Stop the prostaglandin before injecting endothelial cells

Design

Multicentre retrospective cohort study with Kaplan-Meier time-to-detection analysis

Population

57 eyes of 57 patients receiving cultured human corneal endothelial cell injection at three Japanese tertiary centres

Primary outcome

Postoperative detection of cystoid macular oedema on OCT

Effect

26.3% (15/57); prostaglandin analogue use in 60.0% of CME eyes vs 14.3% without (P=.001); 6-month BCVA 0.44 vs 0.60 logMAR (P=.52)

Fifty-seven eyes treated with cultured human corneal endothelial cell injection at three Japanese centres were reviewed for cystoid macular oedema on OCT. It was found in 15, just over a quarter, almost all in the early postoperative period.

The association with prostaglandin analogues is strong: 60 per cent of the eyes that developed oedema had been on one preoperatively, against 14.3 per cent of those that did not. Glaucoma itself, topical glaucoma medication and worse preoperative acuity all tracked with oedema too, which makes disentangling the drug from the disease difficult in a series this size.

The reassuring part is the outcome. All cases resolved with topical NSAIDs or sub-Tenon triamcinolone, and best-corrected acuity at six months was no different. So the practical change is surveillance and drug review rather than a reason to withhold the therapy — with the caveat that preoperative macular status was unknown in 50 of the 57 eyes, so some of this oedema may have been there beforehand.

• Review prostaglandin analogue use before scheduling endothelial cell injection, and consider switching to an alternative agent perioperatively.
• Get a preoperative macular OCT; the study could not exclude pre-existing oedema in most eyes.
• Scan early postoperatively — this is when the oedema appeared.
• Treat with topical NSAID first; all cases in this series resolved.
• Do not stop glaucoma treatment without a plan for pressure control.

Why it matters

The commonest topical drug in the eyes most likely to need this therapy may be the one complicating it.

Don't overread it

Retrospective, univariate, and unable to separate prostaglandin use from the glaucoma it treats.

The statistics, in plain English

This is univariate analysis in 57 eyes, so the prostaglandin association is unadjusted for the glaucoma that led to the prescription — glaucoma, glaucoma drops and worse acuity all showed the same association, and they travel together. P=.001 tells you the difference is unlikely to be chance; it tells you nothing about which of the correlated factors is responsible. Pre-existing oedema was not excluded in 50 eyes, which could inflate the incidence figure.

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.