Fluorescence angiography prevents about one colorectal anastomotic leak for every 24 patients

Design

Systematic review and meta-analysis of 8 RCTs with trial sequential analysis

Population

4712 patients undergoing colorectal resection with anastomosis

Primary outcome

Overall anastomotic leak

Effect

RR 0.68 (95% CI 0.58 to 0.81), I² 0%; ARR 3.6%, NNT about 24

This meta-analysis with trial sequential analysis pooled eight randomised trials (4712 patients) comparing indocyanine green fluorescence angiography with conventional visual assessment of perfusion before colorectal anastomosis.

Angiography reduced overall anastomotic leak (RR 0.68, 95% CI 0.58 to 0.81) with no heterogeneity (I² = 0%). The absolute reduction was 3.6%, a number needed to treat of about 24. Trial sequential analysis crossed the superiority boundary, meaning further trials are unlikely to reverse the primary finding. Benefit was clearest for left-sided anastomoses and for grade A and B leaks. Grade C leaks and reoperation did not differ significantly, and operating time was not prolonged. Infection, ileus and mortality were similar.

After a decade of mixed individual trials, the pooled randomised evidence now supports routine perfusion angiography for left-sided and rectal anastomoses where the equipment is available. The absence of a clear effect on severe leaks means it reduces morbidity more than it prevents catastrophe.

• Use ICG angiography for left-sided and rectal anastomoses where the equipment exists
• Expect about one leak prevented per 24 patients
• Benefit is mainly on grade A and B leaks; severe leaks were not clearly reduced
• It did not lengthen operating time in the trials
• Where laparoscopic fluorescence systems are unavailable, this supports prioritising them in equipment planning

Why it matters

The debate over whether perfusion imaging changes outcomes now has a randomised answer for left-sided resections.

Don't overread it

Severe (grade C) leaks and reoperation were not significantly reduced.

The statistics, in plain English

A risk ratio of 0.68 means about a third fewer leaks. With no heterogeneity, the trials agreed. A number needed to treat of 24 means about 24 patients get angiography to prevent one leak. Trial sequential analysis checks whether enough patients have been studied to trust the result.

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.