Predictive and Prognostic Implication of Bowel Resections During Primary Cytoreductive Surgery in Advanced Epithelial Ovarian Cancer

Abstract

Objectives: The aims of this study were to assess anastomotic leakage (AL) rate and risk factors for AL in patients with advanced epithelial ovarian cancer (EOC) undergoing cytoreductive surgery including bowel resections and to evaluate the prognostic implication of AL.

Methods: Data of 350 consecutive patients with International Federation of Gynecology and Obstetrics EOC stage IIB-IV who underwent cytoreductive surgery at the Department of General Gynecology and Gynecologic Oncology of the General Hospital of Vienna between 2003 and 2017 were collected. Within this cohort, 192 patients (54.9%) underwent at least 1 bowel resection and were further analyzed. Preoperative risk factors for AL were computed using logistic regression models. Prognostic factors for overall survival were evaluated by using log-rank tests and multivariable Cox regression model.

Results: Overall, the AL rate was 4.7% for patients with advanced EOC undergoing cytoreductive surgery with at least 1 bowel resection, including patients with multiple large bowel resections. The AL rate for patients with isolated rectosigmoid resection was 1.9%. In univariate analysis, the number of anastomoses per surgery (P = 0.04) was associated with the occurrence of AL. In multivariable analysis, rectosigmoid resection with additional large bowel resection was associated with a higher risk of AL compared with isolated rectosigmoid resection (P = 0.046; odds ratio, 7.23 [95% confidence interval, 1.04-50.39]). Anastomotic leakage was associated with decreased overall survival (P = 0.04) in univariate but not in multivariable survival analysis.

Conclusions: Anastomotic leakage rate after rectosigmoid resection in advanced EOC is acceptably low and outweighs increased perioperative risks when performed in a high-volume institution. Nonetheless, the occurrence of AL is a severe adverse event, which even seems to negatively affect patients' overall prognosis. As no factor could be identified to clearly predict AL, extensive procedures comprising multiple bowel resections, should be avoided particularly when complete resection cannot be achieved.