Cetirizine, ibuprofen and prednisone use were linked to less vision-threatening diabetic retinopathy

Design

Retrospective propensity score-matched cohort with positive and negative controls

Population

Adults with type 2 diabetes in a US network; 18,762 to 86,846 matched pairs

Primary outcome

Incident diabetic macular oedema and proliferative retinopathy at 3 years

Effect

PDR HR: cetirizine 0.59, ibuprofen 0.48, prednisone 0.36, fenofibrate 0.59, gabapentin 1.04

A propensity-matched cohort in Ophthalmology (10 September 2026) used the US TriNetX network to compare adults with type 2 diabetes who had used cetirizine, ibuprofen or prednisone with matched non-users, using fenofibrate as a positive control and gabapentin as a negative control. Matched groups ranged from about 19,000 to 87,000 pairs.

Over three years, cetirizine users had lower rates of diabetic macular oedema (hazard ratio 0.62) and proliferative retinopathy (0.59), ibuprofen users 0.63 and 0.48, and prednisone users 0.46 and 0.36. Fenofibrate showed similar associations; gabapentin showed none. Results held across nine sensitivity analyses, and negative-control outcomes suggested little bias for cetirizine and ibuprofen.

This is careful work, but it is still an association in prescribing records. Systemic steroids and NSAIDs carry real risks in diabetes. The finding is a reason for trials, not for prescribing any of these drugs to protect the retina.

• Do not prescribe cetirizine, ibuprofen or prednisone to prevent diabetic retinopathy.
• Keep to established measures: glucose, blood pressure and lipid control, and regular screening.
• Remember that prednisone raises glucose and NSAIDs carry renal risk in diabetes.
• Watch for trials of anti-inflammatory approaches to retinopathy prevention.

Why it matters

It points to inflammation as a modifiable driver of diabetic retinopathy worth testing in trials.

Don't overread it

This was observational; it cannot show that these drugs prevent retinopathy.

The statistics, in plain English

A hazard ratio of 0.48 means the rate of proliferative retinopathy was about half in ibuprofen users. An E-value of 2.6 to 5.0 means an unmeasured factor would need to be strongly linked to both drug use and retinopathy to explain the result away, which is possible but not easy.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.