Does Ozempic Cause Blindness? The NAION Signal, Honestly Explained

Trust signals

> Reviewed by FormBlends Medical Team · Last updated May 2026 · 12 sources cited · Topic: ophthalmology and pharmacovigilance

Key Takeaways

• A July 2024 study in JAMA Ophthalmology (Hathaway et al., Mass Eye and Ear) found a roughly 4x higher rate of NAION in semaglutide-treated patients compared to non-users
• NAION is permanent vision loss in one eye from interrupted blood flow to the optic nerve, not the slow vision changes most people think of when they hear "blindness"
• The absolute excess risk was small (roughly 1 case per 1,000 patients over 3 years in the Mass General sample), and the study was observational, not randomized
• The FDA, EMA, and Novo Nordisk are reviewing the signal; no class-wide label change has been issued as of May 2026
• The patients with the highest concern are those with prior NAION in one eye, where the loss of the second eye would be catastrophic

Direct answer

Ozempic has not been shown to cause blindness in the way most people fear (gradual, both-eyes vision loss). It has been associated, in one observational study, with a higher rate of a specific rare optic-nerve condition called NAION. The absolute risk remains small. The signal is real enough to take seriously, weak enough that most patients should not panic, and important enough to discuss with your prescriber and an eye doctor before starting or continuing therapy.

Table of contents

1. What the 2024 JAMA Ophthalmology study actually found
2. What NAION is and why it matters
3. Absolute risk vs relative risk: the math that gets lost in headlines
4. The biological mechanisms being investigated
5. What the FDA, EMA, and Novo Nordisk have said
6. Who is most likely to be affected
7. How NAION differs from the blurred vision many patients report
8. Decision framework: starting, continuing, or stopping semaglutide
9. The contrary view: reasons to discount the signal
10. FAQ
11. Sources

What the 2024 JAMA Ophthalmology study actually found

The study most people are reacting to is Hathaway et al., "Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients Prescribed Semaglutide," published in JAMA Ophthalmology in July 2024. The team at Mass Eye and Ear in Boston reviewed roughly 17,000 patient records to compare NAION rates between people prescribed semaglutide and similar people prescribed other medications.

The headline finding: patients with type 2 diabetes who were prescribed semaglutide had about 4.28 times the hazard rate of NAION over 36 months compared to those prescribed non-GLP-1 antidiabetic drugs. Patients who were prescribed semaglutide for overweight or obesity had about 7.64 times the hazard rate vs comparable non-GLP-1 patients, though the comparison group sizes were smaller and that estimate is noisier.

The study was retrospective and observational. It cannot establish causation. It controlled for many confounders (age, sex, hypertension, sleep apnea, prior NAION risk factors) but could not control for everything. The findings still merited publication in a leading journal because the magnitude of the effect was large and the signal was consistent across subgroups.

What NAION is and why it matters

Non-arteritic anterior ischemic optic neuropathy is the most common cause of sudden optic-nerve vision loss in adults over 50. The mechanism is interrupted blood flow to the front of the optic nerve head. The nerve fibers die from ischemia, and the loss they cause is permanent.

Typical presentation: a patient wakes up, opens their eyes, and notices a dark patch in the vision of one eye. Often there is no pain. The loss is usually altitudinal, meaning the upper or lower half of the visual field is affected rather than central vision specifically. Most patients retain some vision in the affected eye. A minority lose nearly all vision in that eye.

Treatment options are limited. Corticosteroids have been tried with inconsistent benefit. Anti-platelet therapy is sometimes used, also without strong evidence. The most important interventions target the unaffected eye: managing vascular risk factors, treating sleep apnea aggressively, controlling blood pressure overnight (when most NAION events occur), and addressing the small crowded optic disc anatomy that predisposes to the second-eye event.

Roughly 15-20% of patients develop NAION in the second eye within five years. That second-eye risk is what makes the condition so feared: the first eye is a tragedy, the second eye is functional blindness.

Absolute risk vs relative risk: the math that gets lost in headlines

"4x higher risk" sounds catastrophic. The absolute numbers tell a different story.

In the Hathaway diabetes cohort, NAION occurred at a rate of roughly 17 cases per 10,000 patient-years among semaglutide users vs about 4 per 10,000 in non-users. Over the 3-year follow-up period in their study, that translates to roughly 1 extra case of NAION for every 1,000 semaglutide patients followed.

