Ocular complications associated with glucagon-like peptide-1 receptor agonists (GLP-1 RAs) – Clinical evidence and insights

Abstract

Type 2 diabetes mellitus (T2DM) and obesity are commonly treated with glucagon-like peptide-1 receptor agonists (GLP-1 RAs). However, there have been growing concerns over the possibility that these medications may result in ocular adverse events, such as nonarteritic anterior ischemic optic neuropathy (NAION). Standard clinical care can overlook these potentially fatal adverse effects. The purpose of this review was to compile the most recent data on ocular side effects associated with GLP-1 RA therapy, assess their clinical relevance, and highlight necessary preventative and monitoring measures. A thorough search of the literature was conducted using Medline/PubMed/PMC, Google Scholar, Scopus, Web of Science, and ScienceDirect. Relevant literature, including observational studies, randomized trials, case reports, and pharmacovigilance data, was selected for its discussion of GLP-1 RAs and associated eye-related adverse events. Semaglutide, a GLP-1 RA, has been associated in a number of studies and safety assessments with an increased risk of NAION and the progression of diabetic retinopathy, especially in patients with underlying eye disorders. Findings, however, were mixed because no statistically significant link was found in a number of cohort studies and meta-analyses. The necessity for additional research was supported by data from adverse event reporting systems that showed disproportionality signals. Clinicians should be mindful of the potential ocular hazards associated with GLP-1 RAs, even though these medications remain useful in treating T2DM and obesity. Closer ophthalmologic monitoring may be beneficial for patients with known diabetic eye problems. Further prospective research with well-defined ocular outcomes is required to elucidate these correlations and guarantee the secure administration of GLP-1 RAs.

An incretin hormone called glucagon-like peptide-1 (GLP-1) is created and released by the intestine in response to meal consumption. GLP-1 binds to GLP-1 receptors and regulates glucose metabolism via various mechanisms, including the stimulation of the release of insulin from the pancreatic β cells, the suppression of glucagon from the pancreatic α cells, the enhanced survival and function of the pancreatic β cells, delayed gastric emptying, and appetite regulation.[1,2,3]

GLP-1 receptor agonists (GLP-1 RAs) are a class of medications that mimic GLP-1’s actions. They aid in the management of type 2 diabetes mellitus (T2DM) and obesity by regulating blood sugar levels and appetite. By attaching themselves to GLP-1 receptors on cells, they cause the pancreatic α cells to release less glucagon, the β cells to produce more insulin, the stomach to empty more slowly, hunger to decrease, and satiety to increase.[4,5,6] Common GLP-1 RAs include semaglutide, exenatide, liraglutide, dulaglutide, lixisenatide, and albiglutide.[7,8,9]

Injection site responses, headaches, nausea, vomiting, and diarrhea are among the most commonly reported side effects of GLP-1 RAs.[10,11] Furthermore, pancreatitis, acute renal injury, gall bladder problems, and, in rare cases, aggravation of gastroparesis can all be brought on by GLP-1 RAs.[12,13] Furthermore, several pharmacovigilance studies and reports from different regulatory bodies have linked the use of GLP-1 RAs to the prevalence of suicidal ideation and self-harming behavior.[14] In addition, the use of GLP-1 RAs is related to ischemic optic neuropathy (ION) and nonarteritic anterior ischemic optic neuropathy (NAION), as identified in recent pharmacovigilance and observational studies. Moreover, certain ocular side effects of GLP-1 RAs, including worsening of diabetic retinopathy, dry eye disease, blurred vision, and other ocular complications, have been documented.

