GLP-1 Drugs Linked to Higher Risk of a Rare Eye Condition Compared to DPP-4 Inhibitors

What the researchers found

At 1 year, GLP-1 RA initiators had 18.5 NAION events per 100,000 people compared to 7.2 per 100,000 among DPP-4 inhibitor initiators. The adjusted risk ratio was 2.56 (95% CI: 1.44-4.86), with an absolute risk difference of 11.3 per 100,000.

The risk was highest during the first 6 months of use and diminished with longer treatment duration. Subgroup analyses showed higher risk in patients under 50 years old, men, ever-smokers, and those with a hemoglobin A1c reduction of 1% or more — suggesting that rapid blood sugar lowering may be a contributing factor.

Why it matters

GLP-1 receptor agonists are among the most widely prescribed drugs in the world, used by millions for diabetes and weight loss. NAION is a rare but serious eye condition that causes sudden, often permanent vision loss. This large-scale study raises an important safety signal that clinicians and patients should be aware of, even though the absolute risk is small.

How the study worked

This was an active-comparator, new-user cohort study that emulated a pragmatic target trial using the UK Clinical Practice Research Datalink. It compared adults with type 2 diabetes who newly started a GLP-1 RA versus a DPP-4 inhibitor. DPP-4 inhibitors were chosen as the comparator because they serve a similar clinical role but have no known association with NAION. Results were adjusted using propensity-score fine-stratification weighting to account for differences between the groups.

What this study cannot tell us

As an observational study, it cannot prove that GLP-1 drugs directly cause NAION — only that there is an association. Despite propensity-score adjustment, residual confounding from unmeasured factors is possible. The absolute number of NAION events was small (14 in the GLP-1 group), which limits the precision of subgroup analyses. The study used diagnostic codes, which may miss some cases or include misdiagnoses. It also could not distinguish between specific GLP-1 RA medications.

How to read the evidence

This is a large, well-designed observational cohort study using real-world medical records from over 500,000 patients with propensity-score adjustment. While it cannot prove causation, the large sample size, active comparator design, and consistent subgroup findings make this strong epidemiological evidence.

When this study was published

Published in 2026, this is a very recent study directly relevant to the ongoing safety evaluation of GLP-1 receptor agonists.

The bigger picture

This study adds to a growing body of research examining potential eye-related side effects of GLP-1 drugs. Previous reports have raised concerns about semaglutide and NAION, particularly after a smaller study from Harvard in 2024. This larger UK-based study provides stronger epidemiological evidence and identifies specific risk factors, contributing to the evolving safety profile of one of the most important drug classes in modern medicine.

Questions still open

• Is the increased NAION risk a direct effect of GLP-1 drugs, or could rapid blood sugar lowering itself be the underlying cause?
• Are certain GLP-1 receptor agonists (e.g., semaglutide vs. liraglutide) associated with different levels of NAION risk?
• Should patients at higher risk (younger, male, smokers) receive additional eye monitoring when starting GLP-1 therapy?

Common questions

Read the original research

Glucagon-Like Peptide 1 Receptor Agonists and Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients With Type 2 Diabetes.

Diabetes care

Citation

Noh, Yunha; Yin, Hui; Ben Ghezala, Inès; Yu, Oriana H Y; Suissa, Samy; Azoulay, Laurent. (2026). Glucagon-Like Peptide 1 Receptor Agonists and Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients With Type 2 Diabetes.. Diabetes care. https://doi.org/10.2337/dc25-2577