The glucagon like peptide-1 linked to nonarteritic anterior ischemic optic neuropathy

Previous evidence have reported an risk of retinopathy progression following rapid reduction of Hba1c level (~2.5%) with Semaglutide.¹ At risk patients were those who had high baseline HbA1c, long diabetes duration or had evidence of significant diabetic retinopathy prior to treatment initiation. A recent study have raised new evidence linking Semaglutide with an increased risk of developing, nonarteritic anterior ischemic optic neuropathy, an uncommon condition that can cause vision loss.²

This was a retrospective data base analysis involving 16 827 patients at Massachusetts Eye and Ear in Boston. Despite the large number, their analysis only focuses on 710 patients with type 2 diabetes (194 of whom had been prescribed semaglutide) and 979 patients with overweight or obesity (361 prescribed semaglutide). The researchers compared patients prescribed semaglutide with those prescribed a medication other than a GLP-1 agent. They matched patients by factors such as age and sex and whether they had hypertension, obstructive sleep apnea, or coronary artery disease. Cumulative incidence of nonarteritic anterior ischemic optic neuropathy (NAION) were assessed during 36 months of follow-up.

The study showed that Semaglutide use was associated with a 4.3 fold increased risk for NAION in patients with type 2 diabetes and 7.6 fold increased risk in patients with overweight or obesity. Among patients with type 2 diabetes, the cumulative incidence of NAION over 36 months was 8.9% for those prescribed semaglutide versus 1.8% among those taking non-GLP-1 medications. For patients with overweight or obesity, the cumulative incidence of NAION over 36 months was 6.7% for the semaglutide cohort versus 0.8% for those in the other group.

The study has significant limitations. First it is a retrospective study. As such evidence derived from this may not be fully adjusted for residual confounders. Specifically, the study did not adjust for baseline or cumulative HbA1c or diabetes duration—two important factors which determine risk of retinopathy progression. The results were driven by a relatively small number of NAION cases in the patients exposed to semaglutide. The study does not establish that semaglutide directly causes NAION, but rather provide an association which needs to be tested in larger clinical trials or a post-market analysis of all GLP-1 RA drugs. The generalizability of the study is also unclear since the study was undertaken from patients which attend a specialist eye hospital. Interestingly, since risk of NAION was also seen in people without Type 2 diabetes, the potential mechanism for increased risk of NAION is unlikely to be driven by rapid lowering of HbA1c levels as reported in SUSTAIN-6.¹

In clinical practise, the evidence form this study, in my opinion should not detract the continued use of semaglutide—given its well-recognized benefits in improving cardio-metabolic outcomes as shown in multiple randomized clinical trials. Furthermore as stated in an editorial that accompanied the article, given the large numbers of participants who have been recruited to these clinical trials and the large number of people globally who use GLP-1 RAs, we should be confident that if corroborated, the absolute risk of developing NAION in direct relation to taking semaglutide must be rare. Nonetheless it is important to ensure patients remains engaged in attending eye screening regularly.