NAION is a rare but dangerous eye disorder caused by ischemia of the optic nerve, which results in painless vision loss. Diabetes, hypertension, obesity, hyperlipidemia, obstructive sleep apnea, smoking, and certain medications such as amiodarone and phosphodiesterase-5 inhibitors are among the significant risk factors for NAION.[15]

Worsening of diabetic retinopathy is perhaps the most significant ocular concern with GLP-1 RAs. Studies have shown a potential increased risk of DR complications in some patients, particularly early in treatment.[16] Rapid improvement in blood sugar control can sometimes paradoxically worsen existing diabetic retinopathy, which is thought to be due to changes in retinal blood flow and vascular endothelial growth factor (VEGF) levels in response to rapid glycemic control.[17,18,19] The risk of diabetic retinopathy worsening could be mitigated by various actions, including close monitoring of patients with pre-existing diabetic retinopathy while taking GLP-1 RAs, gradual dose titration of GLP-1 RAs to minimize rapid blood sugar fluctuations, regular dilated eye exams for patients with diabetes, and appropriate treatment of any signs of worsening of diabetic retinopathy.[20]

GLP-1 RAs may induce dry eye disease via changes in tear film production or composition.[21] Dry eye can cause discomfort, blurred vision, and a gritty sensation in the eyes. The symptoms of dry eye could be mitigated by using over-the-counter artificial tears that can help lubricate the eyes and relieve symptoms, by using prescription eye drops that can reduce inflammation or increase tear production, and by modifying lifestyle, like avoidance of dry environments, staying hydrated, and taking breaks from screen time.[22]

GLP-1 RAs may induce blurred vision through fluctuations in blood sugar levels that can affect the lens of the eye, causing it to swell or shrink. Mitigation of blurred vision could be achieved by consistent blood sugar control. GLP-1 RAs may also induce other ocular side effects such as diplopia, papilledema, and angle closure glaucoma.[23] Considering all these, our current review focuses on ocular complications associated with GLP-1 RA usage.

Methods

This narrative review primarily focuses on ocular complications associated with the use of GLP-1 receptor agonists. To find pertinent papers published, a thorough literature search was conducted using Medline/PubMed/PMC, Scopus, Web of Science, Google Scholar, and reference lists. To find pertinent publications, the search technique used keywords like glucagon-like peptide-1, GLP-1, and GLP-1 receptor agonists, semaglutide, tirzepatide, ocular complications, nonarteritic anterior ischemic optic neuropathy, ischemic optic neuropathy, macular edema, diabetic retinopathy progression, vitreous hemorrhages, and blurred vision.

The review concentrated on clinical evidence including prospective and retrospective observational studies, pharmacovigilance studies, case series, and meta-analyses pertaining to ocular complications linked to GLP-1 RAs.

After screening titles and abstracts, a full-text evaluation was conducted to make sure the included studies were pertinent. The results were combined to provide a thorough examination of ocular complications linked to GLP-1 RAs. This narrative review includes the English-language articles, eliminating the duplicates.

Results and Discussion

Numerous cases of eye problems related to GLP-1 RAs have been documented, including NAION, ischemic optic neuropathy (ION), progression of diabetic retinopathy, macular edema, retinal hemorrhages, and blurred vision. Semaglutide users reported significantly more vision problems than users of other antidiabetic medications, such as other GLP-1 RAs, DPP-4 inhibitors, SGLT2 inhibitors, metformin, and weight-loss drugs like phentermine and orlistat, according to an analysis of the US Food and Drug Administration Adverse Event Reporting System (FAERS) database.[24] Additionally, 5003 ocular adverse events, including retinopathy and visual impairment, were linked to GLP-1 RAs such as semaglutide, liraglutide, and exenatide in a retrospective pharmacovigilance study analyzing the FAERS database.[25] Similarly, another study examining FAERS data found that GLP-1 RAs like semaglutide and lixisenatide were significantly associated with ocular adverse events such as blurred vision, visual impairment, and diabetic retinopathy.[26]

Nonarteritic Anterior Ischemic Optic Neuropathy (NAION)

