Fenofibrate slows diabetic retinopathy progression, but vision benefit is unproven

Design

Systematic review and meta-analysis of randomised trials, GRADE-rated

Population

3 trials, 3,756 adults with diabetes, mostly type 2

Primary outcome

Diabetic retinopathy progression

Effect

RR 0.77 (95% CI 0.68 to 0.88), 45 fewer per 1,000; VTE RR 1.52 (1.12 to 2.06)

A pre-registered meta-analysis included only randomised trials of fenofibrate reporting an eye outcome in adults with diabetes, mostly type 2. Three completed trials, including LENS, with 3,756 participants contributed to the main outcome.

Fenofibrate reduced retinopathy progression (RR 0.77, 95% CI 0.68 to 0.88; I-squared 0%; moderate certainty), about 45 fewer progression events per 1,000 people. The need for retinal treatment also fell (RR 0.68), but 90% of that estimate came from one older trial and a more conservative analysis crossed 1, so certainty was low. No trial showed better visual acuity. Venous thromboembolism was higher with fenofibrate (RR 1.52, 1.12 to 2.06), an estimate drawn from one large cardiovascular trial.

For an ophthalmologist, this is a reasonable basis to raise fenofibrate with the diabetes team for a patient with progressing non-proliferative retinopathy, while being clear that it slows structural change rather than protects sight. It does not replace control of glucose, blood pressure or lipids.

Fenofibrate is inexpensive and available in India. Kidney function affects dosing and the thrombosis signal deserves weighing in high-risk patients.

• Consider discussing fenofibrate with the diabetes team for patients with progressing retinopathy.
• Explain that it slowed retinopathy progression but has not been shown to preserve vision.
• Check kidney function before starting; dosing depends on it.
• Weigh the higher venous thrombosis risk seen in one large trial.
• Keep glucose, blood pressure and lipid control as the foundation.

Why it matters

It is a cheap oral drug with consistent evidence of slowing retinopathy, which ophthalmologists rarely prompt.

Don't overread it

No trial showed better vision; slowing retinopathy grade is not the same as preventing sight loss.

The statistics, in plain English

A risk ratio of 0.77 means about 23% fewer progression events, or 45 fewer per 1,000 people. I-squared of 0% means the trials agreed. The thrombosis estimate comes from one trial and should be read with that in mind.

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.