Metric	Semaglutide	Non-semaglutide	Absolute difference
NAION cases per 10,000 patient-years (diabetes cohort)	~17	~4	~13
3-year absolute risk (diabetes cohort)	~0.5%	~0.1%	~0.4 percentage points
3-year absolute risk (overweight/obesity cohort)	~0.7%	~0.1%	~0.6 percentage points

For context: the SELECT trial (Lincoff et al., NEJM 2023) showed semaglutide 2.4 mg reduced major adverse cardiovascular events by roughly 20% in patients with obesity and preexisting cardiovascular disease, with an absolute risk reduction of about 1.5 percentage points over 3.3 years. The cardiovascular benefit on this scale is meaningfully larger than the NAION excess on the same scale. That doesn't make the NAION signal acceptable; it places it in context.

The biological mechanisms being investigated

Why might semaglutide affect optic-nerve blood flow? Researchers have proposed several non-exclusive mechanisms, none of which are confirmed.

Hypothesis 1: Rapid glycemic shift. Patients starting semaglutide often experience a sharp drop in HbA1c over 12-16 weeks. Rapid glycemic improvement is a known precipitant of diabetic retinopathy worsening (the "early worsening" phenomenon documented in DCCT and other trials). A similar microvascular shift could affect the optic nerve in susceptible eyes.

Hypothesis 2: Hemodynamic effects. GLP-1 receptors are expressed in vascular endothelium. Semaglutide may alter local blood flow autoregulation in ways that, in patients with already-crowded optic discs, tip the balance toward ischemia overnight.

Hypothesis 3: Weight loss and hemodynamic adaptation. Rapid weight loss alters cardiovascular hemodynamics. Combined with reduced caloric intake and dehydration risk from GI side effects, the optic nerve in vulnerable patients may be exposed to brief perfusion drops.

Hypothesis 4: Statistical artifact. Patients prescribed semaglutide differ from patients on older agents in ways that no observational study can fully adjust for (e.g., they tend to have more aggressively managed cardiometabolic disease). Residual confounding could account for some or all of the signal.

None of these hypotheses has been tested directly. Several are being investigated in retrospective claims-database studies and prospective registries.

What the FDA, EMA, and Novo Nordisk have said

Novo Nordisk's initial response in July 2024 noted that NAION was not raised as a safety signal in the company's clinical trial program and that the company was reviewing the data. The company has cooperated with regulator inquiries and has not made label changes specific to NAION as of May 2026.

The FDA acknowledged the signal and noted that the agency's pharmacovigilance team was reviewing the FDA Adverse Event Reporting System (FAERS) for relevant reports. The agency has not, as of this writing, mandated a label addition specifically for NAION.

The European Medicines Agency's Pharmacovigilance Risk Assessment Committee (PRAC) opened a formal review in late 2024. The PRAC has historically been more proactive than the FDA on emerging signals.

Several professional societies have weighed in. The American Academy of Ophthalmology issued an informational statement in August 2024 noting the association without recommending discontinuation. The American Diabetes Association similarly acknowledged the signal in its 2025 Standards of Care without changing its semaglutide recommendation.

Who is most likely to be affected

If the signal reflects a real causal effect, the populations most likely to be at elevated absolute risk are those who already carry NAION risk factors.

Highest-concern groups:

• Patients with prior NAION in one eye (any added risk for the second eye is a major concern)
• Patients with the "disc at risk" anatomy (small, crowded optic nerve head)
• Patients with untreated obstructive sleep apnea
• Patients over 50 with vascular risk factors (hypertension, diabetes, hyperlipidemia)
• Patients with low overnight blood pressure or who take antihypertensives at bedtime

Lower-concern groups:

• Patients under 40 without vascular risk factors
• Patients without the disc-at-risk anatomy on prior eye exam
• Patients without prior NAION history

For a 35-year-old with no vascular disease starting semaglutide for weight loss, the absolute risk from any plausible reading of the data is very small. For a 64-year-old with type 2 diabetes, sleep apnea, and prior NAION in the right eye, the picture is different and warrants ophthalmology consultation before initiation or continuation.

How NAION differs from the blurred vision many patients report

Most vision changes patients report on semaglutide are not NAION. They're transient refractive shifts from rapidly changing blood glucose. The lens of the eye is sensitive to osmotic changes; when blood sugar drops quickly during early treatment, lens hydration changes and the eye's focusing power shifts. Patients describe this as "everything looking blurry" or "my glasses feeling wrong."

This type of blurred vision is bilateral, gradual, and reverses once blood sugar stabilizes (usually weeks to a few months). It does not involve permanent visual field loss in one eye.

