Background: Robot-assisted radical prostatectomy (RARP) can generally be performed with 1-2 nights of postoperative monitoring before discharge from the hospital. Little is known about what causes individual patients to remain in hospital beyond the second postoperative day.
Methods: Data for RARPs performed between 2013 and 2015 were extracted from the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database. The fraction of cases with prolonged length of stay (PLOS) that can be reasonably attributed to complications was examined. Logistic regression was performed to identify risk factors for PLOS in the overall population and separately in the population of patients with PLOS without any perioperative complications.
Results: Of 11,440 patients, 10,342 (90.4%) were discharged on postoperative days 0-2; 80.6% (887/1101) of patients with PLOS did not experience any perioperative complications. The most common complication was bleeding requiring transfusion, but this was present in only 5.6% (62/1101) of patients with PLOS. Logistic regression identified predictors of PLOS as age, race, wound class, American Society of Anesthesiologists class, smoking, diabetes, dyspnea, dependent functional health status, congestive heart failure, operative time, and pelvic lymph node dissection. Results of this regression were insensitive to the exclusion of patients who experienced no perioperative complications.
Conclusions: This study utilizes logistic regression on NSQIP data to identify risk factors for PLOS after RARP and, in particular, to evaluate the role of postoperative complications in PLOS. The analysis shows that postoperative complications account for a small minority of cases of PLOS after RARP.