Optic Nerve Stroke Risk Higher With Semaglutide in Type 2 Diabetes
New veteran study suggests relative higher risk of optic nerve injury with semaglutide compared with SGLT2 inhibitors, while absolute risk remains low
In a large Veterans Health Administration study, semaglutide use in type 2 diabetes was associated with a higher relative risk of nonarteritic anterior ischemic optic neuropathy, but the absolute risk over follow-up was under 0.3%, underscoring the need for balanced risk discussions.
Study Details
A new observational study published in JAMA Ophthalmology evaluated the risk of nonarteritic anterior ischemic optic neuropathy, or NAION, among veterans with type 2 diabetes who initiated semaglutide, marketed as Ozempic.
NAION is often described as an “optic nerve stroke.” It typically affects adults over 50 and presents as sudden, painless vision loss due to reduced blood flow to the optic nerve head.
The study compared semaglutide initiators with patients starting SGLT2 inhibitors, most commonly Jardiance. Regulators, including the U.S. Food and Drug Administration, are currently reviewing emerging safety data related to NAION and GLP-1 receptor agonists but have not issued formal warnings.
Methodology
Researchers emulated a target trial within the Veterans Health Administration system from 2018 through 2025.
Key elements:
Population: 102,361 veterans with type 2 diabetes on metformin who newly started semaglutide or an SGLT2 inhibitor
Comparison: Semaglutide vs SGLT2 inhibitor initiation
Follow-up: Median 2.1 years, maximum 7.5 years
Outcome: New-onset NAION
Analysis: Overlap weighting to balance baseline differences between groups
Semaglutide users were generally younger, had higher BMI, slightly lower HbA1c, and were more likely to be female or Black compared with SGLT2 inhibitor users.
The analysis did not adjust for dose or duration of therapy, which investigators identified as an important limitation for future research.
Key Findings
Semaglutide initiation was associated with more than a twofold higher relative risk of NAION compared with SGLT2 inhibitors (HR 2.33, 95% CI 1.54–3.54).
Absolute risk remained low: 0.29% with semaglutide vs 0.13% with SGLT2 inhibitors.
Over 7.5 years, 30 NAION events occurred among 11,478 semaglutide users and 143 events among 90,883 SGLT2 inhibitor users.
Incidence rates were 123 per 100,000 person-years with semaglutide vs 67 per 100,000 person-years with SGLT2 inhibitors.
While the relative risk increase is statistically significant, the event rate is rare in absolute terms.
Biological Plausibility
The mechanism linking GLP-1 receptor agonists to NAION remains uncertain. Hypothesized contributors include:
Hypotension
Volume depletion from gastrointestinal adverse effects
Rapid glycemic improvement leading to transient microvascular dysregulation
Impaired vascular autoregulation at the optic nerve head
These remain theoretical and require mechanistic validation.
Implications for Practice
For patients living with type 2 diabetes, semaglutide remains an effective therapy for glucose control, weight reduction, and cardiovascular risk modification. The observed association with nonarteritic anterior ischemic optic neuropathy, often referred to as optic nerve stroke, appears rare in absolute terms. However, sudden, painless vision loss warrants immediate medical evaluation.
Patients with a history of optic nerve disease or prior NAION should discuss individual risk factors with their prescribing clinician before starting or continuing therapy. Shared decision-making remains central, particularly when balancing meaningful cardiometabolic benefit against a rare but potentially serious adverse event.
For clinicians, this study reinforces the importance of contextualizing relative risk alongside absolute risk. Although the hazard ratio suggests a more than twofold increase compared with SGLT2 inhibitors, the cumulative incidence remained under 0.3 percent. Treatment decisions should therefore incorporate baseline ocular risk, cardiovascular profile, glycemic status, and alternative therapeutic options. Clinicians should counsel patients to report new visual symptoms promptly and coordinate closely with ophthalmology when optic neuropathy is suspected.
Ophthalmologists, in turn, should routinely inquire about GLP-1 receptor agonist exposure in cases of new-onset NAION and communicate findings to the patient’s diabetes care team. At this stage, heightened vigilance and patient education may be more appropriate than broad changes in prescribing patterns.