Prior vitrectomy raised post-cataract macular oedema from 1.2% to 4.6%

Design

retrospective propensity-matched cohort study in a US federated electronic health record network, 2005 to 2025

Population

7,318 matched pairs of adults undergoing cataract surgery, with and without pars plana vitrectomy at least 6 months earlier

Primary outcome

cystoid macular oedema within 30 to 90 days of cataract surgery

Effect

4.59% vs 1.23%; absolute difference 3.36% (95% CI 2.82 to 3.90); RR 3.73 (2.97 to 4.70), P < .001

Whether a previous pars plana vitrectomy independently raises the risk of cystoid macular oedema after cataract surgery has been debated for years, usually with small series that could not separate the vitrectomy from the disease that caused it. This study used a US federated electronic health record network covering academic and community hospitals from 2005 to 2025, took adults undergoing cataract surgery, and compared those with a vitrectomy at least six months earlier against those without — excluding anyone with pre-existing macular oedema or known risk factors, and propensity-matching on age, sex, race, hypertension, hyperlipidaemia, diabetes, myopia and retinal detachment history.

Among 7,318 matched pairs, cystoid macular oedema occurred within 30 to 90 days in 336 of 7,318 previously vitrectomised eyes (4.59 per cent) against 90 of 7,318 controls (1.23 per cent): absolute difference 3.36 per cent (95% CI 2.82 to 3.90), risk ratio 3.73 (2.97 to 4.70), P < .001. The excess persisted when vitrectomy had been done for retinal detachment (5.65 vs 1.22 per cent, RR 4.62) and for other indications (3.99 vs 1.23 per cent, RR 3.26), and after excluding every eye with an intraoperative or postoperative cataract surgery complication (4.59 vs 1.26 per cent, RR 3.63).

The authors are careful to say the study cannot determine whether prophylaxis or intensified monitoring helps, and that is right — the diagnoses are coding-based and there are no visual acuity outcomes. But the magnitude and its survival through every subgroup make this a risk worth naming at consent and building follow-up around. A one-in-twenty-two chance of macular oedema is a different conversation from one in eighty.

• Quote the raised risk explicitly at consent in a previously vitrectomised eye
• Schedule a review inside the 30 to 90 day window rather than at three months only
• Take a preoperative macular OCT so the postoperative scan can be interpreted
• Do not assume prophylactic topical NSAIDs solve this — the study cannot say whether they help
• Apply the same caution to vitrectomy done for non-detachment indications; the excess was there too

Why it matters

It settles a long-standing question about an eye most cataract surgeons treat as ordinary once the retina is quiet.

Don't overread it

Coding-based diagnoses with no visual acuity outcomes — this establishes the risk but says nothing about whether prophylaxis or extra monitoring changes it.

The statistics, in plain English

A risk ratio of 3.73 on a 1.23 per cent baseline is a large relative increase on a small absolute one — about three extra cases per hundred operations. Because the outcome comes from diagnostic codes rather than imaging, mild cases will be missed in both arms; that biases the absolute numbers down, not the comparison. Propensity matching handles the confounders that were recorded, and cannot handle the ones that were not.

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.