Predicting incomplete cytoreduction in patients with advanced ovarian cancer.

Purpose: The most important prognostic factor for survival in ovarian cancer patients is complete cytoreduction. The preoperative prediction of suboptimal cytoreduction, considered as any residual disease at the end of surgery, could prevent futile surgery and morbidity. Here, we aimed to identify markers in the preoperative abdominal CT scans of an unselected cohort of patients with ovarian cancer that are predictive of incomplete cytoreduction. Methods: This is a single-institution retrospective analysis of 105 epithelial ovarian cancer (EOC) patients treated with surgical cytoreduction between 2010 and 2020. Twenty-two variables on preoperative abdominal CT scans were compared to the intraoperative macroscopic findings by Fisher's exact test. Parameters with a significant correlation between intraoperative findings and imaging were analyzed by multivariate binary logistic regression analysis regarding the surgical outcome of complete versus incomplete cytoreduction. Results: Complete cytoreduction (CC), indicated by the absence of macroscopic residual disease, was achieved in 79 (75.2%) of 105 patients and 46 (63.9%) of 72 International Federation of Gynecology and Obstetrics (FIGO) stage III and IV patients. Twenty patients (19%) were incompletely cytoreduced due to miliary carcinomatosis of the small bowel, and six patients (5.7%) had various locations of residual disease. Thirteen variables showed a significant correlation between imaging and surgical findings. Large-volume ascites, absence of numerically increased small lymph nodes at the mesenteric root, and carcinomatosis of the transverse colon in FIGO stage III and IV patients decreased the rate of CC to 26.7% in the multivariate analysis. Conclusion: Large-volume ascites, the absence of numerically increased small lymph nodes at the mesenteric root, and carcinomatosis of the transverse colon are markers in preoperative CT scans predicting a low chance for complete cytoreduction in unselected ovarian cancer patients in a real-world setting.