An international collaboration for comparison of two approaches to surgical staging of low-risk endometrial cancer

Abstract

patients who underwent laparoscopic hysterectomy for endometrial cancer or EIN by one of four gynecologic oncologists in a single tertiary care academic hospital between 2011 and 2013. Frequencies of SDD, and postoperative complications were evaluated longitudinally over the study periods. Rapid adoption of routine SDD within the institution was conceptualized as a natural experiment, and a noninferiority analysis, with composite postoperative complications as the outcome of interest, was conducted using a two one-sided test approach with an equivalence margin set at 7%. A multivariate logistic regression model was also constructed to evaluate the association between SDD and postoperative complications. Results: Six hundred ninety-four patients were included in the study. All patients underwent laparoscopic hysterectomy, 38.2% had pelvic and 3.0% had para-aortic lymphadenectomy, and 9.4% underwent omentectomy. The rate of SDD increased from 3.9% in the first quarter of 2011 to 70.9% in the final quarter of 2013 (P b 0.001). During this period, the frequency of postoperative readmission, unscheduled surgery, infection, and composite complications within 30 days of hysterectomy did not differ by year (P≥ 0.1 for all). The absolute risk difference in the rate of the composite complication for patients who had surgery in 2013 compared to 2011 was−3.3% (95% CI−7.7 to 1.0), which supports the hypothesis of noninferiority. Patients who had SDD were younger (mean difference = 4.1 years, P b0.001), had shorter duration of surgery (MD 28min, P b 0.001), and had fewer comorbidities (relative risk for Charlson comorbidity index N5= 0.6, P b 0.001) compared with those who were admitted. After controlling for demographic, intraoperative, and comorbid factors, patients who had SDD were not at increased risk for postoperative complications.