Abstract
OBJECTIVES: In rectal cancer surgery, a diverting stoma (DS) is used to prevent severe anastomotic leakage (AL), peritonitis, reoperation, and longer hospitalization. DS creation has increased in the last few decades, without establishment of clear criteria for construction of a DS. Therefore, the goal of the study was to investigate the validity of DS construction based on risk factors for AL, as an approach to reduce the number of stoma creations. METHODS: The subjects were 143 patients with rectal cancer who underwent laparoscopic or robot-assisted low anterior resection with DS creation from January 2010 to May 2021. Patients and operative characteristics were examined using univariate and multivariate analyses to identify risk factors for AL. RESULTS: Of the 143 subjects, 30 (21%) had AL, including asymptomatic AL in 19 cases (13%) and AL requiring conservative therapy in 11 cases (8%). No case of AL required reoperation. In cases with asymptomatic AL, there was a defect in the anastomosis found by colonoscopy several days after the operation, before discharge. In univariate analysis, preoperative Glasgow prognostic score (GPS) ≥1 (p=0.046) and number of stapler firings ≥3 (p=0.002) were associated with AL. In multivariate analysis, only GPS ≥1 was significantly associated with increased AL (p=0.033; OR=4.225; 95% CI 1.122 to 15.905). CONCLUSIONS: DS creation is effective for avoiding reoperation for AL. Preoperative GPS ≥1 is a risk factor for AL in low anterior resection with a DS for rectal cancer. Thus, DS construction should be considered in a case with an elevated preoperative GPS.