Update: September 24, 2026.
Hundreds of patients in the United States are pursuing their claims in court, claiming that commonly used diabetes medications and weight-control supplements may have contributed to the sudden and, in some cases, permanent loss of vision. The disputes primarily focus on non-ischemic anterior ischemic optic neuropathy (NAION), a rare damage associated with insufficient blood flow to the optic nerve.
These lawsuits once again brought attention to some of the most commercially important drugs in the pharmaceutical industry. The lawsuits include products manufactured by Novo Nordisk and a product from Eli Lilly that belongs to a broader group of GLP-1 receptor agonists. The plaintiffs essentially claim that the manufacturers failed to provide sufficient warnings regarding the potential risk of NAION. The companies, however, dispute the claim that a causal link between their drugs and the injuries described in the lawsuits has been established.
The picture resulting from scientific research is much more complex than the simple headline „GLP-1 drugs cause blindness” would suggest—but also more complex than a complete rejection of a potential safety signal.
Several large observational studies have reported a link between semaglutide, the active substance in Ozempic, Wegovy, and Rybelsus, and NAION. Other analyses have shown neither statistically significant increases in risk nor significantly lower values than the first widely-commented study. However, in randomized clinical trials, there were too few confirmed cases of NAION to allow a definitive resolution of this issue based on them.
Regulatory authorities have therefore taken various decisions regarding the information included in product documentation.
In the European Union and the United Kingdom, NAION is currently listed as a very rare side effect of semaglutide, potentially occurring in up to 1 in 10,000 people taking the drug. Australia has adopted broader warnings covering GLP-1 and dual GIP/GLP-1 agonists on the market. In the United States, the current FDA-approved information for Ozempic includes a warning about the complications of diabetic retinopathy, but does not mention NAION.
This difference in the approach of regulators has become one of the most important elements of both the medical discussion and the ongoing court disputes.
This raises a difficult question regarding the safety of pharmacotherapy: if available data suggest a potential risk, the possible outcome may be irreversible, but the absolute risk remains very small. What information should manufacturers, regulatory authorities, and doctors provide to patients?
What does the documentation from the US federal courts reveal?
Disputes regarding the loss of vision should be clearly separated from the much larger group of lawsuits related to GLP-1 drugs and alleged gastrointestinal complications.
The American Judicial Panel on Multidistrict Litigation has created a separate federal MDL 3163 proceeding for claims related to NAION following the use of GLP-1 receptor agonists.
According to an official federal report, as of September 1, 2026, there were 216 unresolved cases under this proceeding. The court indicates that they relate to the products Ozempic, Wegovy, and Saxenda from Novo Nordisk and Trulicity from Eli Lilly. The lawsuits include, among other things, allegations regarding insufficient warnings, product design defects, and warranties.[1][2]
The separate federal lawsuit, MDL 3094, concerns alleged gastrointestinal tract damage related to GLP-1 medications. According to data as of September 1, 2026, it encompassed 4,022 unresolved cases.[2]
These two numbers should not be combined and presented as thousands of lawsuits concerning the loss of sight.
The proceedings concern other alleged damages and are being conducted separately.
Additional cases were also brought before state courts. In New Jersey, the state judiciary in November 2025 separated the claims related to gastrointestinal complications from the NAION claims as two separate multi-party proceedings.[3]
The mere creation of the MDL procedure does not mean that the medicine caused a specific damage.
It also does not mean that the manufacturer's warnings were insufficient from a legal point of view, that the product was defective, or that the manufacturer is liable.
Multidistrict litigation is primarily a procedural mechanism that allows for the coordinated conduct of preliminary proceedings in federal judicial cases involving common factual issues.
The individual claims must still be evaluated based on the procedural documents, the documents obtained during the proceedings, the opinions of experts, regulatory materials, scientific evidence, and the individual medical history of the plaintiffs. Cases that are not dismissed, settled through a settlement agreement, or resolved in another way may ultimately end up in individual lawsuits.
The existence of hundreds of cases is therefore an important legal event, but the number of lawsuits alone does not constitute scientific evidence of a causal relationship.
What do the patients say?
The main allegation that appears in many cases is similar: the patient was taking a medication associated with GLP-1, then developed NAION, and had not previously received appropriate warning that the condition could potentially be associated with the treatment.
Some patients describe the loss of vision in one eye, followed by the occurrence of a similar event in the other eye. They argue that prior information about an emerging safety signal could influence their decision to start or continue therapy, especially after the first incident involving vision.
USA Today described the case of a 63-year-old man who was prescribed Ozempic for diabetes in 2023. According to the report, a few months later he lost his sight in one eye, continued treatment without knowing about the developing signal regarding NAION, and then lost his sight in the other eye as well. The material described the consequences for his work, driving ability, and family life.[4]
However, this story must be interpreted in accordance with the nature of the source.
