Author: Claude (AI), from PubMed abstracts · Reviewed by: Dr Mina Abdelmseih · Reviewed: 2026-11-15 · Review due: 2027-11-15

Educational content only. Not personalised medical advice. Clinical decisions remain the responsibility of the treating clinician.

GLP-1 receptor agonists and the eye

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) such as semaglutide are widely used for type 2 diabetes and obesity and offer considerable cardiovascular benefit [S5]. Two eye questions matter in clinic: early worsening of diabetic retinopathy (DR) with rapid glucose lowering [S1] [S6], and a possible small increase in non-arteritic anterior ischaemic optic neuropathy (NAION) [S11]. Expert consensus holds that the benefits outweigh the ocular risks, with screening and shared decision-making [S6] [S11].

Key points - Most meta-analyses found no overall increase in DR with GLP-1 RAs [S3] [S5] [S7]; worsening events are thought to be linked primarily to rapid glucose lowering rather than a direct drug effect [S6]. - Larger HbA1c reductions in cardiovascular outcome trials were associated with greater retinopathy risk; retinopathy status should be assessed when intensifying glucose-lowering therapy [S2]. - Observational and pooled data suggest a modest increase in NAION risk (OR 1.70 in one meta-analysis), but absolute risk is low, about 0.09% [S8]; a meta-analysis restricted to RCTs found no significant increase in optic nerve or vision-threatening events [S9]. - The North American Neuro-Ophthalmology Society and AAO advise shared decision-making on starting, continuing or stopping GLP-1 RAs [S11]. - Retrospective data suggest GLP-1 RAs may be associated with less glaucoma, but interpretation should be cautious [S4].

Evidence last checked: 2026-10-18 (PubMed abstracts).

Red flags

• Suspected NAION (new optic nerve-related vision loss) in a patient on a GLP-1 RA, particularly semaglutide [S8] [S10] [S11].
• Pre-existing DR with poor glycaemic control and insulin treatment when semaglutide is started: the SUSTAIN 6 excess in DR complications was mostly in this group [S1].
• Diabetes duration over 10 years or HbA1c above 10% (86 mmol/mol) at the start of treatment: high risk, screen before starting [S6].
• Sight loss in one eye or previous NAION: discuss ocular risk before prescribing [S6].

Recognise it

• Diabetic retinopathy. In SUSTAIN 6, semaglutide was associated with a significant increase in DR complications vs placebo, mostly attributed to the magnitude and rapidity of HbA1c reduction in the first 16 weeks [S1]. There was no imbalance in DR adverse events across SUSTAIN 1 to 5 and the Japanese trials [S1].
• Pooled DR data. A meta-analysis of 20 RCTs (24,832 patients) found no increased DR risk (OR 1.17) [S3]; a meta-analysis of 78 semaglutide trials found no change in DR (OR 1.04) [S5]; a 2025 review of RCTs and observational studies found RR 1.00 [S7].
• NAION. A Bayesian meta-analysis of 5 observational studies found semaglutide associated with higher NAION risk (RR 2.52), more so in diabetes; incidence was 118 per 100,000 semaglutide users (low to moderate certainty) [S10]. Semaglutide trials showed significant odds of NAION (OR 3.92), but the data were insufficient for definitive conclusions [S5].

Confirm it

• For suspected NAION, the consensus notes that the evidence comes mainly from retrospective electronic health record and claims studies with limitations [S11]; the diagnostic work-up itself is not covered by these sources.
• For DR, assess retinopathy status before and during intensification of glucose-lowering therapy [S2] [S6].

Manage it

Before acting on this section, check the national and local guidelines, product information and regulatory approvals that apply where you practise. See the medical disclaimer.

Tier A (primary care / diabetes clinic, no eye specialist)

• Make sure people with diabetes have DR screening before starting a GLP-1 RA, especially those at high risk [S6].
• Do not withhold GLP-1 RAs because of eye concerns alone: consensus is that benefits outweigh potential ocular risks and should not limit access [S6].
• Tell patients to report sudden vision loss promptly (see Refer when) [S11].
• Dosing and titration are not covered here: follow local formularies and diabetes guidelines.

Tier B (general eye clinic)

• Grade retinopathy at baseline and follow patients with pre-existing DR more closely when rapid HbA1c reduction is expected; similar recommendations to those for insulin-related early worsening may be appropriate for semaglutide [S1] [S2].
• When a patient with NAION or sight loss in one eye is on, or considering, a GLP-1 RA, discuss the low absolute risk and the uncertainty, and share the decision with the prescriber [S6] [S11].

Tier C (specialist centre)

• Confirm NAION and coordinate with the prescribing team on whether to continue or stop, using shared decision-making [S11].
• Report suspected ocular adverse events; long-term post-marketing safety studies are needed [S8].

Refer when

Pitfalls

• Stopping a GLP-1 RA on eye grounds without weighing its cardio-metabolic benefits [S6] [S8].
• Missing early DR worsening in patients with pre-existing retinopathy and rapid HbA1c falls [S1] [S2].
• Overstating NAION risk: absolute risk is low and RCT data do not show a significant increase [S8] [S9].
• Dismissing the NAION signal: semaglutide-specific observational data show a consistent association [S10].

Evidence notes

• Conflict on NAION: pooled observational data and semaglutide trials suggest increased risk [S5] [S8] [S10], whereas a meta-analysis of RCTs did not [S9]. RCTs reported vision events as adverse events, not prespecified outcomes [S9]; observational data rely on health records and claims [S11].
• Cardiovascular outcome trials were not powered for retinopathy, and median follow-up of 3.4 years was short [S2].
• The glaucoma finding depended on excluding one study and had high heterogeneity [S4].
• Not covered by these sources: dual GIP/GLP-1 agonists such as tirzepatide, use in people without diabetes in detail, periocular or ocular surface effects, and DR screening intervals.

Related articles

• Non-Arteritic Anterior Ischaemic Optic Neuropathy
• Diabetic Macular Oedema

References

See the numbered source list below. Each marker [S#] links to its PubMed and DOI record.