Less is more—the best test for anastomotic leaks in rectal cancer patients prior to ileostomy reversal

PURPOSE: No clear consensus exists on how to routinely assess the integrity of the colorectal anastomosis prior to ileostomy reversal. The objective of this study was to evaluate the accuracy of contrast enema, endoscopic procedures, and digital rectal examination in rectal cancer patients in this setting. METHODS: A systematic literature search was performed. Studies assessing at least one index test for which a 2 × 2 table was calculable were included. Hierarchical summary receiver operating characteristic curves were calculated and used for test comparison. Paired data were used where parameters could not be calculated. Methodological quality was assessed with the QUADAS-2 tool. RESULTS: Two prospective and 11 retrospective studies comprising 1903 patients were eligible for inclusion. Paired data analysis showed equal or better results for sensitivity and specificity of both endoscopic procedures and digital rectal examination compared to contrast enema. Subgroup analysis of contrast enema according to methodological quality revealed that studies with higher methodological quality reported poorer sensitivity for equal specificity and vice versa. No case was described where a contrast enema revealed an anastomotic leak that was overseen in digital rectal examination or endoscopic procedures. CONCLUSIONS: Endoscopy and digital rectal examination appear to be the best diagnostic tests to assess the integrity of the colorectal anastomosis prior to ileostomy reversal. Accuracy measures of contrast enema are overestimated by studies with lower methodological quality. Synopsis of existing evidence and risk–benefit considerations justifies omission of contrast enema in favor of endoscopic and clinical assessment. TRIAL REGISTRATION: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42019107771 SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00384-021-03963-1.