This is an allegation presented in the context of the legal proceedings and media reports.
This is not an independent clinical study, a determination of a causal relationship, or a legally binding court decision.
The lawsuits also allege that the manufacturers knew or should have known about the emerging signal based on sources such as clinical trial data, adverse event reports, medical publications, and regulatory reviews conducted outside the United States.
The producers dispute these allegations.
They also draw attention to an important scientific complication: people treated for diabetes and obesity often already have conditions related to the vascular system, the retina, and the optic nerve.
This initial risk is crucial when assessing the causal relationship.
The recognition of the disease after the start of treatment confirms the temporal sequence, but the temporal sequence alone does not prove that the drug caused the disease in question.
What is NAION?
Non-ischemic anterior ischemic neuropathy of the optic nerve, abbreviated NAION, is an acute injury involving the anterior part of the optic nerve.
The optic nerve transmits visual information from the eye to the brain. If the blood supply to the nerve portion becomes insufficient, damage to its tissue may occur. This can result in a sudden deterioration in vision or the loss of part of the visual field.
NAION usually occurs without pain and often initially affects only one eye.
The condition is commonly referred to as „an eye attack”. Although this wording can make the problem easier to understand, it can also mislead.
NAION is not the same process as the closure of the retinal artery and is not equivalent to a stroke.
It is also not the same as diabetic retinopathy.
Diabetic retinopathy primarily affects the damage to the small blood vessels of the retina associated with diabetes, while NAION is the ischemic damage to the optic nerve.
NAION occurs rarely.
Published estimates of the incidence vary depending on age, population, and diagnostic methods, but the disease is most commonly described in people over 50 years of age.
Recognized or suspected risk factors include diabetes, hypertension, lipid disorders, obstructive sleep apnea, smoking, anemia, nocturnal decreases in pressure, and certain anatomical features of the optic nerve. [5]
This is where one of the main difficulties in studying the relationship between GLP-1 drugs and NAION arises.
Many factors related to NAION are also common among people taking medications for type 2 diabetes or obesity.
Currently, there is no proven treatment that would reliably restore sight lost due to NAION. Clinical management therefore focuses on confirming the diagnosis, excluding vascular causes that may require urgent treatment, monitoring of significant co-morbidities, and assessing the risk to the second eye.[5]
Sudden partial or complete loss of vision is a condition requiring immediate medical attention regardless of the suspected cause.
A person who experiences sudden loss of vision requires urgent clinical evaluation, rather than relying on self-diagnosis to determine whether NAION, diabetic retinopathy, retinal artery disease, neurological disorder, or another cause is responsible.
The medicines are related to each other, but they are not identical.
The term „GLP-1 medications” is a convenient shorthand, but it can obscure important pharmacological differences.
The drugs discussed in this context contain different active substances, exhibit different receptor activity, have different approved indications, dosages, and scientific bases.
Ozempic, Wegovy and Rybelsus contain semaglutide and are produced by Novo Nordisk. Trulicity contains dulaglutide and is produced by Eli Lilly. Saxenda contains liraglutide and is produced by Novo Nordisk.
Mounjaro and Zepbound contain tirzepatide and are produced by Eli Lilly. Tirzepatide activates both the glucose-dependent polypeptide insulinotropic polypeptide (GIP) receptor and the GLP-1 receptor. From a pharmacological point of view, therefore, it differs from a selective GLP-1 receptor agonist.
These are not purely technical differences.
The strongest observational and regulatory data currently available regarding NAION relate to semaglutide.
Evidence regarding a single active substance should not be automatically extrapolated to all medicines belonging to a broader therapeutic group.
Australia in 2026 decided on a broader warning class, however. A previous review conducted by the Australian advisory committee clearly indicated that most of the published data concerned semaglutide, and the available information at the time did not confirm a similar signal for dulaglutide or tirzepatide.[6]
The later Australian approach to product labeling should therefore be understood as precautionary and covering the entire class, and not as evidence that regulators have determined identical, quantitatively defined NAION risks for each of these drugs.
A 2024 study that drew attention to a potential signal
The current discussion gained momentum after the study was published in „JAMA Ophthalmology” in 2024.
The researchers analyzed the documentation of a specialized neuroocularist center and identified 16,827 patients without a previous diagnosis of NAION. The main analyses included 710 people with type 2 diabetes and 979 people with overweight or obesity. [7]
Among participants with diabetes, 17 cases of NAION were reported in patients prescribed semaglutide, compared with six cases among people receiving non-GLP-1 diabetes medications. The estimated hazard ratio was 4.28. [7]
In the analysis of people with overweight or obesity, 20 events were recorded in the semaglutide group and three in the comparison group, giving an estimated hazard ratio of 7.64.[7]
The results attracted considerable attention due to the high values of the relative risk.
