Key Takeaways

• A Swedish nationwide cohort study found GLP-1 receptor agonist users had a higher relative risk of anterior ischemic optic neuropathy than SGLT-2 inhibitor users, but the absolute risk remained very low
• At 1 year, AION occurred in 0.04% of GLP-1 RA users versus 0.02% of SGLT-2 inhibitor users, representing an absolute risk difference of just 0.02%

A large nationwide cohort study from Sweden found that glucagon-like peptide-1 receptor agonist (GLP-1 RA) use was associated with a higher relative risk of anterior ischemic optic neuropathy (AION), which predominantly consists of nonarteritic anterior ischemic optic neuropathy (NAION), compared with sodium-glucose cotransporter-2 (SGLT-2) inhibitors.¹ However, investigators emphasized that the absolute risk remained very low and that much of the observed association may be explained by differences in underlying diabetes severity rather than the medications themselves.

The findings, published in Annals of Internal Medicine, add to the growing body of evidence examining the potential relationship between GLP-1 RAs and ischemic optic neuropathy following reports that raised concerns about a possible safety signal.

Researchers from Karolinska Institutet and collaborating institutions conducted a nationwide, register-based cohort study using Swedish health data collected between 2013 and 2024. The investigators compared 107,518 adults with type 2 diabetes who initiated GLP-1 RA therapy with 185,898 patients who initiated SGLT-2 inhibitors, using propensity score weighting to balance baseline characteristics between groups.

Over a median follow-up of approximately 1.6 years among GLP-1 RA users and 1.5 years among SGLT-2 inhibitor users, AION occurred in 62 GLP-1 RA users and 64 SGLT-2 inhibitor users. At 1 year, the cumulative risk was 0.04% among GLP-1 RA users compared with 0.02% among SGLT-2 inhibitor users, corresponding to an adjusted risk ratio of 1.93. At 5 years, cumulative risks were 0.12% and 0.07%, respectively, with an adjusted risk ratio of 1.69.

Despite the elevated relative risk, the absolute risk differences were small—0.02% at 1 year and 0.05% at 5 years. Importantly, analyses restricted to patients already receiving metformin at baseline, intended to better account for diabetes severity, substantially attenuated the association. In these analyses, the relative risk estimates were no longer statistically significant, suggesting that residual confounding may explain at least part of the observed increase in risk.

The authors noted several limitations, including the small number of AION events, the potential for unmeasured confounding, and limited generalizability to patients using GLP-1 RAs for obesity without diabetes.

The Swedish findings were published alongside a separate U.S. target trial emulation that also reported a modest increase in ischemic optic neuropathy among GLP-1 RA users while emphasizing the rarity of the event and the potential influence of confounding factors. Together, the studies suggest that although clinicians should remain aware of the possible association, the overall risk appears to be low and should be weighed against the established cardiovascular and metabolic benefits of GLP-1 receptor agonists.

Reference

1. Ludvigsson JF, Pasternak B, et al. Glucagon-like peptide-1 receptor agonists and risk for anterior ischemic optic neuropathy: a nationwide cohort study. Ann Intern Med. Published online 2026. doi:10.7326/ANNALS-25-02096.