The glucagon-like peptide 1 receptor agonist (GLP-1RA) semaglutide has rapidly gained prominence as a major treatment option for managing type 2 diabetes and obesity. Recent discoveries have raised serious concerns about a potential association between semaglutide and non-arteritic anterior ischemic optic neuropathy (NAION). NAION accounts for most cases of anterior ischemic optic neuropathy (AION) with the remaining classified as giant cell arteritis-associated arteritic AION (AAION).
NAION is a leading cause of vision loss in adults after glaucoma where it is numerically the second in frequency for optic neuropathies. With optic disc edema, NAION manifests as sudden, painless, and monocular vision loss. Vision loss typically is irreversible and there is no treatment available. Therefore, this potential severe adverse effect associated with semaglutide demands the researchers extend the investigation into this association using the national Danish and Norwegian health registries.
In detail, this cohort study contrasted the risk of NAION among individuals with type 2 diabetes by using semaglutide compared to those using sodium-glucose co-transporter 2 inhibitors (SGLT-2is) and a self-controlled study analyzing the risk of NAION among semaglutide users with regardless of indication of use.
To further investigate the putative link between semaglutide use and NAION. A recent study published in the MedRxiv. This study included a mixed-effects meta-analysis and a propensity score weighting to account for confounding within a bi-national active comparator new-user cohort. An additional self-controlled analysis (symmetry analysis) was also included.
The data set contained patient registers, drug-prescribed registers, and civil registrations, and was shaped from actual national health registries by countries such as Denmark and Norway. Participants used sodium-glucose co-transporter inhibitors (SGLT-2is) or semaglutide for the first time in Norway between 2018 and 2022, and in Denmark between 2018 and 2024. The self-controlled design investigated a wide range of reasons for semaglutide exposure, while the general case added exposure to SGLT-2is in conjunction with semaglutide use in the management of type 2 diabetes. Incidence rates of NAION among users of SGLT-2is and semaglutide, with strong 95 percent CIs for HRs and IRDs. These were to be combined using a fixed effects model (inverse variance weighting) between Denmark and Norway. The supplemental self-controlled study provided the sequence symmetry ratios (SRs).
The study identified 44,517 potential users of semaglutide for type 2 diabetes management in Denmark and 16,860 in Norway. A total of 32 cases of NAION (non-arteritic anterior ischemic optic neuropathy) were recorded among semaglutide users, with 24 cases in Denmark and 8 in Norway. In Denmark, the unadjusted NAION incidence rate for semaglutide users was 2.19 per 10,000 person-years, compared to 1.18 for SGLT-2 inhibitor users. In Norway, the corresponding rates were 2.90 and 0.92 per 10,000 person-years, respectively.
After adjustments, the pooled hazard ratio (HR) was 2.81 (95% CI: 1.67–4.75), with an incidence rate difference (IRD) of +1.41 (95% CI: +0.53 to +2.29). While findings were consistent across both countries, Norway’s estimates were higher and less precise (HR 7.25; 95% CI: 2.34–22.4) compared to Denmark’s (HR 2.17; 95% CI: 1.20–3.92). Sensitivity and additional analyses supported these results, with a post hoc per-protocol analysis showing a stronger association (HR 6.35; 95% CI: 2.88–14.0). In a supplementary self-controlled study, the NAION rate in Denmark was 1.14 (95% CI: 0.55–2.36), while in Norway, it was 2.67 (95% CI: 0.91–8.99).
In this Danish-Norwegian cohort study sampling the overall population, an increased risk for NAION was observed among semaglutide users for T2D concerning users of SGLT-2i. Such findings repeated across all the sensitivity and complementary analyses. In absolute terms, however, the risk of NAION was still low. Analyses of the association of semaglutide for obesity with NAION have yielded inconclusive results.
The study also has several weaknesses. First, the very limited number of NAION events made detailed subgroup analyses impossible, limiting the understanding of how different characteristics (age, sex, and severity of diabetes) may influence the risk. Second, the validity of the AION diagnostic codes from our data is speculative because it has a code for arteritic and non-arteritic diagnoses of AION. However, most (AAION) cases of AION falling under-diagnosis (5-10% only) have a median age of 73 years and increase exponentially with age, with less than 10% occurring before age 60. The cohorts had a median age of 65 years; as a result, one can expect that the lower limit of 5-10% is what applies to our data.
Furthermore, a post hoc analysis in the Danish cohort that censored events of outcome for patients with coexisting diagnosis of giant cell arteritis (as such could underlie any potential events of AAION) resulted in slightly stronger estimates of risk, providing further support to view the outcomes observed as real occurrences of NAION.
On the whole, conclusions indicate that semaglutide use for type 2 diabetes is associated with NAION risk with an increased hazard ratio of more than two. However, the absolute risk of developing NAION is still quite low in those using semaglutide. Analyses regarding associations between semaglutide for obesity and NAION were not conclusive.
Reference: Simonsen E, Lund LC, Ernst MT, et al. Use of semaglutide and risk of non-arteritic anterior ischemic optic neuropathy: A Danish–Norwegian cohort study. medRxiv. 2024. doi:10.1101/2024.12.09.24318574