However, they require a significant context.
The study was conducted at a specialized neuroophthalmic center. This type of facility naturally receives significantly more patients with complex eye nerve diseases than would be expected in the general population.
The study was retrospective; the classification of exposures was based on prescriptions; the number of NAION cases was relatively small; and the confidence intervals were wide.
The study design also did not completely eliminate the influence of the error associated with referring patients to a specialized center, the differences between individuals receiving individual medications, or other confounding factors.
The authors therefore described the results as indicating a relationship and called for further research.
They did not state that the study proved a causal relationship.
The most plausible interpretation of this study is a significant and relatively strong signal of pharmacovigilance, rather than a definitive estimate of the frequency of NAION in the general population of semaglutide users.
What did larger studies show?
The studies published after the first work from 2024 presented a more moderate, albeit still potentially worrying, picture.
The study published in 2025 in „JAMA Ophthalmology” analyzed data involving approximately 37.1 million people with type 2 diabetes, including 810,390 new users of semaglutide. [8]
In one of the large databases of insurance claims, the incidence of NAION among people taking semaglutide was 14.5 cases per 100,000 people per year.
However, the results differed depending on the way NAION was defined and the drugs used as a comparison.
In several comparisons, no statistical significance was obtained, but an analysis in which the patient was their own control indicated a more moderate increase.
The authors therefore concluded that the potential increase in risk was significantly smaller than suggested by the first study conducted at a specialized center.[8]
Studies based on Nordic registries also showed an increased relative risk.
A Danish cohort of 424,152 people with type 2 diabetes showed that exposure to semaglutide administered once weekly was associated with approximately twice the five-year risk of NAION, although the authors emphasized that the event itself remained rare.[9]
A subsequent analysis of data from Denmark and Norway, comparing semaglutide with SGLT-2 inhibitors, also showed an increased association, while highlighting a small absolute risk [10].
However, not all analyses of data from real clinical practice yielded the same result.
In other studies, no statistically significant association was found in the general population.[11]
Differences regarding the populations studied, comparator drugs, the length of the follow-up period, the methods for identifying NAION, and the methods for controlling confounding factors may partially explain why the estimates are not fully consistent.
Uncertainty also remains in the case of data from randomized clinical trials.
A systematic review and meta-analysis from 2025, comprising 78 randomized studies and 73,640 participants, did not show an overall increase in the incidence of eye diseases or diabetic retinopathy.[12]
However, the analysis found a higher cumulative value of the chance ratio for NAION, with a very wide confidence interval. The sequential analysis of the studies indicated that the available data on NAION are still insufficient to formulate a definitive conclusion.[12]
This is important because randomized studies usually provide better protection against the effects of confounding factors than retrospective observational studies. However, even a randomized study cannot reliably answer the question about a very rare event if only a few cases occurred and the study itself was not specifically designed to detect and independently assess them.
In May 2026, an meta-analysis of observational studies involving people with type 2 diabetes was published in „PLOS Medicine”.
The estimated hazard ratio for semaglutide compared to other treatment regimens that lowered glucose levels was 2.17. [13]
From the perspective of communicating with patients, however, the absolute risk may be more understandable.
The authors estimated about one additional NAION case for every 7,000 people treated with semaglutide over the course of a year.[13]
The certainty of these evidence was assessed as low or very low, primarily because the studies included were retrospective and still remained susceptible to the influence of confounding factors and diagnostic classification errors.
The overall picture of the research can therefore be described cautiously:
Several independent observational datasets repeatedly indicate a higher relative frequency of NAION in people taking semaglutide. However, NAION remains very rare; the estimated incidence varies significantly between studies, and the available data do not allow for a complete separation of the drug's potential effect from differences in the baseline characteristics of the individuals receiving treatment.
Why is it so difficult to establish a causal relationship?
The safety assessment of medicines rarely relies on a single study.
Scientists and regulatory authorities usually check whether different types of data point in the same direction.
Among other things, temporal dependence, repeatability of results in independent datasets, biological probability, the dose-response relationship, information after discontinuation or re-administration of the drug, data from randomized studies, and the possibility of explaining the observed association with other factors may be relevant.
In the case of semaglutide and NAION, some of these components are available, but others remain incomplete.
There is a temporal relationship, because cases of NAION occurred after initiation of treatment.
In many observational databases, increased risk estimates have been obtained.
Regulatory authorities also analyzed spontaneous reports of adverse reactions and other information from pharmacovigilance systems.
However, people taking these medications often have diabetes, obesity, sleep apnea, hypertension, dyslipidemia, and vascular diseases, which complicates the assessment of the risk related to the optic nerve.
The medication may also be prescribed more frequently to patients whose baseline risk profile differs from that of people receiving the comparator medication.
