Intraoperative fluorescence angiography with indocyanine green to prevent anastomotic leak in rectal cancer surgery (IntAct): an unblinded randomised controlled trial

Abstract

Background: Data are mixed on whether indocyanine green (ICG) fluorescence angiography can reduce the high rate of anastomotic leaks in patients undergoing surgery for rectal cancer. Therefore, we aimed to investigate the safety and efficacy of ICG fluorescence angiography in reducing the rate of clinical anastomotic leaks in these patients.

Methods: IntAct was an unblinded randomised controlled trial conducted at 28 specialist rectal cancer centres across eight European countries. Adults (≥18 years) with rectal cancer (lower margin of cancer ≤15 cm from the anal verge) medically fit for elective, curative, laparoscopic or robotic high or low anterior resection were eligible. Patients not undergoing colorectal or anal anastomosis and those with synchronous colonic tumours or recurrent or locally advanced rectal cancer requiring extended or multi-visceral excision were excluded. Eligible participants were randomly assigned (1:1) by use of minimisation with a random element to undergo surgery with or without ICG (standard care). Resections and anastomoses were done per surgeon preference. In the ICG group, surgeons first marked proximal transection levels via standard white-light laparoscopy and then administered an intravenous bolus of 0·1 mg/kg of ICG for perfusion assessment. A second 0·1 mg/kg ICG assessment was done following anastomosis. In the standard care group, only a white-light assessment of bowel perfusion was performed. The primary endpoint was the rate of clinical anastomotic leak (grades B or C, per the International Study Group of Rectal Cancer) within 90 postoperative days. Analyses were done in the intention-to-treat population for complete cases. This trial is registered with the ISRCTN registry (ISRCTN13334746) and is now complete.

Findings: Between Oct 20, 2017, and Aug 15, 2023, 2534 patients were assessed for eligibility and 766 participants were randomly assigned (383 to the ICG group and 383 to the standard care group). 501 (65%) of 766 participants were male, 726 (95%) were of White ethnicity, and the median age was 64·0 years (IQR 56·0-72·0). 343 patients in the ICG group and 355 in the standard care group were included in the intention-to-treat analysis. The rates of anastomotic leak were 11 (3%) of 343 in the ICG group and 20 (6%) in the standard care group for grade A, 11 (3%) and 31 (9%) for grade B, and 25 (7%) and 23 (6%) for grade C. Within 90 days, a clinical anastomotic leak occurred in 90 (13%) of 698 participants: 36 (10%) of 343 in the ICG group and 54 (15%) of 355 in the standard care group (adjusted odds ratio 0·667 [95% CI 0·419-1·060]; p=0·087). There were no serious adverse events related to ICG.

Interpretation: Although IntAct did not show a significant benefit for ICG fluorescence angiography, a signal towards a reduction in clinical anastomotic leak rate was observed. The benefit of ICG could be in preventing grade A or B leaks, given similar rates of grade C leaks between groups. Future research is needed to standardise ICG fluorescence assessment and understand its relevance to anastomotic leak.

Funding: National Institute for Health and Care Research Efficacy and Mechanism Evaluation Programme.