Semaglutide Linked to Increased Risk of Severe Eye Condition NAION in New Study

A recent study published in the journal JAMA Ophthalmology evaluated associations between semaglutide and the risk of non-arteritic anterior ischemic optic neuropathy (NAION).

NAION is a significant cause of adult blindness and the second most common optic neuropathy. Semaglutide is a glucagon-like peptide (GLP)-1 receptor agonist (RA) approved by the United States (US) Food and Drug Administration to treat obesity and type 2 diabetes (T2D). Weekly new prescriptions of these and other GLP-1 RAs have increased by 60% in the US between 2021 and 2023. However, anecdotal reports suggest that semaglutide might be associated with NAION.

Study: Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients Prescribed Semaglutide. Image Credit: MattL_Images / Shutterstock

About the study

In the present study, researchers evaluated the risk of NAION among semaglutide users. Patients referred to a clinic for neuro-ophthalmology indications between December 2017 and November 2023 were identified for inclusion from the Massachusetts General Brigham centralized data registry. Electronic health records were queried to identify NAION events.

Each identified record was reviewed manually to confirm the NAION diagnosis. People with overweight/obesity or T2D were separately analyzed. Individuals who did not use semaglutide were included for comparison. Sex/gender and race were self-reported. The study populations were T2D patients prescribed semaglutide or non-GLP-1 RA anti-diabetic medicines and obese/overweight patients prescribed semaglutide or non-GLP-1 RA weight loss medicines.

Non-GLP-1 RA anti-diabetic medicines included thiazolidinediones, sulfonylureas, metformin, insulin, and inhibitors of sodium-glucose transport protein 2, dipeptidyl peptidase 4, or α-glucosidase. Further, non-GLP-1 RA weight loss medicines included setmelanotide, phentermine, topiramate, orlistat, naltrexone, and bupropion.

For balance between cohorts (semaglutide users and non-users), 1:2 nearest-neighbor propensity score matching was applied to account for demographic factors, contraindications for semaglutide, indications for semaglutide use, and use of NAION-associated drugs. The primary outcome was a NAION event. Cumulative incidence of NAION was estimated using the Kaplan-Meier method.

Person-time was calculated from prescription to NAION event, end of follow-up (36 months), or death. Cox proportional hazards regression analyzed the associations between covariates and NAION risk. The log-rank test was used to assess survival times. In addition, secondary analyses were performed with 1:1 nearest-neighbor propensity score matching and an exact match for baseline variables differing by ≥ 20% between cohorts for each study population.

Findings

The researchers identified nearly 17,300 unique patient records; of these, 16,827 patients were included, following the exclusion of people under 12 years. Overall, 979 individuals were overweight or obese, and 710 had T2D. In the T2D group, NAION occurred in 17 patients using semaglutide and six non-semaglutide users. The average follow-up duration was 33.3 and 34.5 semaglutide users and non-users, respectively.

The cumulative incidence of NAION at 36 months was 8.9% for the semaglutide cohort and 1.8% for the non-semaglutide cohort. The survival probability declined sharply over the first year for semaglutide users. Semaglutide users had a higher risk of NAION than non-users, with a hazard ratio of 4.28. In secondary analyses, 264 patients with no history of NAION were included; the elevated risk of NAION was sustained for semaglutide users.

In the overweight/obesity group, 20 semaglutide users and three non-users developed NAION. The semaglutide and non-semaglutide cohorts were followed up for 34.1 and 35.4 months, respectively. The cumulative incidence of NAION was 6.7% and 0.8% for the semaglutide and non-semaglutide cohorts, respectively, at 36 months.

Likewise, the survival probability showed a steep decline in the first year for semaglutide users. Compared to the non-semaglutide cohort, the semaglutide cohort had an elevated risk of NAION, with a hazard ratio of 7.64. Secondary analyses included 442 patients without NAION, and the risk of NAION was consistently higher in the semaglutide cohort.

Conclusions

The findings indicate that semaglutide use is associated with an increased risk of NAION. The relatively high hazard ratios underscore the substantially elevated risk of NAION among semaglutide users compared to those prescribed non-GLP-1 RA medications for obesity and T2D. The risk did not appear to be due to differences in baseline characteristics between cohorts.

The study’s limitations include limited generalizability to other settings and ethnic/racial groups and non-assessment of adherence to the prescribed medicine. Taken together, the study uncovered an association between semaglutide and NAION; further large-scale, multicenter, population-based studies are required to corroborate these findings.