Several clinical studies have connected the use of GLP-1 RAs to NAION [Table 1]. Exposure to semaglutide for 2, 3, and 4 years raised the risk of developing NAION, according to a cohort study that analyzed data from the TriNetX database, which included 3,344,205 diabetic individuals.[27] Furthermore, 96 documented cases of NAION were found in a pharmacovigilance analysis that examined the FAERS database until December 2024; 83 of these cases were connected to semaglutide. The constant disproportionality signal and the increasing trend of NAION cases, especially from Denmark, indicated a possible safety risk, the study concluded, even though causality could not be verified. The results emphasized the need for additional research on risk factors connected to drugs, patients, and regions.[28] Additionally, a cohort study comprising 44,517 semaglutide users from Denmark and 16,860 from Norway, using data from national health registries in Denmark (2018–2024) and Norway (2018–2022), found that semaglutide users had a threefold higher risk (32 cases) of developing NAION in comparison to those who used sodium-glucose co-transporter 2 (SGLT-2) inhibitors.[29]

Table 1.

GLP-1 RA-induced Nonartertic Anterior Ischemic Optic Neuropathy (NAION)

Type of Study		Outcomes		Reference
Cohort study of TriNetX data		The risk of developing NAION increased after 2 years, 3 years, and 4 years of semaglutide exposure		[27]
Pharmacovigilance study of FAERS data		Identified 96 reported cases of NAION, with 83 cases linked to semaglutide, until December 2024		[28]
Cohort study of national health registries in Denmark and Norway		Threefold increased risk (32 cases) of developing NAION in semaglutide users		[29]
Case series		Four male patients undergoing semaglutide therapy developed NAION within the first year of treatment		[30]
Case report		A 63-year-old male experienced a sudden loss of vision in the lower part of his right visual field, which gradually extended upward, accompanied by pain in the right temporal region, while taking semaglutide for 1 year		[31]
Retrospective case series		9 patients developed ophthalmic complications while using semaglutide or tirzepatide including 7 patients with NAION, one patient with bilateral papillitis, and one patient was diagnosed with paracentral acute middle maculopathy.		[32]
Retrospective study of Observational Health Data Sciences and Informatics (OHDSI) network		The incidence rate of NAION among semaglutide users was 14.5 per 100,000 person-years		[33]
Disproportionality analysis of FAERS data		Significant disproportionality signal for semaglutide, indicating a stronger association with NAION compared to other GLP-1 receptor agonists		[34]
Cohort study		218 patients developed NAION, and the incidence rate of NAION was higher among semaglutide users compared to nonusers		[35]
Case report		A 73-year-old male developed a rare presentation of bilateral NAION in both eyes while using semaglutide		[36]
Retrospective matched cohort study		17 cases of NAION among semaglutide users		[37]
Pharmacovigilance analysis of FAERS data		31,118 adverse events linked to semaglutide (6% were classified under eye disorders, including 10 events of optic ischaemic neuropathy)		[38]
Multicenter retrospective study		No significant increase in NAION risk among semaglutide users across all groups at 1, 2, or 3 years of follow-up		[39]
Retrospective matched cohort study using data from the TriNetX network		Patients with T2DM on semaglutide did not have a significantly higher 5-year risk of NAION compared to controls		[40]
Seven retrospective cohort analyses using a nationally		No significant increase in NAION risk among semaglutide users		[41]
Meta-analysis of 69 RCTs		No cases of NAION in 64 RCTs. In the remaining 5 RCTs, there were 6 cases of ischemic optic neuropathy in patients receiving semaglutide and 4 in those receiving comparator treatments.		[42]

The patients receiving semaglutide therapy experienced NAION within the first year of treatment, according to a case series of four male patients. Optical coherence tomography revealed intraretinal fluid, a packed optic disc, a short Bruch’s membrane opening diameter, and significant optic disc enlargement in all four individuals.[30] Furthermore, a case report of a 63-year-old man detailed how, over a year of semaglutide use, the patient developed pain in the right temporal area and a sudden loss of vision in the lower part of his right visual field that gradually spread upward.[31] A retrospective case series of nine patients who experienced ocular complications while taking tirzepatide or semaglutide also identified seven patients with NAION, one with bilateral papillitis, and one with a diagnosis of paracentral acute maculopathy.[32]

