OBJECTIVE: To investigate the association between glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and nonarteritic ischemic optic neuropathy (NAION).
RESEARCH DESIGN AND METHODS: The PubMed, Embase, and Cochrane Library databases were searched during August 2025. Odds ratios (ORs) and absolute risk for NAION values were pooled using random-effects Peto and inverse-variance models. Any ocular event was a secondary outcome.
RESULTS: Fifteen longitudinal studies (n = 8 trials; n > 1.5 million patients) were included. GLP-1 RA use was associated with higher NAION risk (OR 1.70; 95% CI 1.23-2.36), consistent across randomized (2.36; 0.85-6.53) and nonrandomized studies (1.64; 1.15-2.35) (P = 0.51, for heterogeneity). Absolute NAION risk in the GLP-1 RA group was 0.09%, corresponding to a 0.037% risk difference (number needed to harm ~ 2,700). There was no association with overall ocular events (OR 0.95; 95% CI 0.86-1.05).
CONCLUSIONS: GLP-1 RA use was associated with a modest increase in NAION risk but not overall ocular adverse events. Findings underscore the need for long-term postmarketing safety studies and should be interpreted against their well-established mortality and cardio-kidney-metabolic benefits.
| Discipline Area | Score |
|---|---|
| Surgery - Ophthalmology |  |
| Internal Medicine | |
| Endocrine | |
| Public Health | |
| Family Medicine (FM)/General Practice (GP) | |
| General Internal Medicine-Primary Care(US) | |
| Special Interest - Obesity -- Physician |  |
This large meta-analysis (>1.5 million patients; 8 RCTs and 7 observational studies) shows a statistically significant but clinically small increase in nonarteritic ischemic optic neuropathy with GLP-1 receptor agonists (OR 1.70; 95% CI 1.23–2.36), corresponding to an absolute excess risk of ~0.04% (NNH ˜2700). No increase was seen for overall ocular adverse events. For primary care, this provides important quantitative context when counselling patients starting GLP-1 RAs, particularly those with diabetes or vascular risk factors who are already at baseline risk for NAION. The analysis is limited by non-standardized NAION ascertainment and reliance on secondary safety reporting, but the signal appears robust across sensitivity analyses. Importantly, the very small absolute risk must be weighed against the cardio-renal and mortality benefits of GLP-1 therapy.
Findings in this meta-analysis shed light on the ongoing association between GLP-1 RA and the risk of NAIOP, which is very helpful given the widespread use of these agents.
This is a rare outcome that may just be a chance finding because there were no overall excess ocular adverse events. This merits further investigation.
The association may be indirect due to rapid improvement of diabetic control contributing to optic nerve ischaemia.
This meta-analysis assessed the prevalence of NAION in individuals who take GLP-1 inhibitors. GLP-1 RA was associated with a modest increase in NAION risk but not overall ocular adverse events.
