Highlight
1. GLP-1 receptor agonists (GLP-1RAs) are associated with a higher 18-month risk of ischemic optic neuropathy (ION) compared to sodium-glucose cotransporter-2 inhibitors (SGLT2is) and dipeptidyl peptidase-4 inhibitors (DPP4is) in patients with type 2 diabetes.
2. The absolute risk of ION remains very low, with risk differences of 3.0 and 3.6 per 10,000 patients versus SGLT2is and DPP4is, respectively.
3. Increased risk is most pronounced in men, patients aged ≥50 years, and those with pre-existing cardiovascular or ophthalmic disease.
4. Residual confounding and lack of NAION-specific diagnostic codes limit causal inference.
Background
Nonarteritic anterior ischemic optic neuropathy (NAION) is the leading cause of ischemic optic neuropathy (ION) in adults, accounting for approximately 75% of cases. NAION results from inadequate blood flow to the optic nerve head, causing sudden vision loss, often irreversible. Patients with type 2 diabetes (T2D) are at increased risk due to microvascular dysfunction and comorbidities.
Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have emerged as effective antidiabetic agents with cardiovascular benefits. However, their vascular effects on the microcirculation, including ocular circulation, are not fully understood. Rare but serious ocular adverse events such as NAION have been reported anecdotally, raising safety concerns.
Robust epidemiological data evaluating whether GLP-1RAs increase NAION risk compared with other glucose-lowering agents are limited. Understanding such risks is critical given the expanding use of GLP-1RAs for glycemic control and cardiovascular protection.
Study Design
This investigation used an observational target trial emulation approach leveraging a large U.S. commercial claims database from January 2017 to December 2022. The cohort included adults aged 18 to 65 years with T2D initiating GLP-1RAs, SGLT2is, or DPP4is. The comparator drugs, also second-line antidiabetic agents with favorable cardiovascular profiles, served as active comparators to minimize bias.
The primary outcome was incident ischemic optic neuropathy (ION) identified by diagnostic codes serving as a proxy for NAION, since NAION-specific codes were unavailable. Covariates exceeding 80 variables, including demographics, comorbidities, medication use, and healthcare utilization, were adjusted via inverse probability of treatment weighting to reduce confounding.
The main analysis estimated 18-month cumulative incidence, risk differences (RDs) per 10,000 patients, and numbers needed to harm (NNH) comparing GLP-1RAs to each comparator group.
Key Findings
Among the study population, the 18-month incidence of ION was 8.5 per 10,000 GLP-1RA initiators versus 5.5 per 10,000 SGLT2i initiators, yielding an RD of 3.0 (95% confidence interval [CI], 0.4 to 5.7). Compared to DPP4i initiators, the incidence was 7.8 versus 4.2 per 10,000, indicating an RD of 3.6 (CI, 1.1 to 6.1). Corresponding NNH values were 3333 and 2778 for GLP-1RAs versus SGLT2is and DPP4is, respectively, highlighting a very low absolute risk increase.
The majority (85.2%) of ION events among GLP-1RA users occurred in those aged ≥50 years, and 70.3% were male. Stratified analyses showed higher RDs among men, patients older than 50 years, those with cardiovascular disease, or pre-existing ophthalmic conditions, compared with women and younger patients where differences were minimal.
Patients on metformin monotherapy showed smaller risk increases (RDs 2.0 and 4.1) compared to those on two or more diabetes medications (RDs 5.7 and 4.0), potentially indicating disease severity or polypharmacy effects.
These findings suggest that GLP-1RA use might be linked with modestly elevated ION risk primarily in vulnerable subgroups, though causality remains uncertain due to potential unmeasured confounding such as diabetes duration and body mass index.
Expert Commentary
This study is novel in its use of a large, real-world dataset and rigorous target trial emulation methodology to address an important safety question concerning modern antidiabetic therapies.
The reported association between GLP-1RAs and ION is biologically plausible, as GLP-1 analogs influence vascular tone, endothelial function, and inflammatory pathways that could affect optic nerve perfusion. However, the absolute risk increment is very small, which is reassuring for clinical practice.
Critical limitations include reliance on diagnostic codes lacking specificity for NAION, absence of detailed clinical ophthalmologic findings, and missing data on key confounders such as BMI, duration of diabetes, and smoking status. These factors may induce residual confounding that partially explains the observed associations.
Moreover, differential surveillance or diagnostic bias could inflate identification of ocular events in GLP-1RA recipients who may be medically monitored more intensively.
Despite these caveats, physicians prescribing GLP-1RAs should remain vigilant about potential ocular ischemic events, especially in older male patients with cardiovascular and ophthalmic comorbidities. Prompt recognition and referral for ophthalmologic evaluation is recommended when visual symptoms arise.
Conclusion
In this large emulated target trial, initiation of GLP-1 receptor agonists in adults with type 2 diabetes was associated with a small but statistically significant increase in the risk of ischemic optic neuropathy compared to SGLT2 inhibitors and DPP4 inhibitors. While the absolute risk is very low, heightened awareness is warranted for high-risk patient subgroups.
Given the observational design and residual confounding potential, these findings do not establish causality but underscore the need for further prospective studies with detailed ophthalmic assessments. Clinicians should balance the proven cardiovascular and metabolic benefits of GLP-1RAs against this rare ocular risk during individualized treatment planning.
Funding and Registration
The study was funded by the National Institutes of Health. No clinical trial registration was applicable due to the observational design.
References
1. Reynolds KR, O’Malley KM, Roy JA, Dave CV. Glucagon-Like Peptide-1 Receptor Agonists and Risk for Ischemic Optic Neuropathy: A Target Trial Emulation. Ann Intern Med. 2026 Jul 14. PMID: 42441967.
2. Hayreh SS. Ischemic optic neuropathy. Prog Retin Eye Res. 2009 Jan;28(1):34-62.
3. Marso SP, Daniels GH, Brown-Frandsen K, et al. Liraglutide and Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2016 Jul 28;375(4):311-22.
4. Chen Y, Tao L, et al. Effects of GLP-1 receptor agonists on ocular microcirculation: a review of mechanisms and clinical implications. Diabetes Metab Res Rev. 2023;39(2):e3509.
