Associations between Hospital Volume and Outcomes of Robot-Assisted Radical Prostatectomy.

Abstract

Robot-assisted radical prostatectomy has become the predominant surgical modality to manage localized prostate cancer in the U.S. However, there are few studies focusing on the associations between hospital volume and outcomes of robot-assisted radical prostatectomy. We identified robot-assisted radical prostatectomies for clinically localized (cT1-2N0M0) prostate cancer diagnosed between 2010 and 2014 in the National Cancer Database. We categorized annual average hospital robot-assisted radical prostatectomy volume into very low, low, medium, high and very high by most closely sorting the final included patients into 5 equal-sized groups (quintiles). Outcomes included 30-day mortality, 90-day mortality, conversion (to open), prolonged length of stay (more than 2 days), 30-day (unplanned) readmission, positive surgical margin and lymph node dissection rates. A total of 114,957 patients were included in the study, and hospital volume was categorized into very low (3 to 45 cases per year), low (46 to 72), medium (73 to 113), high (114 to 218) and very high (219 or more). Overall 30-day mortality (0.12%), 90-day mortality (0.16%) and conversion rates (0.65%) were low. Multivariable logistic regressions showed that compared with the very low volume group, higher hospital volume was associated with lower odds of conversion to open surgery (OR 0.23, p <0.001 for very high), prolonged length of stay (OR 0.25, p <0.001 for very high), 30-day readmission (OR 0.53, p <0.001 for very high) and positive surgical margins (OR 0.61, p <0.001 for very high). Higher hospital volume was also associated with higher odds of lymph node dissection in the intermediate/high risk cohort (OR 3.23, p <0.001 for very high). Patients undergoing robot-assisted radical prostatectomy at higher volume hospitals are likely to have improved perioperative and superior oncologic outcomes compared to lower volume hospitals.