Hathaway · JAMA ophthalmology 2024 · retrospective matched cohort study · n=1689

Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients Prescribed Semaglutide.

Cited 283 times in the scientific literature.

Level 3 - non-randomized controlled study

Retrospective propensity-matched cohort study

PubMed 38958939 · doi:10.1001/jamaophthalmol.2024.2296 · record verified 2026-08-26

What was done

A retrospective matched cohort study analyzed electronic health record data from 16,827 patients evaluated at a single academic neuro-ophthalmology center between December 2017 and November 2023. Propensity score matching was used to compare patients prescribed semaglutide versus non-GLP-1 receptor agonist medications across two separate cohorts: type 2 diabetes (T2D, n=710) and overweight or obesity (n=979). Covariates included age, sex, systemic hypertension, sleep apnea, hyperlipidemia, and coronary artery disease. The primary outcome was cumulative incidence and hazard ratio of nonarteritic anterior ischemic optic neuropathy (NAION) over 36 months using Kaplan-Meier analysis and Cox proportional hazards regression.

What was found

In the T2D cohort, 17 NAION events occurred in the semaglutide group (n=194) versus 6 in the non-GLP-1 RA group (n=516). The 36-month cumulative incidence was 8.9% (95% CI, 4.5% to 13.1%) for semaglutide versus 1.8% (95% CI, 0% to 3.5%) for controls, corresponding to an adjusted hazard ratio (HR) of 4.28 (95% CI, 1.62 to 11.29; P < .001). In the overweight/obesity cohort, 20 NAION events occurred with semaglutide (n=361) versus 3 with controls (n=618). The 36-month cumulative incidence was 6.7% (95% CI, 3.6% to 9.7%) versus 0.8% (95% CI, 0% to 1.8%), with an adjusted HR of 7.64 (95% CI, 2.21 to 26.36; P < .001).

Why it matters

This study identifies a potential safety signal linking semaglutide prescriptions to an increased risk of NAION, a rare but potentially blinding optic nerve condition, warranting vigilance and replication in larger population-based cohorts.

Limits

The study is observational and cannot establish causality. Data came entirely from a single specialized academic neuro-ophthalmology clinic, introducing substantial referral bias and inflating baseline risk compared to the general population. The absolute number of NAION events was small (total 46 cases across all groups), yielding wide confidence intervals around effect estimates, and residual confounding from unmeasured disease severity or medication adherence remains possible.

Cited by