GLP-1 Drugs and Rare Vision Loss: Two New Studies, Two Different Conclusions
Two separate studies on GLP-1 drugs and a rare eye condition were published within about a week of each other in late August 2026, and they reached different conclusions on the same specific question. That's not a contradiction to be resolved by picking the "right" one; it's a useful window into how much study design shapes what gets found in the first place.
Key Takeaways
Non-arteritic anterior ischemic optic neuropathy (NAION) is a rare condition caused by reduced blood flow to the optic nerve; it is the second-leading cause of optic-nerve-related blindness and has no effective treatment once it occurs.
A large analysis covering nearly 1.6 million patients across five studies found semaglutide users had a 152% higher relative risk of developing NAION compared with users of other drugs, and a 141% higher relative risk among patients with diabetes specifically.
In absolute terms, that translated to 118 NAION cases per 100,000 semaglutide users overall, and 125 cases per 100,000 among semaglutide users with diabetes, a small fraction of a percent of all users.
A separate propensity-matched cohort study of more than 98,000 patient pairs, published August 27, 2026 in Diabetes, Obesity and Metabolism, found a modest increase in diabetic retinopathy with GLP-1 drugs compared to DPP-4 inhibitors, but no statistically significant increase in NAION specifically.
Physicians interviewed on this topic are consistent that the absolute risk is very low, that GLP-1 drugs have not been recommended to be stopped, and that it remains unclear whether the drugs directly cause NAION or whether the same patients who are good candidates for GLP-1 therapy already carry NAION risk factors like diabetes, high blood pressure, and sleep apnea.
What Is NAION, and Why Does It Matter Here?
Non-arteritic anterior ischemic optic neuropathy, or NAION, happens when blood flow to the optic nerve head is reduced. It’s described as the second-leading cause of optic-nerve-related blindness, can come on suddenly, and currently has no effective treatment once it occurs. It’s more common in people who already have diabetes, high blood pressure, sleep apnea, or a physically crowded optic nerve, all risk factors that overlap heavily with the population most likely to be prescribed a GLP-1 drug in the first place. That overlap is exactly what makes this research hard to interpret cleanly.
What Did the Larger Study Find?
An analysis covering nearly 1.6 million patients across five studies, 682,456 of them semaglutide users and 911,098 users of other medications, found that semaglutide use was associated with a 152% higher relative risk of developing NAION compared with other drugs. Among patients with diabetes specifically, that relative risk increase was 141%. The researchers’ own stated conclusion was direct: semaglutide significantly increases the risk of NAION relative to non-GLP-1 receptor agonist drugs, particularly in patients with diabetes, and they said the findings warrant further investigation and should inform clinical risk-benefit conversations. In absolute terms, though, the numbers are far less alarming than the relative risk percentages suggest on their own: 118 cases per 100,000 semaglutide users overall, and 125 cases per 100,000 among those with diabetes, meaning the substantial majority of users in the study did not develop the condition.
What Did the Diabetes, Obesity and Metabolism Study Find?
Published online on August 27, 2026, this was a propensity-matched cohort study, a design that pairs patients with similar underlying health profiles to isolate the drug’s effect more precisely, covering more than 98,000 matched patient pairs. It found a modest increase in diabetic retinopathy among GLP-1 users compared with users of DPP-4 inhibitors, a different class of diabetes medication used as the comparison group. But on the specific question of NAION, this study found no statistically significant increase at all.
Why Do the Two Studies Disagree?
The two studies weren’t measuring the exact same thing in the exact same way, which is the most likely explanation for the different conclusions. They used different comparison groups (other medications broadly in one case, DPP-4 inhibitors specifically in the other), different study designs (a multi-study analysis versus a single propensity-matched cohort), and different underlying patient populations. This kind of disagreement between studies published within days of each other isn’t a sign that the research is worthless; it’s a sign that a genuinely small, hard-to-detect signal is being measured with different tools that don’t all resolve it the same way. A 2024 study in JAMA Ophthalmology adds another data point to this pattern: it found that semaglutide users with diabetes were more than four times more likely to be diagnosed with NAION than non-users, and among patients who were overweight or had obesity, that risk climbed to seven times more likely, an even larger relative signal than the 2026 findings, from an earlier and differently designed study.
How Should This Risk Actually Be Weighed?
Physicians commenting on this research are consistent on a few points. Bavand Youssefzadeh, DO, an ophthalmologist at Cedars Sinai, notes that most patients who develop NAION already have risk factors like diabetes, high blood pressure, sleep apnea, or a crowded optic nerve, meaning the people most likely to be prescribed a GLP-1 drug for diabetes or obesity may already be more susceptible to NAION independent of the drug itself. He also points to a plausible but unconfirmed mechanism: dehydration, lower blood pressure, or rapid blood sugar changes from these medications could temporarily affect blood flow to the optic nerve. Mir Ali, MD, medical director of a surgical weight loss center, frames the practical takeaway plainly: the complication is very rare, GLP-1 drugs have been used for diabetes far longer than for weight loss, and no one has recommended stopping these medications for patients with diabetes or high blood pressure on the basis of this signal.
The Bottom Line
Two credible studies published within about a week of each other looked at GLP-1 drugs and NAION and reached different conclusions, one finding a statistically significant increased relative risk with real but small absolute numbers, the other finding no statistically significant NAION link at all while still flagging a modest retinopathy signal. Both can be true at once given their different designs and comparison groups. The consistent thread across physicians discussing this research is that the absolute risk remains very low, the mechanism isn’t confirmed, and the patients most likely to be prescribed these drugs may already carry NAION risk factors independent of the medication itself. This is exactly the kind of finding that deserves continued monitoring rather than either dismissal or alarm.
Comments