This problem is referred to as confounding by indication, which is interference related to the indication for treatment.
Statistical methods can reduce the impact of such factors, but they are not always able to completely eliminate them.
The next problem is the identification of cases.
Large administrative databases may rely on diagnostic codes for ischemic optic neuropathy, which do not always allow for an ideal differentiation of NAION from other diseases.
Confirmation of the diagnosis by a specialist can increase its accuracy, but studies conducted in specialized centers may involve atypically selected groups of patients.
The rarity of NAION creates an additional statistical problem.
With a very small number of events, adding or removing only a few cases from the analysis can significantly change the estimated hazard ratio.
Randomized studies limit many types of confounding factors, but most GLP-1 studies were never specifically designed to assess NAION. The condition usually did not constitute a primary endpoint for safety requiring systematic ophthalmological evaluation, and there were very few cases.
The absence of a statistically significant signal in randomized studies is therefore not equivalent to evidence that the risk does not exist.
No definitive biological mechanism has also been established.
The researchers considered hypotheses regarding changes in visual nerve perfusion, autonomic nervous system activity, arterial pressure, hydration, and metabolic control.
They remain hypotheses rather than confirmed mechanisms.
It is already known that a rapid improvement in glycemic control can temporarily worsen diabetic retinopathy in some patients. However, diabetic retinopathy and NAION are different diseases.
Data regarding one of them should not be automatically treated as mechanistic evidence regarding the other.
What did the European regulators decide?
In June 2025, the Risk Assessment Committee within the European Medicines Agency’s Pharmacovigilance Oversight Network completed a review that included non-clinical data, information from clinical trials, post-marketing data, and published studies.
The Committee concluded that NAION should be listed as a „very rare” adverse reaction of semaglutide-containing drugs, corresponding to a frequency of up to 1 case per 10,000 users.[14]
European information regarding the products Ozempic, Wegovy and Rybelsus was subsequently updated.
The recommendations indicate that patients should immediately contact a doctor in case of sudden loss of vision or rapidly worsening vision. If NAION is confirmed, European recommendations suggest discontinuation of semaglutide.[14]
The United Kingdom published the corresponding safety recommendation in February 2026.
The Medicines and Healthcare products Regulatory Agency stated that available studies suggest a very rare association.
The MHRA reported three spontaneous UK reports suggesting NAION by August 1, 2025, while approximately 10.2 million semaglutide packs were dispensed over the previous five years.[15]
These two numbers should not be used to calculate a reliable frequency of occurrence.
Systems for spontaneous reporting of adverse reactions primarily serve to detect safety signals. Not every event is reported; the willingness to report may change after the problem is brought to light, and the mere reporting does not prove that the event was caused by the medication.
However, such reports remain part of a broader pharmacovigilance system.
The World Health Organization subsequently publicized the EMA’s findings and drew attention to the need to respond to sudden eye-related symptoms.[16]
Australia has adopted a broader approach that covers the entire class
The Australian Therapeutic Goods Administration initially examined whether the available data confirmed a safety signal for individual drugs.
In the advisory process, it was considered that the evidence is strongest for semaglutide and weaker for dulaglutide and tirzepatide.
In July 2026, however, the TGA announced a broader update to the warnings covering products available in Australia containing semaglutide, dulaglutide, liraglutide, and tirzepatide.[6]
The Australian wording indicates that cases of NAION have been reported in patients treated with products exhibiting an agonist activity against the GLP-1 receptor, and that sudden loss of vision should lead to an ophthalmological evaluation.
The Australian decision therefore reflects a broader precautionary approach.
It should not be interpreted as representing the same risk assessment for semaglutide that European regulators have adopted for individual substances.
This difference highlights an important aspect of pharmacovigilance: regulatory authorities can analyze a similar set of data and yet make different decisions regarding how narrowly or broadly a warning should be formulated.
Why is the American information about the drug different?
The current US information regarding the prescribing of the drug Ozempic, updated in 2026, includes a warning regarding the complications of diabetic retinopathy.
NAION is not listed in the section regarding warnings, precautions, or adverse reactions. [17]
This difference must be described precisely.
This does not mean that American regulators considered the connection to NAION impossible.
However, it means that the publicly available, FDA-approved product information currently does not include a warning regarding NAION, which has been adopted in Europe.
Regulatory authorities can analyze the evolving evidence and reach various conclusions regarding whether the signal is sufficiently substantiated to change the labeling, what wording should be used, whether the warning should apply to a single substance or the entire class, and what clinical recommendations should accompany it.
The American Academy of Ophthalmology has also taken a cautious stance.