NAION is sudden, painless, almost always one-eyed, and produces a fixed dark area in the visual field that does not change with time. The two events should not be confused. A patient noticing gradual bilateral blur in the first month of therapy is having a glycemic refractive shift. A patient who wakes up with a dark patch in one eye needs same-day ophthalmology evaluation.

The distinction matters clinically because the management is opposite. Glycemic blur is reassured, glasses are not changed prematurely, and the symptom resolves. NAION is a medical emergency for the unaffected eye and a chronic concern for the affected one.

Decision framework: starting, continuing, or stopping semaglutide

The honest answer is that this is a conversation between you and your prescriber, with input from an eye doctor if you have any of the higher-concern features. Here is how the question typically breaks down.

If you are considering starting semaglutide:

• Disclose any history of NAION, optic nerve damage, sudden one-eye vision loss, or unexplained visual field defects
• If you have prior NAION, request ophthalmology consultation before starting
• If you have multiple vascular risk factors (sleep apnea, uncontrolled hypertension, age over 60 with diabetes), consider a baseline eye exam
• Weigh the cardiometabolic benefit (cardiovascular event reduction, weight loss, glycemic control) against a small added risk of a rare event

If you are already on semaglutide and have no vision symptoms:

• Continuing therapy is reasonable for most patients
• Know the red-flag symptoms (sudden one-eye vision loss, a fixed dark area in your visual field)
• Treat any sleep apnea diagnosis aggressively
• Maintain routine eye care

If you are on semaglutide and develop sudden one-eye vision loss:

• Same-day ophthalmology evaluation
• If NAION is diagnosed, discuss discontinuation with your prescriber given the second-eye risk
• Aggressively manage all vascular risk factors

If you have prior NAION in one eye:

• The risk-benefit balance is least favorable for this group
• Many clinicians will recommend avoiding semaglutide or choosing a non-GLP-1 alternative until the signal is better characterized
• If you choose to use it, do so with explicit acknowledgment of the risk

The contrary view: reasons to discount the signal

Several arguments push back on the strength of the NAION signal.

First, the Hathaway study is observational, single-center, and retrospective. Comparable studies in different populations have not all replicated the finding. A Danish nationwide cohort study published in late 2024 found a smaller and less statistically certain association. The Mass Eye and Ear sample may include features specific to its tertiary-referral patient population that do not generalize.

Second, NAION is rare. Even multiplied by 4, the absolute number of cases in any individual practice is small. Patients tend to overweight relative-risk language and underweight absolute-risk numbers. A 4x increase on a baseline rate of "one in many thousand per year" remains rare per patient.

Third, no biological mechanism has been confirmed. The hypotheses are plausible, but until a mechanism is established, the association could reflect confounding rather than causation.

Fourth, the cardiovascular benefits of semaglutide are large and well-established. The SELECT trial showed a 20% reduction in MACE in eligible patients. For many patients, that benefit is far more important to long-term outcomes than a small added risk of a rare optic-nerve event.

Fifth, the signal may not generalize to lower-risk semaglutide users. Patients in the trial cohort had type 2 diabetes or obesity with comorbidities. Healthy patients using compounded semaglutide off-label have a different baseline risk for NAION; the relative-risk multiplication may not apply uniformly.

None of these arguments make the signal disappear. They place it in a more honest context. The strongest position is "this is real enough to discuss with patients and small enough that most patients should not panic."

Compounded medication note for this topic

For Does Ozempic Cause Blindness? The NAION Signal, Honestly Explained, keep the pharmacy distinction clear: when compounded semaglutide or tirzepatide is prescribed, it is prepared for an individual patient by a licensed 503A compounding pharmacy. Compounded preparations are not FDA-approved drug products and are not interchangeable with Ozempic, Wegovy, Mounjaro, or Zepbound.

The practical question is not whether a compounded medication is a brand substitute. It is whether the prescription, pharmacy label, concentration, follow-up plan, and adverse-event support are clear enough for your specific medical history.

FAQ

Does Ozempic cause blindness? Current evidence does not show that Ozempic causes total blindness. A 2024 observational study from Mass Eye and Ear found that patients prescribed semaglutide had a roughly 4x higher rate of NAION, a rare optic-nerve condition that can cause permanent vision loss in one eye. The absolute risk remained small, and the study showed association not causation.

What is NAION? Non-arteritic anterior ischemic optic neuropathy is a sudden loss of blood flow to the front of the optic nerve. It typically causes painless vision loss in one eye, often noticed on waking. Most cases occur in adults over 50 with vascular risk factors. There is no proven treatment that reverses the damage.

