Trends and comparative effectiveness of inpatient radical hysterectomy for cervical cancer in the United States (2012-2015)

Objective. Report the up-to-date trends in surgical approach for cervical cancer and compare outcomes between open and minimally invasive routes. Methods. Radical Hysterectomy (RH) cases from the National Inpatient Sample (NIS) dataset between 2012 and 2015 were grouped into abdominal (ARH) and Minimally Invasive Surgery (MIS). The MIS group was subdivided as Laparoscopic, Robotic, and Converted. Univariate and multivariable logistic regression were used to analyze differences in complication rates. The National Surgical Quality Improvement Dataset 2015 was used for validation. Results. A total of 7180 cases from NIS were identified. Overall, there was 44% decline in RH cases from 2012 (n = 2220) to 2015 (n = 1255). A proportionate increase in robotic cases from 31.5% in 2012 to 41.4% in 2015 was noted. By intention to treat analysis, the rate of at least one complication for abdominal cases was 24.8% compared to 10% for MIS (p < 0.001). On multivariate analysis, abdominal cases had higher odd of any one complication (aOR 2.9,95% CI 2.12-4.00), medical complication (aOR 3.25,95% Cl 2.15-4.19), infectious complication (aOR 3.76,95% CI 2.1-6.1) but not for surgical complications (aOR 1.7,95% CI 0.5-5.6). AH resulted in longer hospital stay compared to MIS (4.3 vs 1.9 days, p < 0.001). Median cost of AH was $12,624, laparoscopic $12,873, robotic $14,029 and converted cases $17,036. NSQIP analysis supplemented the outcomes to 30-days and showed similar findings. Conclusions. Perioperative complications are significantly lower for MIS procedures. These data should be used for contemporary cost-effective analysis and comprehensive counseling regarding risks and benefits of the surgical approach for cervical cancer. (C) 2018 Elsevier Inc. All rights reserved.