Perfusion Assessment in Left-Sided/Low Anterior Resection (PILLAR III): A Randomized, Controlled, Parallel, Multicenter Study Assessing Perfusion Outcomes With PINPOINT Near-Infrared Fluorescence Imaging in Low Anterior Resection.

BackgroundIndocyanine green fluoroscopy has been shown to improve anastomotic leak rates in early phase trials.ObjectiveWe hypothesized that the use of fluoroscopy to ensure anastomotic perfusion may decrease anastomotic leak after low anterior resection.DesignWe performed a 1:1 randomized controlled parallel study. Recruitment of 450 to 1000 patients was planned over 2 years.SettingsThis was a multicenter trial.PatientsIncluded patients were those undergoing resection defined as anastomosis within 10 cm of the anal verge.InterventionPatients underwent standard evaluation of tissue perfusion versus standard in conjunction with perfusion evaluation using indocyanine green fluoroscopy.Main outcome measuresPrimary outcome was anastomotic leak, with secondary outcomes of perfusion assessment and the rate of postoperative abscess requiring intervention.ResultsThis study was concluded early because of decreasing accrual rates. A total of 25 centers recruited 347 patients, of whom 178 were randomly assigned to perfusion and 169 to standard. The groups had comparable tumor-specific and patient-specific demographics. Neoadjuvant chemoradiation was performed in 63.5% of perfusion and 65.7% of standard (p > 0.05). Mean level of anastomosis was 5.2 ± 3.1 cm in perfusion compared with 5.2 ± 3.3 cm in standard (p > 0.05). Sufficient visualization of perfusion was reported in 95.4% of patients in the perfusion group. Postoperative abscess requiring surgical management was reported in 5.7% of perfusion and 4.2% of standard (p = 0.75). Anastomotic leak was reported in 9.0% of perfusion compared with 9.6% of standard (p = 0.37). On multivariate regression analysis, there was no difference in anastomotic leak rates between perfusion and standard (OR = 0.845 (95% CI, 0.375-1.905); p = 0.34).LimitationsThe predetermined sample size to adequately reduce the risk of type II error was not achieved.ConclusionsSuccessful visualization of perfusion can be achieved with indocyanine green