The incidence rate of NAION among semaglutide users was 14.5 per 100,000 person-years, based on a retrospective study analyzing data from the Observational Health Data Sciences and Informatics (OHDSI) network, which includes 14 databases combining administrative claims and electronic health records, and involved a total of 37.1 million patients, including over 810,000 new semaglutide users.[33] Furthermore, adverse event reports related to GLP-1 receptor agonists submitted to FAERS between January 2004 and September 2024 were analyzed using disproportionality methods. This analysis identified a significant disproportionality signal for semaglutide, indicating a stronger link with NAION than other GLP-1 receptor agonists.[34] Moreover, in a cohort study of 424,152 patients with type 2 diabetes from 2018 to 2024, 218 participants developed NAION; of these, 106,454 patients took semaglutide, while 317,698 did not. Semaglutide users experienced a higher incidence rate of NAION compared to nonusers. The median time between starting semaglutide and onset of NAION was nearly 22 months in 67 patients who were on semaglutide.[35] Additionally, after experiencing significant weight loss and postural hypotension while on semaglutide, a 73-year-old male with a history of T2DM, obesity, and hypertension developed bilateral NAION in both eyes, an uncommon presentation. The study underscores the importance of recognizing NAION as a potential adverse effect of semaglutide, especially in patients with existing risk factors for optic nerve ischemia.[36]

Seventeen cases of NAION were found among semaglutide users, compared to six cases in the non-GLP-1 receptor agonist antidiabetic medication group, in a retrospective matched cohort study that included 16,827 patients with T2DM (194 patients were treated with semaglutide, while 516 received non-GLP-1 receptor agonist antidiabetic medications). This suggests that semaglutide users are at a higher risk than those who did not use semaglutide. Semaglutide was prescribed to 361 patients in the overweight or obese category, whereas other weight-loss drugs were administered to 618 patients. In this cohort, only three cases of NAION occurred in the non-GLP-1 RA group, but 20 cases were found in the semaglutide group. According to the study’s findings, patients with T2DM and those who were overweight or obese had a noticeably increased risk of developing NAION if they took semaglutide.[37] Furthermore, using data from the FAERS from 2017 to June 2024, a pharmacovigilance analysis identified 31,118 adverse events associated with semaglutide. Six percent of them were categorized as having eye disorders. Ten events were reported as “optic ischemic neuropathy,” all of which were deemed significant, with the majority of these events happening in female patients. However, no instances were officially documented as NAION.[38]

On the other hand, a multicenter retrospective analysis examined electronic health records from 160 healthcare facilities across 21 countries. The study included data from January 2017 to August 2023, with follow-up through August 2024. It involved 130,216 people with both diabetes and obesity, 37,314 with diabetes alone, and 129,690 with obesity alone. Patients on non-GLP-1 receptor agonist glucose-lowering or weight-loss medications were compared to semaglutide users. After 1, 2, or 3 years of follow-up, the results showed no significant increase in NAION risk among semaglutide users in any group.[39] Additionally, a matched retrospective cohort study using data from the TriNetX network in the United States, which included about 120,000 T2DM patients prescribed semaglutide and 220,000 receiving other GLP-1 receptor agonists, found no significant 5-year difference in NAION risk between semaglutide users and controls.[40] Moreover, seven retrospective cohort studies utilizing a nationally representative dataset of 66 million electronic health records and claims found no apparent rise in NAION risk among semaglutide users. Sensitivity analyses further confirmed that there was no significant increase in NAION risk, despite initial findings suggesting higher risk among patients receiving semaglutide and other GLP-1RAs before adjusting for covariates.[41] Additionally, a meta-analysis of 69 RCTs involving 144,226 patient-years on GLP-1RA treatments (including semaglutide, liraglutide, dulaglutide, and exenatide) and 132,922 patient-years in control groups reported no instances of NAION. In five other RCTs, eight cases of ischemic optic neuropathy occurred in patients on GLP-1RAs and four cases in patients on comparator therapies. Six of these eight cases involved semaglutide. The estimated incidence in the GLP-1RA group was 5.6 cases per 100,000 patient-years, compared to 3 cases per 100,000 in the control group.[42]