The organization recognizes the possibility of a relationship, emphasizes the absolute small risk, and recommends urgent clinical evaluation in the event of sudden loss of vision. Its recommendations also indicate ongoing uncertainty, especially when the decision to treat involves a person who has previously experienced NAION in one eye or has other risk factors for the optic nerve. [18]
The unresolved regulatory question in the USA is therefore more complex than simply determining whether NAION ever occurred following exposure to semaglutide.
The question is whether the totality of the evidence meets the FDA threshold required to change product information, and if so — should the warning apply specifically to semaglutide or to a broader group of drugs.
What do the producers say?
Novo Nordisk stated that data from clinical trials and the post-marketing period did not show a justified possibility that semaglutide causes NAION, and the overall benefit-risk ratio of its drugs remains favorable.
The company, however, agreed to cooperate with European regulators on the required changes to product information. [19]
Eli Lilly also denies any responsibility in disputes regarding GLP-1.
The company's position should be analyzed separately for each product, because its medicines do not contain the same active substance.
Trulicity contains dulaglutide, while Mounjaro and Zepbound contain tirzepatide.
The official description of the federal MDL 3163 proceedings specifically lists Trulicity as one of the Eli Lilly products currently under investigation.[1]
The questioning of manufacturers' liability is not surprising in ongoing product liability proceedings.
It also reflects an important scientific principle.
Temporal causality, spontaneous reporting of an adverse event, or an inflated risk estimate in an observational study do not automatically indicate that the drug caused damage in a particular patient.
However, the reverse relationship is also true.
The uncertainty regarding the causal relationship does not mean that the recurring safety signal is scientifically irrelevant.
One of the reasons for the existence of pharmacovigilance systems is precisely the fact that serious adverse reactions can occur too rarely to be reliably characterized before a drug is used on millions of people.
The most important questions that a pharmacotherapy safety study should ask
Several important issues remain unresolved, and answers to them could significantly strengthen or weaken the suspicion of a connection.
One of the most important questions is whether independently assessed cases still show the same association when rigorous NAION diagnostic criteria are applied. The accuracy of diagnosis is particularly important in the case of a rare disease, because even a small number of incorrectly classified cases can significantly change risk estimates.
It is also necessary to determine whether the risk changes depending on the dose, the duration of exposure, the rate of dose increase, or the indication for treatment. A repetitive relationship between exposure and risk could provide important information regarding causality.
The next fundamental question is whether the signal is indeed specific to semaglutide or whether reliable data will emerge over time for dulaglutide, liraglutide, tirzepatide, or other drugs that show activity against the GLP-1 receptor.
Potential modulators also require further investigation. It should be determined whether rapid weight loss, dehydration, decreases in nocturnal blood pressure, or rapid improvement in glycemic control can affect susceptibility to NAION.
The baseline ocular and vascular risk may also be important. People with obstructive sleep apnea, a specific structure of the optic nerve, previous NAION in one eye, or numerous vascular risk factors may have a baseline risk that is different from that of the average patient.
The next question concerns the original clinical trial programs. It is necessary to determine whether sufficiently standardized ophthalmological information was collected in them to enable the detection or reliable exclusion of such a rare event.
Legal disputes also raise historical questions.
When did each of the producers for the first time identify a potential signal in their own data?
What analyses were carried out before and after the publication of the 2024 study?
What information was provided to the FDA, EMA, MHRA, TGA and other regulatory authorities, and when was this done?
Answers to these questions require documentation and information regarding communication with regulators, and not just inference based on published epidemiological studies.
More precise comparisons of the observed NAION incidence with appropriately selected background values in comparable populations are also needed.
Regulators and doctors must also consider whether a precisely formulated warning could improve informed decision-making without implying that a causal relationship involving an entire class of drugs has already been proven.
Some of these questions may be clarified by future epidemiological studies, prospective registries, data from clinical trials, and regulatory documentation.
Others may be examined based on materials disclosed in the course of legal proceedings.
Until the relevant documents are made public, claims regarding what the producer knew at a particular time should be clearly presented as allegations rather than established facts.
What can be stated with a high degree of certainty at present?
Based on the data available on September 24, 2026, several conclusions can be drawn.
NAION is a real, rare, and potentially irreversible damage to the optic nerve.
Several observational studies have shown a link between semaglutide and NAION.
The observed association is not the same as evidence that semaglutide caused damage in a specific patient.
European and British regulators have issued warnings specifically regarding semaglutide, while Australia has adopted broader precautionary wording. The current US product information remains the same.
At the same time, federal and state courts are conducting proceedings, but the allegations themselves contained in the lawsuits do not prove the responsibility of the producers.
It would therefore not be correct to state that it has been proven that GLP-1 medications as a whole cause blindness.
However, it would also be incomplete to state that there is no significant safety signal.
The most justified formulation at present is narrower:
Semaglutide was associated with a very small absolute increase in the risk of NAION in several observational analyses. The magnitude and causal nature of this association remain uncertain; regulatory authorities have adopted different approaches to warnings, and scientific and legal questions regarding product-specific liability remain under investigation and litigation.
