Introduction

Although previous studies assessed the learning curve phenomenon associated with the introduction of robot-assisted radical prostatectomy (RARP), evidence is scarce regarding the impact of previous experience with open radical prostatectomy (ORP) as well as annual robotic caseload on outcomes.

Methods

We evaluated the first ≤200 patients with localized prostate cancer treated with RARP by 6 surgeons between 2006 and 2016 (n=1,010). Patients were stratified according to previous experience with ORP of the surgeon (no experience vs. at ≥500 open cases). The rates of positive surgical margins (PSM) were compared using the chi-square test. Multivariable logistic regression analyses assessed the impact of previous experience with ORP on the risk of PSM. Surgeons without experience with ORP were stratified according to their annual caseload during the first 200 cases in two groups: ≤40 vs. >40 cases per year. Multivariable logistic regression analyses tested the impact of annual caseload on the risk of PSM.

Results

Overall, 4 (n=610) vs. 2 surgeons (n=400) had no vs. high experience with ORP. Differences were observed with regards to clinical stage, preoperative PSA, and biopsy Gleason score (all P≤0.02). The rate of PSM was lower among surgeons with previous experience with ORP (13.8 vs. 22.3%; P<0.001). Surgeons with no experience had 2.7-fold higher odds of PSM compared to those with experience with ORP (P=0.01). Overall, 2 surgeons with no experience with ORP had a caseload of ≤40 RARP per year while 2 surgeons performed >40 procedures per year during their first 200 cases. Among surgeons with no experience with ORP, the PSM rates were different according to annual caseload (33.7 vs. 17.4% for ≤40 vs. >40 cases per year). In multivariable analyses, surgeons with a lower caseload had a 2-fold higher probability of PSM (P=0.001). No differences were observed in the rate of PSM between surgeons without previous experience with ORP performing >40 procedures per year and their counterparts with experience with open surgery (17.4 vs. 13.8%; P=0.1). This was confirmed at multivariable analyses (P=0.1).

Conclusions

Surgical experience with previous ORP reduces the risk of PSM in the first robotic cases. A higher annual caseload of initial robotic cases allowed non-open surgeons to reach PSM rates comparable to what achieved by robotic surgeons with previous significant ORP experience. A sufficient annual surgical caseload in the initial learning curve should be guaranteed to optimize perioperative surgical outcomes.

Funding

none