Although the study was unable to prove that semaglutide caused NAION, an advisory article published in 2024 by the American Academy of Ophthalmology and the North American Neuro-Ophthalmology Society recognized that a Harvard-based neuro-ophthalmology research team’s analysis had found a possible correlation. Both organizations suggested that a post-marketing surveillance study would shed light on the possible connection between semaglutide and NAION. The advisory recommended that patients seek medical attention right away if they had unexpected vision loss, but it did not suggest stopping semaglutide. Blurred vision, loss of peripheral vision, color changes, black or gray areas in the visual field, and decreased light or contrast sensitivity were among the symptoms of NAION that patients experienced. Additionally, the advisory mentioned that semaglutide had previously been linked to temporary vision changes such as blurred vision and worsening of diabetic retinopathy, often attributed to rapid changes in blood sugar levels. These effects were generally reversible within a few months.[43]

The development of NAION is generally considered to be multifactorial. The genetic mutations associated with NAION (such as certain variants in NOS3, PON1, MT-ND1, and GP1bα) often exhibit incomplete penetrance. This means that some individuals who carry the gene variants may never develop the disease, while others do. Genetic factors contribute to a predisposition, but an acute ischemic event usually occurs only when combined with other local and systemic risk factors.[44]

Progression of diabetic retinopathy

In many clinical studies, the use of GLP-1 RAs has been linked to the progression of diabetic retinopathy [Table 2]. Using data from FAERS covering late 2003 to mid-2024, a disproportionality analysis assessed the potential association between GLP-1 RAs (semaglutide, liraglutide, dulaglutide, lixisenatide, and tirzepatide) and diabetic retinopathy (DR). It was found that semaglutide and dulaglutide showed a strong association with reported cases of diabetic retinopathy, with semaglutide showing the highest risk.[45] Furthermore, a population-based retrospective cohort analysis revealed that among patients with pre-existing DR, GLP-1 RA users experienced a significantly higher rate of DR progression than those using SGLT2 inhibitors. Conversely, the risk of ocular outcomes for individuals without baseline DR was similar across both treatment groups. The study concluded that in patients with established DR, GLP-1 RAs were associated with an increased risk of retinopathy progression compared to SGLT2 inhibitors.[46] Additionally, compared to the SGLT-2 inhibitor group, GLP-1 RA users had a higher rate of progression to proliferative diabetic retinopathy at 1 year and 3 years after starting treatment, based on a retrospective cohort study utilizing data from the TriNetX electronic health record network involving 6481 patients. Moreover, at 3 months, 6 months, 1 year, and 3 years, the GLP-1 group consistently showed a higher incidence of diabetic macular edema. The findings indicate that patients treated with GLP-1 agonists are more likely than those on SGLT-2 inhibitors to experience worsening retinopathy and retinal issues.[47]

Table 2.

GLP-1 RA-induced Diabetic Retinopathy Progression

Type of Study		Outcomes		Reference
Disproportionality analysis of FAERS data		Semaglutide and dulaglutide showed a strong association with reported cases of diabetic retinopathy, with semaglutide having the highest risk		[45]
Population‐based retrospective cohort study		GLP-1RAs users experienced a significantly higher incidence of DR progression compared to the SGLT2 inhibitors users among patients with pre-existing DR		[46]
Retrospective cohort study using data from the TriNetX electronic health record network		GLP-1 RA users showed a higher rate of progression to proliferative diabetic retinopathy at both 1 year and 3 years following treatment initiation, compared to the SGLT-2 inhibitor users		[47]
Meta-analysis of 93 RCTs		Albiglutide appeared to be associated with a significantly higher risk of early-stage DR		[48]
Meta-analysis of 13 RCTs		GLP-1 RAs, including liraglutide, semaglutide, and dulaglutide, were linked to an increased risk of accelerated DR		[49]
Retrospective cohort		No significant association between GLP-1RA use and DR progression		[50]
Clinical cohort study		DR progression was more frequent in the GLP-1 RA users		[51]
Retrospective cohort study		In GLP-1RA users, 10.1% developed vision-threatening diabetic retinopathy (VTDR), 1.2% developed proliferative diabetic retinopathy (PDR), and 8.9% developed diabetic macular oedema (DME)		[52]
Nationwide cohort study		GLP-1 RA use was not linked to a higher risk of VTDR		[53]
Real-world data study		Semaglutide was not associated with increased risk of DR worsening, visual decline, or greater treatment burden during the follow-up		[54]
Meta-analysis of 20 RCTs		No significant increase in the risk of diabetic retinopathy in patients receiving GLP-1RA therapy compared to those on placebo		[55]
Meta-analysis of 8 cardiovascular outcome trials		No significant increase in the risk of hemorrhagic stroke or diabetic retinopathy		[56]