Frequently asked questions
Has it been proven that Ozempic and other GLP-1 medications cause blindness?
No. Studies have not shown that GLP-1 medications as a whole cause blindness. In several observational studies a link between semaglutide and NAION has been found, and European regulators currently classify NAION as a very rare side effect of semaglutide.[14] However, the absolute frequency of events remains low, risk estimates vary between studies, and a significant portion of the evidence comes from observational studies. Data on other active substances are less developed and should not be automatically treated as equivalent to data on semaglutide.
How is NAION different from diabetic retinopathy?
NAION is an ischemic optic nerve injury that often presents with sudden, painless loss of vision or part of the visual field in one eye. Diabetic retinopathy involves damage to the blood vessels of the retina associated with diabetes and usually has a different biological and clinical course. The current American information regarding semaglutide already includes a warning regarding the complications of diabetic retinopathy in certain patients.[17] However, this is not the same as the warning regarding NAION, and data regarding diabetic retinopathy do not show that NAION is caused by the same mechanism.
How big can the risk be?
The exact magnitude remains uncertain and depends, among other things, on the study design, the population being studied, and the method of identifying NAION. European regulators classify NAION as a very rare side effect of semaglutide, potentially occurring in up to 1 in 10,000 people taking the drug.[14] A 2026 observational meta-analysis estimated approximately one additional case of NAION per 7,000 patients treated with semaglutide annually among people with type 2 diabetes, while assessing the certainty of the evidence as low or very low.[13] High relative risk values can sound alarming for a very rare event, therefore the absolute risk is particularly important in communication regarding the potential scale of the threat.
Which medications are affected by the warning?
Warnings in the European Union and the United Kingdom specifically concern products containing semaglutide, such as Ozempic, Wegovy, and Rybelsus.[14][15] Australia has introduced broader warnings covering GLP-1 and dual GIP/GLP-1 agonists on the market, including products containing semaglutide, dulaglutide, liraglutide, and tirzepatide.[6] The current US information on Ozempic does not mention NAION.[17] Regulatory information should therefore be checked for the specific jurisdiction.
Why didn’t the FDA put out the same warning?
Publicly available information confirms that the current US labeling differs from that of Europe and Australia. However, it does not provide a complete picture of all the FDA internal analyses, communications with manufacturers, or any ongoing regulatory assessments. Agencies may use different evidentiary thresholds and differ in their assessment of whether a warning should apply to a specific substance or a broader group of drugs. The absence of NAION in the current US product information should therefore not be presented as evidence that the FDA has considered the risk to be zero. This merely means that such a warning is not currently included in the approved public product information.
Are there thousands of lawsuits regarding the loss of vision after GLP-1 medications?
Official federal data do not justify describing the current situation as thousands of cases involving vision loss. As of September 1, 2026, MDL 3163, the special federal procedure for NAION, included 216 unresolved cases. [1] [2] Separately, MDL 3094 included 4022 cases involving damage to the digestive tract. [2] There are also cases before state courts, but cases involving the digestive tract and NAION should not be combined and presented as thousands of lawsuits involving blindness.
Do the lawsuits prove that the manufacturers were hiding the risk?
No. Plaintiffs are alleging, among other things, insufficient warning of the risk, while the manufacturers deny liability. Filing a lawsuit or merging cases under the MDL does not mean that the allegations have been proven. Questions regarding internal safety analyses, communication with regulators, labeling decisions, and individual medical histories require evaluation during the evidentiary process, with the participation of experts and as part of the further legal process.
Is there an effective treatment for NAION?
No treatment has been shown to reliably restore the visual acuity lost due to NAION.[5] Clinical management therefore focuses on urgent diagnosis, exclusion of other causes that may require immediate treatment, control of significant co-morbid conditions, and assessment of the risk to the second eye. Sudden loss of vision requires urgent medical attention regardless of whether the person is taking GLP-1 medication.
Should a patient stop taking GLP-1 after reading such information?
Neither the press release nor the information about the legal dispute allows for determining the individual balance of benefits and risks of the treatment. Patients should not routinely change therapy solely based on media reports of legal proceedings. Sudden loss of vision or rapidly worsening vision require urgent medical evaluation. European recommendations indicate that in the event of confirmation of NAION semaglutide should be discontinued. [14] In other situations, decisions should be made jointly with the prescribing physician, taking into account the approved product information applicable in the given country and the individual medical history of the patient.
What data could best resolve this issue?
Significantly stronger evidence would be provided by large, independent studies using validated and best independently confirmed NAION diagnoses, appropriately selected active comparator groups, previously defined statistical analyses, and detailed information regarding dose, treatment duration, changes in glycemic control, and baseline ocular risk. Randomized trials or prospective registries involving standardized assessment of ocular events could further reduce uncertainty. Since NAION is rare, such studies would require very large populations and potentially a long observation period.
