UCSF

BACKGROUND: Although attempts have been made in the past to establish consensus regarding the definitions and grading of the severity of colorectal anastomotic leakage, widespread adoption has remained limited. OBJECTIVE: A systematic review of the literature was conducted to examine the various elements used to report and define anastomotic leakage in colorectal cancer resections. DATA SOURCES: A systematic review was conducted using the PubMed, Embase, and Cochrane Library Database. STUDY SELECTION: All published randomized controlled trials, systematic reviews, and meta-analyses containing data related to adult patients undergoing colorectal cancer surgery and reporting anastomotic leakage as a primary or secondary outcome, with a definition of anastomotic leakage were included. MAIN OUTCOME MEASURES: Definitions of anastomotic leakage, clinical symptoms, radiological modalities and findings, findings at reoperation, and grading terminology or classifications for anastomotic leakage. RESULTS: Of the 471 articles reporting anastomotic leakage as a primary or secondary outcome, a definition was reported in 95 studies (45 randomized controlled trials, 13 systematic reviews, and 37 meta-analyses) involving a total of 346,140 patients. Of these 95 articles, 68% reported clinical signs and symptoms of anastomotic leakage, 26% biochemical criteria, 63% radiological modalities, 62% radiological findings, and 13% findings at reintervention. Only 45% (n = 43) of included studies reported grading of anastomotic leakage severity or leak classification, and 41% (n = 39) included a time frame for reporting. LIMITATIONS: There was a high level of heterogeneity between the included studies. CONCLUSIONS: This evidence synthesis confirmed incomplete and inconsistent reporting of anastomotic leakage across the published colorectal cancer literature. There is a great need to develop and implement a consensus framework for defining, grading, and reporting anastomotic leakage. REGISTRATION: Prospectively registered at PROSPERO (ID 454660).