Update: Can GLP1 medications cause loss of vision (NAION)?

While clinical trials of GLP1 medications [semaglutide (Ozempic/Wegovy), tirzepatide (Mounjaro/Zepbound), liraglutide (Victoza/Saxenda), dulaglutide (Trulicity)] had not originally identified a risk of vision loss caused by non arteritic anterior ischemic optic neuropathy (NAION), some database and registry cohort studies have suggested there may be an increased risk.

Since the first publication suggesting a possible increased risk in 2024, what do the data tell us now? Is there a potential increased NAION risk with GLP1s, and if so, how big is the potential risk? And is there anything we can do to identify who might be at risk, or to reduce the risk of NAION occurring?

Background: As blogged previously, NAION is an extremely rare swelling of the optic nerve (the nerve responsible for vision), thought to be caused by decreased blood flow to the nerve. Risk factors for NAION include older age, diabetes, high blood pressure, high cholesterol, cardiovascular disease, sleep apnea, and kidney failure. Retinal (back of the eye) exam findings of an anatomically crowded optic disc, such as a small optic nerve cup and optic disc drusen, are the strongest known risk factors for the development of NAION. NAION is very rare, occuring in 2-10 per 100,000 people per year in the general population. People who have NAION present with a loss of vision without pain, usually in one eye, occurring over the course of hours to days. NAION is usually irreversible with no known treatment.

Since the first registry cohort study suggesting an increased risk of NAION with semaglutide, several additional studies have been published – with conflicting results. As reviewed in a recent article published in The Journal of Endocrinology & Metabolism, almost all of the data studied so far has been in users of semaglutide (Ozempic/Wegovy), as it has been the most widely used medication of all the GLP1s to date. Two national registry data studies from Scandinavia suggested an increased risk of NAION with semaglutide (about 1 case per 10,000 people in a year) in people with type 2 diabetes, as well as a higher risk (about 1.4 cases per 10,000 people in 1 year) with semaglutide compared to people taking SGLT2 inhibitors (a different class of diabetes medication). A systematic review and meta-analysis of 69 studies (52 diabetes studies and 17 obesity studies) did not find a statistically significant difference (though numerically higher – 8 cases with GLP1 medications vs 4 in the comparator groups). A large database study of over 37 million found that people with type 2 diabetes showed an increased risk of NAION with semaglutide (1.5 people per 10,000 per year), but also found a similar risk for people taking other diabetes medications. Three studies using the TriNetX database in USA have not found an increased risk of NAION with GLP1s, but this database censors data when there are less than 10 people affected, so this database may underreport very rare conditions like NAION. A case series of three people developing NAION while on tirzepatide (Mounjaro/Zepbound) have been reported, but no large cohort studies have yet been done to look at NAION risk specifically with tirzepatide.

A recent meta-analysis of 15 GLP1 studies (8 randomized) published in Diabetes Care found an increased risk of NAION of 0.04% vs comparator groups – said another way, for every 2700 people treated with GLP1, one person would develop NAION.

The authors of both papers conclude that the data overall suggests an association between GLP1s and NAION. Both the World Health Organization and the European Medicines Agency issued safety warnings in 2025 highlighting the potential link, potentially affecting 1 in 10,000 users.

So, if the NAION risk is (very low but) real, how could it happen, and is there anything we can do to reduce the risk? While there are no studies or data to give us any clear actions to decrease risk, low blood flow to the optic (eye) nerve is widely regarded to be a key contributing factor to developing NAION. Low blood flow can be caused by low blood pressure. Many people taking GLP1s are on blood pressure medications. BP decreases with the GLP1 medications (especially tirzepatide and semaglutide); as blogged previously, it is extremely important for people on BP medications to watch their BP (ideally with a home BP cuff for close tracking) so that BP medication can be reduced if needed, to prevent low blood pressure. A TriNetX database study suggested that semaglutide can be of benefit to treat idiopathic intracranial hypertension (more on this topic on a future blog post, stay tuned!), and may impact similarly on perfusion to the optic nerve.

Another hypothesis that has been suggested to contribute to NAION risk includes rapid inprovement in blood sugar control in people with diabetes, which isn’t well supported by the existing data. On a related note, there is not evidence to support diabetic retinopathy as a risk factor for NAION. Whether faster weight loss could be a factor has been suggested but not yet studied. As noted above, an anatomically crowded optic disc is the strongest risk factor for NAION; there are no current guidelines to suggest routine retinal screening at this time.

BOTTOM LINE: GLP1 medications may increase the risk of NAION by about 4 cases per 10,000 people. This potential low risk should be taken in the context of the known and important health benefits of GLP1 based medications, including diabetes control, weight loss, cardiovascular benefits, improvement in heart failure, MASH (fatty liver disease), sleep apnea, diabetes prevention, and so on. In my view, watching blood pressure for people taking BP medications, and adjusting these meds as needed to avoid low blood pressure, may be an important preventive strategy. Whether people starting GLP1s should have eye exams to screen for retinal anatomical risk factors is an interesting idea certainly worthy of research.

Disclaimer: I am/have been an investigator in clincal trials of semaglutide, tirzepatide, liraglutide, and dulaglutide. I receive honoraria as a continuing medical education speaker and consultant from the makers of semaglutide and liraglutide (Novo Nordisk), and tirzepatide and dulaglutide (Eli Lilly).

Check me out on X/twitter! @drsuepedersen

Subscribe to this blog! (box upper right)

Share this blog post using your favorite social media link below!