Limitations of current evidence
The first significant limitation is that most data suggesting an increased risk of NAION come from observational studies. These studies are of great importance for detecting rare safety events, especially after drugs have been widely introduced, but they remain susceptible to confounding factors.
This is particularly important in this case, because diabetes, obesity, hypertension, sleep apnea, and vascular diseases themselves may be associated with the risk of NAION.
The second limitation is the rarity of this event.
When the number of cases is small, estimates of the relative risk can vary significantly after including or excluding only a few additional events. This may partially explain why the initial estimates were significantly higher than some later results.
The third limitation is the diagnostic classification. Administrative databases provide huge numbers of studied populations, but they may be based on diagnostic codes that do not always provide the same certainty as a specialized ophthalmological evaluation.
The fourth limitation is the small number of NAION cases in randomized studies, which were usually not designed specifically to investigate this event. They therefore cannot currently provide either definitive confirmation or the strong reassurance that would be desirable when assessing the causal relationship.
The fifth limitation is the lack of a validated biological mechanism. Several hypotheses are being considered, but no mechanism has yet been established that convincingly explains how semaglutide would lead to NAION.
The sixth restriction concerns the transfer of applications between medicines.
The strongest data currently pertain to semaglutide. They should not be automatically transferred to dulaglutide, liraglutide, tirzepatide, or future drugs acting on GLP-1-related pathways.
The seventh restriction concerns the interpretation of regulatory decisions.
Different warnings in the European Union, the United Kingdom, Australia, and the United States do not mean that scientific data change with the crossing of the border. They do show that regulatory authorities may interpret evolving evidence differently and adopt different thresholds of caution.
Finally, court proceedings themselves have their own evidentiary limitations.
The lawsuit presents the charges.
The manufacturer's position represents the company's position.
Media reports can provide important information about the experiences of individual people.
However, none of these sources should be considered a substitute for clinical research or a final judgment of a court.
Summary
The evolving body of evidence regarding GLP-1-related drugs and sudden vision loss is much more complex than either alarmist headlines or categorical assurances of no risk suggest.
NAION is a rare but potentially irreversible damage to the optic nerve. In several observational studies, a link between semaglutide and NAION has been identified, and European and British regulators have deemed this signal sufficiently significant to classify NAION as a very rare adverse reaction of semaglutide. [14] [15]
Australia went further, introducing broader precautionary warnings covering drugs on the market that exhibit agonist activity towards the GLP-1 receptor. [6]
The current US information about the drug Ozempic remains different and does not mention NAION. [17]
However, the scientific uncertainty remains significant.
The first widely commented study showed high relative risk values, but it came from a specialized clinical population and included relatively few events. [7] Larger later studies mostly yielded lower estimates, ambiguous statistical results, or data suggesting a more moderate association. [8] [9] [10] [11]
A 2026 observational meta-analysis estimated about one additional case of NAION for every 7,000 people with type 2 diabetes treated with semaglutide over a year, but the certainty of these findings was assessed as low or very low.[13]
Distinguishing between relative risk and absolute risk is crucial here.
A significant relative increase in the risk of a very rare disease may still mean a small absolute increase in the number of additional cases.
At the same time, the rarity of the event does not make the problem insignificant when the potential consequence is irreversible loss of sight.
The legal situation also requires precision.
As of September 1, 2026, the special federal proceedings regarding NAION included 216 unresolved cases, while thousands of separate federal cases concerned damage to the digestive tract. [1] [2] Combining these categories would lead to a misleading picture of the number of lawsuits concerning vision loss.
The mere existence of court proceedings does not prove a causal link or the responsibility of the producers.
Similarly, the manufacturers' denials do not resolve the scientific significance of the observed signal.
The most justified proposal at the current stage therefore remains deliberately narrow:
Semaglutide has been repeatedly associated with a very small absolute increase in the risk of NAION in several observational datasets. European and British regulators now consider NAION to be a very rare side effect of semaglutide; Australia has introduced broader precautionary warnings, while the current US labeling remains unchanged. The exact magnitude of the risk, the biological mechanism, the degree of causal association, and the extent to which a potential signal may affect drugs other than semaglutide remain subject to further study.
Editorial and medical note
The article distinguishes the patients' complaints, the claims presented in court proceedings, the associations observed in the studies, the conclusions of regulatory authorities, the positions of manufacturers, and the established clinical facts.
The text does not determine the cause of the eye damage in any particular patient and does not prejudge the legal liability of Novo Nordisk, Eli Lilly or any other company.
The article is informational and educational in nature and concerns the safety of pharmacotherapy. It does not constitute medical or legal advice.
