.7 per 100,000 person-years | Depends on how NAION was defined |
The Danish–Norwegian study compared people starting semaglutide with people starting SGLT-2 inhibitors, another class of diabetes medication. Comparing two groups who both start an active treatment is called an active-comparator design, and it reduces some of the bias of comparing users with non-users. The study found higher risk with semaglutide, but the authors stressed that absolute risk remains low (Diabetes Obes Metab, 2025). It counted 32 NAION events among semaglutide users across both countries. The gap in the unadjusted Danish rates works out to about one additional case per 10,000 person-years.
The OHDSI (Observational Health Data Sciences and Informatics) network study pooled 14 health databases. Among new semaglutide users, NAION occurred at 14.5 per 100,000 person-years with a broad, sensitive definition and 8.7 with a strict, specific one. That is roughly one to one-and-a-half cases per 10,000 person-years (JAMA Ophthalmol, 2025). Those figures are total rates among semaglutide users, not the extra risk on top of a baseline.
The practical reading: rare enough that most people taking semaglutide will never encounter it, and serious enough that everyone taking it should know the warning sign.
Does semaglutide cause NAION? What the studies can and can't show
Not proven. The studies show an association that shrinks as studies get larger and better controlled, and the authors of the largest study say more research is needed to establish causality.
Where the signal started. A 2024 retrospective matched cohort, prompted by anecdotal experience, compared semaglutide with non-GLP-1 medications in two separate groups (JAMA Ophthalmol, 2024). In the type 2 diabetes cohort there were 17 NAION events with semaglutide versus 6 in the comparator group. Later studies cite a hazard ratio of 4.28 for that cohort, with a 95% confidence interval (CI), the range the true value plausibly falls in, of 1.62 to 11.29. The overweight or obesity cohort reported 7.64 (95% CI 2.21 to 26.36). A hazard ratio compares how fast an event occurs in two groups. Both confidence intervals were wide. Small event counts like these make estimates unstable and can overstate risk. This 2024 study has been cited in regulator coverage as the study that first raised the signal.
Larger studies found smaller effects. In the OHDSI network, risk was higher than with empagliflozin, an SGLT-2 inhibitor, under the strict definition (hazard ratio 2.27, 95% CI 1.16 to 4.46). It was not significantly different from non-GLP-1 drugs under the broad definition. A self-controlled case series compares each person with themselves at different times, which cancels out fixed traits like genetics. That analysis gave an incidence rate ratio of 1.32, and half the contributing databases showed no increase. The authors concluded the risk is small and smaller than previously reported (JAMA Ophthalmol, 2025).
Pooled evidence is inconsistent. A 2026 systematic review and meta-analysis of observational studies found a pooled risk estimate of 1.93 (95% CI 1.22 to 3.08). It also found very high heterogeneity (I² of 91%), meaning the individual studies disagreed a lot (J Clin Pharmacol, 2026). The link was significant in people with diabetes but not in people with overweight or obesity, and the authors described a possible association. A meta-analysis of clinical trials found no significant increase, though it was too small to detect an effect of this size (Diabetes Obes Metab, 2025).
Confounding is hard to remove. Diabetes, high blood pressure and other vascular risk factors raise NAION risk on their own. People prescribed semaglutide often have more of them. NAION is also diagnosed and coded inconsistently in health records.
The fair summary: the signal has appeared in several study designs, it shrinks in larger studies, it is small in absolute terms, and it is not proven causal.
Who has a higher baseline risk of NAION, and what screening covers
Eye anatomy is one of the strongest known risk factors, whether or not someone takes a GLP-1. A small, crowded optic disc, often called "disc at risk," is the classic example.
Other factors that raise baseline risk:
- A previous NAION in either eye
- Smoking
- High blood pressure (hypertension)
- High cholesterol (hypercholesterolemia)
- Diabetes itself
An optometrist or ophthalmologist can assess whether someone has disc-at-risk anatomy. That is not a blanket recommendation for everyone to get an exam before starting. Raise it with your clinician if you have eye concerns or several risk factors. People who have had NAION should talk with their clinician about whether a GLP-1 is appropriate.
Eye history is one part of GLP-1 screening. People with a personal or family history of medullary thyroid carcinoma (MTC) or multiple endocrine neoplasia type 2 (MEN 2), or a history of pancreatitis, should not use these medications, and a clinician screens for this. For the pancreas question in depth, see semaglutide pancreatitis risk. During any intake, mention prior NAION, known optic disc problems or recent sudden vision changes.
Vision symptoms on a GLP-1 that need urgent care
Sudden loss of vision in one eye, including partial loss, needs urgent in-person care the same day. Do not wait it out, and do not handle it through a telehealth message.
- Recognize it. Watch for sudden vision loss in one eye, a missing area of vision, or blurring or cloudiness that comes on suddenly.
- Get seen in person. Go to an ophthalmologist or an emergency department the same day.
- Tell them what you take. Name semaglutide or any GLP-1. Privately prescribed or compounded semaglutide may not appear on every medical record, so tell any clinician you see that you take it.
- Follow clinician guidance on the medication. If NAION is confirmed, your clinician will advise whether to stop semaglutide. Do not stop or change a medication on your own.
Gradual or fluctuating blurriness is a different pattern. It still deserves a conversation with your clinician, but it is not the same emergency.
Does this apply to compounded semaglutide and tirzepatide?
For compounded semaglutide, treat the same symptoms the same way, because it contains the same active ingredient. For tirzepatide, the studies cited here did not examine it, so the evidence here is about semaglutide only.
Compounded medications are not FDA-approved. Compounded semaglutide has the same active ingredient as Ozempic and Wegovy, which are FDA-approved brand products. It is not equivalent to them and is not a generic version. Compounded medications are prepared by state-licensed 503A compounding pharmacies or FDA-registered 503B outsourcing facilities. Our explainer on 503A vs 503B oversight covers the difference.