The incidence of diabetic retinopathy, ocular adverse effects, and patient characteristics were compared between patients treated with GLP-1 RAs and those receiving placebo, insulin, or oral antidiabetic medications in a meta-analysis that evaluated data from 93 RCTs that were reported on ClinicalTrials.gov. Albiglutide seemed to be linked to a noticeably increased risk of early-stage DR among the medications examined.[48] Additionally, GLP-1 RAs, including liraglutide, semaglutide, and dulaglutide, were associated with a higher risk of rapid DR deterioration in cardiovascular-focused trials, according to a meta-analysis of 13 RCTs. Using GLP-1 RA was linked to an increased risk of rapid DR progression in four significant RCTs that assessed cardiovascular benefits. In studies that lasted more than 52 weeks and that contrasted GLP-1 RAs with a placebo, the connection persisted as significant. Subgroup analyses involving patients with a diabetes duration of 10 years or more, and those including participants from multiple countries, showed a similar trend, but without reaching statistical significance.[49]

However, a retrospective study of 981 diabetic patients from 2012 to 2023 revealed no significant correlation between the use of GLP-1RAs and the progression of DR (692 patients used GLP-1RAs and 289 used SGLT2 inhibitors). In contrast to SGLT-2 inhibitor use, the study found that GLP-1RA use did not raise the likelihood of DR worsening.[50] Furthermore, a clinical cohort research that evaluated 4500 DR patients discovered that GLP-1 RA users had a higher incidence of DR progression. Only 146 of the DR patients had GLP-1 RA therapy for at least a year with documented ophthalmologic follow-up.[51]

In a retrospective cohort analysis, 14,122 matched controls who were taking other oral antidiabetic medications were compared to 6093 GLP-1RA users. 8.9% of the GLP-1RA group experienced diabetic macular edema (DME), 1.2% proliferative diabetic retinopathy (PDR), and 10.1% vision-threatening diabetic retinopathy (VTDR). These rates were comparable to those in the control group, which had a VTDR of 9.5%, a PDR of 1.2%, and a DME of 8.1%. According to the study’s findings, using GLP-1RA instead of other oral antidiabetic medications did not raise the chance of developing vision-threatening diabetic retinopathy or problems associated with it.[52] Furthermore, countrywide cohort research that included Taiwanese patients with type 2 diabetes and nationwide insurance data found no association between GLP-1 RA use and an increased risk of VTDR. GLP-1 RAs were linked to a much lower risk than DPP-4 inhibitors.[53] Furthermore, semaglutide was not linked to a higher risk of DR deterioration, visual impairment, or increased treatment burden during the follow-up, according to real-world data research from a global federated database. Only 87 of the 4086 individuals with DR whose data were analyzed in the study satisfied the requirements of having taken semaglutide for at least a year and having records of their DR status, visual acuity, and retinal thickness.[54]

Patients taking GLP-1RA medication did not significantly increase their risk of developing diabetic retinopathy as compared to those receiving a placebo, according to a meta-analysis that examined 20 RCTs involving 24,832 people with type 2 diabetes mellitus.[55] Additionally, a different meta-analysis that analyzed data from eight cardiovascular outcome trials with 60,081 individuals with type 2 diabetes showed that patients treated with GLP-1 RAs had a statistically significantly lower risk of both ischemic and total stroke when compared to placebo. Nevertheless, the treatment had no discernible effect on the risk of diabetic retinopathy or hemorrhagic stroke.[56]

Vision-related adverse events

A number of clinical studies determined that the use of GLP-1 RA is linked to vision-related adverse events, including macular edema, vitreous hemorrhage, and blurred vision.