Anyone who experiences sudden loss of vision, sudden loss of visual field, or rapidly worsening vision should seek medical attention immediately—regardless of whether they are using semaglutide, another GLP-1-related drug, or not using any of these medications.
References
- U.S. District Court for the Eastern District of Pennsylvania. MDL 3163: GLP-1 RAs Non-Arteritic Anterior Ischemic Optic Neuropathy Products Liability Litigation. https://www.paed.uscourts.gov/mdl/mdl-3163-re-glucagon-peptide-1-receptor-agonists-glp-1-ras-non-arteritic-anterior-ischemic-optic
- U.S. Judicial Panel on Multidistrict Litigation. Pending MDL Dockets by Actions Pending, September 1, 2026. https://www.jpml.uscourts.gov/sites/jpml/files/Pending_MDL_Dockets_By_Actions_Pending-September-1-2026.pdf
- New Jersey Courts. GLP-1 Medications—Gastrointestinal Injuries and NAION Vision Loss Multicounty Litigation notices. https://www.njcourts.gov/multicounty-litigation/glp-1-medications-gastrointestinal-injuries/case-information
- Walrath-Holdridge M. Are GLP-1s causing blindness? What we know about rare cases. USA Today. September 21, 2026.
- American Academy of Ophthalmology. NAION: Diagnosis and Management. https://www.aao.org/eyenet/article/naion-diagnosis-and-management
- Therapeutic Goods Administration. GLP-1 RAs and rare, but severe, eye disorders. July 23, 2026. https://www.tga.gov.au/safety/safety-monitoring-and-information/safety-alerts/glp-1-ras-and-rare-severe-eye-disorders
- Hathaway JT, Shah MP, Hathaway DB, et al. Risk of nonarteritic anterior ischemic optic neuropathy in patients prescribed semaglutide. JAMA Ophthalmology. 2024;142(8):732–739. https://doi.org/10.1001/jamaophthalmol.2024.2296
- Cai CX, et al. Semaglutide and nonarteritic anterior ischemic optic neuropathy. JAMA Ophthalmology. 2025. https://pubmed.ncbi.nlm.nih.gov/39976940/
- Grauslund J, et al. Once-weekly semaglutide doubles the five-year risk of nonarteritic anterior ischemic optic neuropathy in a Danish cohort of 424,152 persons with type 2 diabetes. 2024. https://pubmed.ncbi.nlm.nih.gov/39696569/
- Simonsen E, et al. Use of semaglutide and risk of non-arteritic anterior ischemic optic neuropathy: a Danish–Norwegian cohort study. 2025. https://pubmed.ncbi.nlm.nih.gov/40098249/
- Chou CC, et al. Association between semaglutide and nonarteritic anterior ischemic optic neuropathy. 2024. https://pubmed.ncbi.nlm.nih.gov/39491755/
- Natividade GR, et al. Ocular adverse events with semaglutide: a systematic review and meta-analysis. 2025. https://pubmed.ncbi.nlm.nih.gov/40810985/
- Chrzanowski J, Walicka M, Burzyński J, et al. Semaglutide-associated risk of nonarteritic anterior ischemic optic neuropathy in patients with type 2 diabetes: a systematic review and meta-analysis of observational studies. PLOS Medicine. 2026;23(5). https://doi.org/10.1371/journal.pmed.1005064
- European Medicines Agency. PRAC concludes eye condition NAION is a very rare side effect of semaglutide medicines Ozempic, Rybelsus and Wegovy. June 6, 2025. https://www.ema.europa.eu/en/news/prac-concludes-eye-condition-naion-very-rare-side-effect-semaglutide-medicines-ozempic-rybelsus-wegovy
- UK Medicines and Healthcare products Regulatory Agency. Semaglutide: risk of non-arteritic anterior ischemic optic neuropathy. February 5, 2026. https://www.gov.uk/drug-safety-update/semaglutide-wegovy-ozempic-and-rybelsus-risk-of-non-arteritic-anterior-ischemic-optic-neuropathy-naion
- World Health Organization. The use of semaglutide medicines and the risk of NAION. June 27, 2025. https://www.who.int/news/item/27-06-2025-27-06-2025-semaglutide-medicines-naion
- U.S. Food and Drug Administration. Ozempic (semaglutide) Prescribing Information. 2026. https://www.accessdata.fda.gov/drugsatfda_docs/label/2026/209637s038lbl.pdf
- American Academy of Ophthalmology. Glucagon-Like Peptide-1 Receptor Agonists and the Risk of NAION. 2026. https://www.aao.org/education/clinical-statement/glucagon-like-peptide-1-receptor-agonists-risk-of-
- Reuters. Novo’s Ozempic linked to rare cases of serious eye disorder, EU regulator says. June 6, 2025.