Key Takeaways

  • Tirzepatide was associated with lower 12-month risks of new or worsening diabetic retinopathy compared with lifestyle intervention alone
  • Tirzepatide users had fewer vision-threatening complications and were less likely to require intravitreal anti-VEGF injections or panretinal photocoagulation
  • The retrospective study included 173,846 propensity-matched patients with diabetes and overweight or obesity across 70 US health systems

Patients with diabetes who initiated tirzepatide had lower rates of diabetic retinopathy, vision-threatening complications, and retinal interventions than matched patients receiving lifestyle intervention alone, according to a large retrospective study published in Ophthalmology.1

The findings suggest that tirzepatide, a dual glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1 receptor agonist, may have a different ocular risk profile than has been reported with some other rapidly acting glucose-lowering medications. However, the researchers say the observational study establishes an association and does not demonstrate that tirzepatide directly protects against diabetic retinopathy.

Investigators used the TriNetX US Collaborative Network to identify patients with diabetes and overweight or obesity who initiated tirzepatide. These patients were propensity score matched with individuals who received lifestyle interventions, such as nutrition therapy or exercise counseling, but had not been exposed to weight-loss medications.

After matching for demographic, metabolic, and systemic variables, the analysis included 173,846 patients—86,923 in each group. The mean patient age was 56.9 years, and 52% of participants were women. During 12 months of follow-up, tirzepatide use was associated with a significantly lower risk of several diabetic retinopathy diagnoses and complications. Compared with lifestyle intervention alone, tirzepatide was associated with reduced risks of:

  • Incident mild nonproliferative diabetic retinopathy (relative risk [RR], 0.864; 95% CI, 0.758-0.985)
  • Proliferative diabetic retinopathy (RR, 0.705; 95% CI, 0.564-0.882)
  • Diabetic retinopathy with macular edema (RR, 0.624; 95% CI, 0.536-0.727)
  • Vitreous hemorrhage (RR, 0.607; 95% CI, 0.429-0.860)
  • Tractional retinal detachment (RR, 0.370; 95% CI, 0.179-0.765)

The investigators also found lower utilization of treatments typically required for more advanced diabetic eye disease. Tirzepatide use was associated with a reduced likelihood of receiving intravitreal anti-VEGF injections (RR, 0.479; 95% CI, 0.368-0.625) or panretinal photocoagulation (RR, 0.610; 95% CI, 0.403-0.924).

In an example of the absolute event rates, incident mild nonproliferative diabetic retinopathy was recorded in 0.49% of patients treated with tirzepatide compared with 1.2% of matched controls, according to Weill Cornell Medicine.

“Based on findings obtained from a large database of patients across multiple clinical practices, those with diabetic retinopathy may be less concerned that taking tirzepatide is going to make their condition worse,” senior author Szilárd Kiss, MD, said in a statement. “The findings suggest they may have a reduced risk of requiring more eye treatments with lasers or injections, which are typically required when retinopathy becomes severe.”

Tirzepatide activates both the GLP-1 and GIP receptors and is used to improve glycemic control and support weight loss. Interest in the ocular effects of incretin-based therapies has grown amid concerns that rapid improvement in glycemic control may temporarily worsen diabetic retinopathy in certain patients—a phenomenon previously reported with intensive diabetes treatment and investigated in relation to semaglutide.

The authors said the findings raise the possibility that tirzepatide and semaglutide may not affect diabetic retinopathy in the same way. Differences in metabolic effects, insulin sensitivity, inflammation, and retinal microvascular responses could contribute to differing outcomes, although those mechanisms were not evaluated by the study.

The researchers warn that the retrospective analysis relied on electronic health records and diagnostic and procedural coding, which may not capture retinopathy severity as precisely as fundus photography, OCT, or standardized reading-center grading. Propensity matching can reduce measured differences between groups but cannot eliminate unmeasured confounding. Patients prescribed tirzepatide also may have differed from controls in medication access, health behaviors, diabetes management, or ophthalmic follow-up. In addition, the lifestyle-intervention group was not an active medication comparator, and the study did not establish whether the observed associations resulted from tirzepatide itself, improved glycemic control, weight loss, reduced systemic inflammation, or other factors.

The investigators concluded that the results may help inform treatment selection for patients at risk of diabetic retinopathy but said further research—including prospective studies incorporating retinal imaging, visual acuity, and retinal thickness measurements—is needed.

Reference

1. Shah J, Razavi P, Festok M, et al. Tirzepatide and reduced risk of diabetic retinopathy and related complications. Ophthalmology. 2026;133(6):728-732. doi:10.1016/j.ophtha.2026.01.013