Bowel resection at the time of interval cytoreductive surgery for advanced ovarian cancer adversely impacts survival: results from two orthogonal cohorts

Abstract

<h3>Objectives:</h3> To evaluate the survival impact of bowel resection at the time of interval debulking surgery (IDS) for primary advanced epithelial ovarian cancer (EOC). <h3>Methods:</h3> Patients with stage IIIC or IV EOC, diagnosed after 2010, who underwent neoadjuvant chemotherapy followed by IDS were identified from a single institution tumor registry (SITR) and from the National Cancer Data Base (NCDB). Kaplan-Meier survival analysis and Cox proportional hazards models were performed comparing outcomes for patients who underwent bowel resection to those that did not. Overall survival over three years after patients started neoadjuvant chemotherapy was compared between patients who did and did not undergo bowel resection. This study used two distinct datasets which collected varying post-operative outcomes– thus some findings were not able to be compared directly. Progression free survival (PFS) and peri-operative outcomes were evaluated in the SITR only. <h3>Results:</h3> In total, 195 patients met inclusion criteria from the SITR; 9,824 met criteria from the NCDB. The rates of at least one bowel resection were identical between the two cohorts: 43 (22%) and 2,126 (22%). The rates of R0 resection were also similar: 52% and 45%. Sub-optimal resection (residual disease >1cm) was 4% among SITR and 24% among the NCDB; these rates were not different between bowel resection and non-bowel resection cohorts. Patients that required bowel resection had worse three-year survival in both cohorts. Hazard ratio (HR) for death with bowel resection was 2.13 (p<0.01) among SITR and 1.15 (p<0.01) among NCDB controlling for age, R0 resection, and stage of disease. The NCDB survival analysis also controlled for Charlson-Deyo Score. Progression free survival (PFS) was only calculated from SITR and did not differ between groups (HR 1.0, p=1.0). Among the SITR cohort, patients who underwent bowel resection were found to experience significant perioperative complications [intensive care unit admission, acute kidney injury, ileus, surgical injury to bowel or urinary tract, post-operative infection], HR 3.44 (p<0.01), and longer hospital stay (median stay 7.5 vs. 4 days, p<0.01). Among the NCDB cohort, bowel resection was associated with increased rates of unplanned hospital readmission within 30 days of surgery (OR 1.9, p<0.001). <h3>Conclusions:</h3> We demonstrate in both a single academic institution and in a national cohort a significantly increased risk of 3-year mortality among patients who undergo bowel resection at the time of IDS. Identifying modifiable risk factors that contribute to bowel resection and decreasing the related sequelae of bowel resections for patients undergoing IDS is warranted.