What the research found

The FDA conducted a meta-analysis across 126 clinical trials involving over 130,000 patients and identified a potential elevated risk of NAION (non-arteritic anterior ischemic optic neuropathy)—a rare condition causing sudden vision loss through reduced blood flow to the optic nerve—among GLP-1 users. The analysis found roughly 6.4 cases per 100,000 patient-years in GLP-1 recipients versus 3.8 per 100,000 in comparison groups, suggesting an absolute difference of about 2.6 additional cases per 100,000 patient-years.

The signal is weak and difficult to isolate with certainty. Most trial participants were obese people with diabetes, who carry multiple pre-existing conditions that complicate causal interpretation. Notably, the increased risk appeared when comparing GLP-1s to DPP-4 inhibitors but not to SGLT-2 inhibitors—an inconsistency that raises questions about whether GLP-1s themselves are the culprit or whether confounding factors are at play.

Why it matters for you

If you use GLP-1 medications for weight loss or metabolic health, NAION remains statistically rare. To put it in perspective: your annual risk sits well below 0.01%, and many competing risks—particularly diabetic retinopathy—shift decisively in your favour on GLP-1 therapy. Diabetes is the leading cause of adult blindness in the US, and GLP-1s substantially lower rates of vision-threatening complications like vitreous haemorrhage, neovascular glaucoma, and overall blindness in people with pre-existing diabetic retinopathy.

For users tracking biomarkers like A1C, fasting glucose, or lipid panels, GLP-1 efficacy at normalising these values creates a large protective effect against microvascular complications. Beyond vision, GLP-1 use associates with meaningful reductions in cardiovascular events and all-cause mortality. The decision to continue, start, or adjust dosing is one to make with your doctor and your own risk profile—but wholesale discontinuation based on a rare signal would likely swap a small, uncertain risk for much larger, established ones.

Caveats

• Post-hoc analysis of trials not designed to detect rare events: identifying 15 cases of NAION across 130,000 patients leaves limited statistical power to establish causation
• Confounding abundant in study population: obese diabetics have multiple comorbidities; NAION aetiology remains poorly understood
• Inconsistent comparisons: no increased risk versus SGLT-2 inhibitors, only versus DPP-4 inhibitors, weakening the GLP-1-specific signal
• Early mechanistic understanding: no proven biological pathway proposed for how GLP-1 agonism would elevate NAION risk
• Author disclaimer present: source concludes with promotional content and personal medical claims, introducing potential bias