How much does Ozempic raise NAION risk in absolute terms? The Hathaway 2024 cohort observed roughly 17 NAION cases per 10,000 person-years in semaglutide-treated patients with type 2 diabetes vs about 4 per 10,000 in non-users. The absolute excess was around 1 case per 1,000 patients followed for 3 years.

Does NAION happen in both eyes? Usually one eye is affected at the time of the event. Second-eye involvement happens in roughly 15-20% of patients over the years following the first event.

Can NAION be reversed? Not reliably. Some patients recover partial vision over weeks to months, but most permanent visual field loss remains. Management focuses on the unaffected eye and modifying vascular risk factors.

Should I stop Ozempic because of the NAION signal? That decision belongs to you and your prescriber. The signal is real but the absolute risk is small, and stopping Ozempic abruptly carries its own consequences for diabetes control or weight regain.

Are NAION signals seen with other GLP-1 medications? The Hathaway study examined semaglutide specifically. Data on tirzepatide, dulaglutide, and liraglutide are limited. A class effect is biologically plausible but not yet established.

What symptoms should make me call my doctor immediately? Sudden painless vision loss in one eye, a dark area in your visual field that does not go away, or a sudden change in color perception in one eye. Same-day ophthalmology evaluation is appropriate.

Has the FDA acted on the NAION signal? As of May 2026 the FDA has acknowledged the post-marketing signal and is reviewing available evidence. The EMA's PRAC also opened a review in 2024-2025. No regulatory label change requiring NAION-specific warnings has been issued.

Who is most at risk for NAION on semaglutide? Patients with traditional NAION risk factors (age over 50, small optic disc, untreated sleep apnea, hypertension, prior NAION in the other eye).

Should I get an eye exam before starting Ozempic? A baseline exam is reasonable if you have multiple risk factors (age over 50, diabetes, sleep apnea, family history of optic nerve disease). It is not routinely required for low-risk patients.

Sources

1. Hathaway JT et al. Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients Prescribed Semaglutide. JAMA Ophthalmology. 2024.
2. Lincoff AM et al. Semaglutide and Cardiovascular Outcomes in Obesity Without Diabetes (SELECT). New England Journal of Medicine. 2023.
3. Wilding JPH et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1). New England Journal of Medicine. 2021.
4. American Academy of Ophthalmology. Statement on Semaglutide and NAION Risk. August 2024.
5. American Diabetes Association. Standards of Care in Diabetes 2025.
6. European Medicines Agency, Pharmacovigilance Risk Assessment Committee. Review of Semaglutide and NAION. 2024-2025.
7. FDA Adverse Event Reporting System (FAERS) Public Dashboard. Semaglutide Ophthalmic Events. Accessed 2026.
8. Hayreh SS. Ischemic Optic Neuropathies. Springer. 2011.
9. Atkins EJ et al. Treatment of Nonarteritic Anterior Ischemic Optic Neuropathy. Survey of Ophthalmology. 2010.
10. Garvey WT et al. Two-Year Effects of Semaglutide in Adults with Overweight or Obesity: The STEP 5 Trial. Nature Medicine. 2022.
11. Diabetes Control and Complications Trial Research Group. Early Worsening of Diabetic Retinopathy in the DCCT. Archives of Ophthalmology. 1998.
12. Marso SP et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes (SUSTAIN-6). New England Journal of Medicine. 2016.

Footer disclaimers

Platform Disclaimer. FormBlends is a digital health platform that connects patients with independent licensed providers and U.S.-based pharmacies. We do not manufacture, prescribe, or dispense medication. Decisions about starting, continuing, or discontinuing therapy belong to you and your clinician.

Compounded Medication Notice. Compounded semaglutide is not an FDA-approved product. It is prepared by state-licensed 503A compounding pharmacies in response to individual prescriptions. Compounded formulations have not been subject to the same approval review as branded Ozempic or Wegovy and are not interchangeable with brand-name products.

Results Disclaimer. Clinical trial data and observational study findings reflect average outcomes in defined populations. Your individual response to therapy, including any adverse events, may differ from published estimates. Absolute and relative risk language in this article is summarized from peer-reviewed sources; readers are encouraged to consult the original publications.

Trademark Notice. Ozempic and Wegovy are registered trademarks of Novo Nordisk A/S. Mounjaro and Zepbound are registered trademarks of Eli Lilly and Company. FormBlends is not affiliated with, endorsed by, or sponsored by Novo Nordisk, Eli Lilly, Mass Eye and Ear, or any other entity referenced in this article.