Objective
To evaluate perioperative complications for open and robotic‐assisted RP for patients enrolled in the PREVENTER trial to determine predictors and impact on opioid consumption.
Patients and Methods
A prospective cohort of 500 patients undergoing open and robotic‐assisted RP were followed to determine rates of complications and opioid use. Complications were classified 30 days post‐RP using the Clavien‐Dindo system. Patient characteristics and outcomes were compared using appropriate statistical tests. Logistic and linear regression were performed to identify predictors of complications and evaluate the relationship between complications and post‐operative opioid use.
Results
A total of 124 (24.8%) men underwent open and 376 (75.2%) robotic‐assisted RP with 418 men (83.6%) receiving pelvic lymph node dissection (PLND). While 83 patients (16.6%) experienced Clavien complications, only 19 (3.8%) were major (≥Clavien III) with no differences by surgical approach. PLND (OR: 2.96,95%CI:1.25‐8.71,p=0.03) and stage pT3b (OR: 2.76,95%CI:1.23‐6.00,p=0.01) were the only predictors of complications after controlling for potential confounders. Patients who experienced complications had greater inpatient (p=0.02) and outpatient (p=0.005) opioid use, which persisted after controlling for patient‐reported pain, attending surgeon variation, surgical approach, and undergoing PLND (inpatient β:77.2,95%CI:17.9‐136.5,p=0.03 and outpatient β:21.9,95%CI:4.7‐39.1,p=0.01).
Conclusion
In an analysis of prospectively collected data, overall and major complications rates did not differ by surgical approach. Patients receiving PLND and with stage pT3b disease experienced more complications. Complications were independently associated with higher inpatient and outpatient post‐operative opioid use.