Macular edema

GLP-1 RA usage has been connected to macular edema in a number of clinical studies [Table 3]. Individuals on GLP-1 RAs had a consistently higher incidence of diabetic macular edema at 3 months, 6 months, 1 year, and 3 years compared to the SGLT-2 group, according to a retrospective study using data from 6481 individuals across various healthcare facilities via the TriNetX network. Furthermore, at 1 and 3 years after starting treatment, the progression rate to proliferative diabetic retinopathy was noticeably greater for those using GLP-1 RAs. According to these results, individuals receiving GLP-1 agonists were more likely than those receiving SGLT-2 inhibitors to experience retinal problems and progressive retinopathy.[57] Furthermore, GLP-1 receptor agonists plus insulin were associated with a higher risk of macular edema and diabetic retinopathy than SGLT2 inhibitors plus insulin, according to a retrospective cohort analysis that evaluated roughly 2 million patients with type 2 DM across 97 healthcare organizations worldwide.[58] Additionally, compared to other GLP-1 receptor agonists, semaglutide was associated with a higher prevalence of ocular complications, according to an examination of FDA Adverse Event Reporting System (FAERS) data. The study looked at semaglutide, albiglutide, dulaglutide, and liraglutide reports in the FAERS. Since 2018, semaglutide has been linked to 2109 adverse events, 140 of which have been ocular. These include 47 cases of impaired vision, 4 cases of retinal problems, and 23 cases of diabetic retinopathy.[59]

Table 3.

GLP-1 RA-induced Macular Edema

Type of Study		Outcomes		Reference
Retrospective study		The patients on GLP-1 RAs had a consistently higher incidence of diabetic macular edema at 3 months, 6 months, 1 year, and 3 years compared to the users of SGLT-2 inhibitors		[57]
Retrospective cohort analysis		GLP-1 receptor agonists plus insulin carried a greater risk for both macular edema and diabetic retinopathy compared to SGLT2 inhibitors plus insulin		[58]
Pharmacovigilance analysis of FAERS data		Higher frequency of ocular complications linked to semaglutide compared to other GLP-1 receptor agonists		[59]
Study using proportional hazard models		No evidence of an increased risk of DME associated with the use of GLP-1 RAs, DPP-4 inhibitors, and SGLT2 inhibitors		[60]

On the other hand, a trial using proportional hazard models did not prove a higher risk of DME linked to using DPP-4 inhibitors, SGLT2 inhibitors, or GLP-1 RAs. The use of one or more of these medications also did not significantly alter the risk of developing DME.[60]

Vitreous hemorrhages

The development of vitreous hemorrhage during dulaglutide medication was documented in a case study of a 64-year-old African American man with T2DM and a 4-year history of severe diabetic retinopathy. At week 17 of dulaglutide 4.5 mg therapy, he presented with sudden vision loss in the left eye and was diagnosed with vitreous hemorrhage, suspected to be spontaneous. His ophthalmologist was consulted, and dulaglutide was resumed at a weekly dose of 1.5 mg. He reported no recurrence of symptoms during the next 4 weeks, and an ophthalmologic test revealed no advancement of diabetic retinopathy. The patient was on dulaglutide 4.5 mg weekly and presented to the emergency department with a 1-day history of vision loss in the left eye due to vitreous hemorrhage. Then he was restarted on dulaglutide 1.5 mg weekly, and no ocular complications were noted after 4 weeks of reinitiation. This case illustrated that a history of diabetic retinopathy should not necessarily preclude the use of GLP-1 receptor agonists.[61]

