|
Although Rare, GLP-1s Tied to Higher Risk for Eye Complications
The New York Times | By Andrew Rhoades
August 22, 2025
People with type 2 diabetes treated with GLP-1 receptor agonists may face a slightly higher risk for eye complications compared with those who receive other antidiabetic medications, two studies show.
The findings, published simultaneously in JAMA Network Open, indicate a need for patient-provider discussions about possible risks and for more studies to better understand the associations, according to researchers.
Healio previously reported that semaglutide (Novo Nordisk) use was tied to a higher risk for nonarteritic anterior ischemic optic neuropathy (NAION), though past literature on this link has been “inconsistent,” which “may be due to limited sample sizes and different study designs,” Rong Xu, PhD, a professor of biomedical informatics at Case Western Reserve University, and colleagues wrote.
In the current study, the researchers assessed the first-time diagnoses of NAION and other optic nerve disorders among 159,398 matched patients with type 2 diabetes. Among them, 79,699 were prescribed semaglutide or tirzepatide (Zepbound, Eli Lilly). The researchers compared their outcome with patients who received other antidiabetic medications, including insulin, metformin, dipeptidyl-peptidase-4 inhibitors, sodium-glucose cotransporter-2 inhibitors, sulfonylureas, thiazolidinediones and other GLP-1 receptor agonists like liraglutide.
They found there were 35 patients diagnosed with NAION in the semaglutide or tirzepatide group and 19 patients in the comparison group (HR = 1.76; 95% CI, 1.01-3.07) over a 2-year follow-up period.
Meanwhile, there were 93 and 54 patients with other optic nerve disorders in the semaglutide or tirzepatide group and comparison group, respectively (HR = 1.65; 95% CI, 1.18-2.31).
Since the comparison group included other GLP-1s, Xu and colleagues noted that the underlying mechanisms may be specific to semaglutide and tirzepatide, “which may include their indirect associations with abrupt changes in metabolic parameters.”
“NAION is rare in general,” Xu told Healio. “For patients with high risk of developing NAION (eg, those with diabetes or hypertension) who are taking GLP-1 receptor agonists, ophthalmologists may increase vigilance.”
The researchers concluded that further research is warranted “to replicate these findings, explore underlying mechanisms, identify individuals at increased risk for these potential complications and examine other eye disorders.”
Diabetic retinopathy
In another analysis, David J. Ramsey, MD, PhD, MPH, director of ophthalmic research at the Lahey Hospital & Medical Center, and colleagues wrote that past studies have suggested associations between GLP-1s and diabetic retinopathy (DR).
A significant limitation of these reports, however, “is that they did not specifically assess for sight-threatening complications from DR, which account for most of the long-term ocular morbidity associated with type 2 diabetes,” they wrote.
The researchers analyzed a cohort of 185,066 people aged 18 years or older with type 2 diabetes prescribed GLP-1s to determine possible elevated risks for DR, NAION and several DR complications.
They reported the use of GLP-1s was tied to a higher risk for DR (HR = 1.07; 95% CI, 1.03-1.11), though not NAION, over a 2-year follow-up period.
They added that GLP-1s were not associated with progression to proliferative DR or diabetic macular edema in a subgroup analysis of 32,695 patients with preexisting DR, but were tied to a lower occurrence of:
- blindness (HR = 0.77; 95% CI, 0.73-0.82);
- neovascular glaucoma (HR = 0.78; 95% CI, 0.68-0.88); and
- vitreous hemorrhages (HR = 0.74; 95% CI, 0.68-0.8).
“These findings suggest that GLP-1 RAs may be a factor in reduced rate of vision loss leading to blindness, even among individuals with preexisting DR,” Ramsey and colleagues concluded.
They that added HbA1c improvement “is the key factor in delaying DR progression and preventing sight-threatening complications.”
‘Rare but serious’
In a related editorial, Sally S. Ong, MD, an assistant professor of ophthalmology at Wake Forest School of Medicine, and Francisco J. Pasquel, MD, MPH, an associate professor of medicine and global health at Emory School of Medicine, said that patients initiating GLP-1s should be counseled about the “rare but serious” risk for NAION, “especially if they have preexisting optic-disc crowding.”
Meanwhile, patients with DR “require close ophthalmic monitoring when [incretin receptor agonists (IRAs)] are initiated or intensified,” they wrote. “Slower titration to avoid abrupt glycemic swings or lower maintenance doses in individuals at highest risk merit formal evaluation.”
Ong and Pasquel also underlined the need for “strong synergy” between diabetes care providers and ophthalmologists, as such collaborations “may allow timely IRA dose adjustments rather than abrupt initiation or discontinuation, which itself may worsen cardiometabolic risk.”
“The path forward is one of balance: preserve the broad advantages of IRAs and their successors while prospectively elucidating, stratifying and mitigating ocular risk,” they concluded.
| 