The studies cited here looked at people prescribed semaglutide in health records. None of them addresses compounded products specifically. There is no basis to assume the risk is lower or higher.
Tirzepatide is also a GLP-1 receptor agonist, but all four studies cited here examined semaglutide only. They cannot tell you whether tirzepatide carries the same, a smaller or no increased risk. If you take tirzepatide, treat sudden vision loss in one eye exactly the same way: same-day, in-person care, and tell the treating clinician what you take.
Questions to ask your clinician about eye risk on a GLP-1
The most useful conversation is short and specific: your eye history, your vascular risk factors and what to do if vision changes.
- Does my eye history, including any prior NAION or known optic disc problem, change whether a GLP-1 is appropriate for me?
- Given my blood pressure, cholesterol, smoking history or diabetes, would an eye exam before starting make sense?
- Which vision changes should send me to in-person urgent care, and which can wait for a routine message?
- If I already see an eye specialist, should I let them know I am starting a GLP-1?
- How should I make sure other clinicians know I take a compounded medication that may not be on my record?
None of these questions has a single right answer for everyone. That is the point of individual clinical review.
How access works through New Blue
New Blue is a technology and administrative services platform. Medical services are provided by independent licensed clinicians; medications are dispensed only if prescribed.
- Intake. The Semaglutide GLP-1 pathway uses an asynchronous intake reviewed by a licensed clinician. Include any eye history. Some treatments or states may require a live visit.
- Clinical review. A licensed clinician decides whether treatment is appropriate. Not everyone qualifies. Eligibility depends on your intake and clinical review, and that's how it should work.
- Pharmacy fulfillment. If prescribed, a licensed partner pharmacy ships the medication directly to you. It is a compounded medication and is not FDA-approved.
Availability varies by state; check state availability. The Semaglutide GLP-1 pathway has delivery limits in some states. The Tirzepatide GLP-1 pathway also has delivery limits in some states.
New Blue is LegitScript certified and serves adults 21 and older. It is not for emergencies, and sudden vision loss needs in-person urgent care. Consultation timing varies by clinician availability. For current pricing, see the Semaglutide GLP-1 pathway page. For the broader research, see the semaglutide evidence summary. If treatment isn't appropriate, refunds follow our refund policy.
Frequently Asked Questions
Is NAION from semaglutide reversible?
NAION can cause lasting vision loss. Vision lost to NAION often does not fully return, and there is no established treatment that reliably restores it. That is why sudden vision loss in one eye needs urgent in-person assessment the same day.
Does NAION risk increase the longer you take semaglutide?
The studies cited here do not answer this. They do not establish whether risk changes with time on treatment or a specific risk window after starting semaglutide. Any sudden vision loss in one eye warrants urgent care, whether treatment started last month or years ago.
Should I get an eye exam before starting semaglutide?
There is no blanket recommendation for everyone. An optometrist or ophthalmologist can assess disc-at-risk anatomy, one of the strongest known risk factors for NAION. Discuss your eye history and risk factors with your clinician to decide whether an exam makes sense for you.
How big is the absolute risk of NAION on semaglutide?
The absolute risk is low. In a Danish cohort of people with type 2 diabetes, unadjusted NAION rates were 2.19 per 10,000 person-years with semaglutide versus 1.18 with SGLT-2 inhibitors, about one extra case per 10,000 person-years (Diabetes Obes Metab, 2025). The study authors noted that absolute risk remains low.
Is NAION the same as diabetic retinopathy getting worse?
No. NAION affects blood flow to the optic nerve and causes sudden vision loss, usually in one eye. Diabetic retinopathy is damage to blood vessels in the retina and is a separate condition with its own monitoring.
Does the NAION link apply to tirzepatide?
The studies cited in this article examined semaglutide only, so they do not show whether tirzepatide carries a similar risk. Tirzepatide is also a GLP-1 receptor agonist. If you take it, treat sudden vision loss in one eye the same way: get same-day, in-person care and tell the clinician what you take.
Is the NAION risk the same for people using semaglutide for weight loss?
The evidence is less clear for weight loss than for diabetes. A 2026 meta-analysis found a significant association in people with diabetes but not in people with overweight or obesity (J Clin Pharmacol, 2026). The 2024 study that raised the signal reported higher risk in both groups, based on small numbers of events.
Sources
- Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients Prescribed Semaglutide. JAMA Ophthalmology, 2024. https://pubmed.ncbi.nlm.nih.gov/38958939/
- Semaglutide and Nonarteritic Anterior Ischemic Optic Neuropathy. JAMA Ophthalmology, 2025. https://pubmed.ncbi.nlm.nih.gov/39976940/
- Use of semaglutide and risk of non-arteritic anterior ischemic optic neuropathy: A Danish-Norwegian cohort study. Diabetes, Obesity and Metabolism, 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC12046482/
- Semaglutide-Induced Nonarteritic Anterior Ischemic Optic Neuropathy: A Systematic Review and Meta-Analysis. Journal of Clinical Pharmacology, 2026. https://pubmed.ncbi.nlm.nih.gov/42612051/
Medical disclaimer: This article is for educational purposes only and is not medical advice. Treatment decisions are made by a licensed clinician based on your intake and clinical review. New Blue Health is a technology and administrative services platform, not a medical provider. If you experience sudden vision loss, seek emergency or urgent in-person eye care immediately.
Related treatment guide
Related reading
Content review & editorial standards
This page is educational content from the New Blue Health Editorial Team. It is checked under the New Blue Health Content Review Policy and Editorial Policy and should not replace individualized medical advice from a licensed clinician. For how we evaluate evidence, see Evidence Methodology and Clinical Sources & References.