Additionally, patients receiving GLP-1 RAs had a higher risk of vitreous hemorrhage than those receiving a placebo, according to a systematic review and meta-analysis that examined data from 60 RCTs. However, using GLP-1 RA did not increase the incidence of diabetic retinopathy, macular edema, retinal detachment, or retinal hemorrhage.[62]

Possible mechanisms of GLP-1 RA-induced ocular complications

The fundamental mechanisms of the ocular adverse effects linked to GLP-1 RA usage are not well understood. However, several theories have been proposed [Fig. 1], such as the idea that GLP-1 RAs cause a sharp drop in HbA1c, which may lead to changes in retinal blood flow, osmotic shifts, retinal ischemia, and the collapse of the blood-retinal barrier. These factors may ultimately lead to the development of DR and ocular complications like NAION. Glycemia reduction is also associated with retinal vasodilation, which leads to the hyperperfusion and leakage of fragile capillaries, resulting in the development of DR and ocular complications like NAION.[33,63] Moreover, GLP-1 RAs may alter vascular endothelial growth factor (VEGF) signaling, which worsens capillary leakage and vascular permeability, ultimately resulting in ocular morbidity like macular edema.[64] Furthermore, GLP-1 RAs can bind to GLP-1 receptors in the retina, and overstimulating these receptors might have contradictory consequences, especially in the retina.[65]

Figure 1.

Possible Mechanisms of GLP-1 RA-induced Ocular Complications

Recommendations from ADA and AAO

Guidelines for managing diabetes, including recommendations for retinopathy assessment and monitoring, are provided by the current American Diabetes Association (ADA) standards of care. Since GLP-1 RAs have been associated with retinal deterioration, the standards specifically advise assessing the patient’s diabetic retinopathy status before starting or escalating GLP-1 RA treatment. However, the standards do not specify how often eye exams should be performed in this context.[66] If a patient experiences vision loss while using semaglutide, the American Academy of Ophthalmology (AAO) recommends stopping the medication until an ophthalmologist has performed a comprehensive eye exam.[67]

Clinicians should adhere to the general guidelines for retinopathy screening and monitoring as outlined in standards of care, which vary depending on the severity of retinopathy. Typically, patients without diabetic retinopathy should be monitored every 1–2 years, those with mild nonproliferative diabetic retinopathy (NPDR) should have annual follow-ups, patients with moderate NPDR should be checked every 6–9 months and referred to an ophthalmologist within 3–6 months, and patients with severe NPDR should be reviewed every 3–6 months and referred immediately to an ophthalmologist.[68] The ADA emphasizes the importance of managing diabetes from a patient-centered approach. While the standards of care provide general guidelines, physicians should also consider each patient’s individual risk factors, comorbidities, and preferences when deciding on the appropriate frequency of eye exams.[69]

Conclusion

This review emphasizes that although GLP-1 RAs are effective in T2DM and obesity, particularly in terms of enhancing glycemic control and cardiovascular outcomes, more research is needed to understand their potential for ocular adverse effects fully. Concerns about the links between GLP-1 RA use, particularly semaglutide, and various eye-related complications, such as NAION, worsening of DR, macular edema, vitreous hemorrhages, and visual disturbances, have been raised by several studies, including observational analyses, pharmacovigilance reports, and meta-analyses. Findings from extensive cohort studies and consistent signals of disproportionate reporting suggest a potential connection, even though causality has not consistently been demonstrated.

Individual factors, such as comorbidities, the structural features of the optic nerve, and the severity of pre-existing ocular conditions, appear to influence the risk of these issues. Early retinal changes may result from GLP-1 RAs’ rapid reductions in blood glucose levels. While some studies report increased risks, others show no significant difference or even possible benefits compared to alternative treatments.

Based on the available data, healthcare providers should be careful when recommending GLP-1 RAs to patients with existing eye conditions and ensure they have regular eye exams. To better understand these relationships, more well-designed, long-term clinical trials focusing on eye-specific outcomes are needed. In the meantime, a patient-centered approach that carefully considers the benefits of treatment versus potential risks to the eyes is recommended.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.