Retzius-sparing robot-assisted radical prostatectomy: early learning curve experience in three continents.

OBJECTIVE: To assess the effect of surgical experience on peri-operative, functional and oncological outcomes during the first 50 Retzius-sparing robot-assisted radical prostatectomy (RsRARP) cases performed by surgeons naïve to this novel approach. MATERIALS AND METHODS: We retrospectively evaluated the initial cases operated by 14 surgeons in 12 different international centres. Pre-, peri- and postoperative features of the first 50 patients operated by each surgeon in all the participating centres were collected. The effect of surgical experience on peri-operative, functional and oncological outcomes was firstly evaluated after stratification by level of surgical experience (initial [≤25 cases] and expert [>25 cases]) and after using locally weighted scatterplot smoothing to graphically explore the relationship between surgical experience and the outcomes of interest. RESULTS: We evaluated 626 patients. The median follow-up was 13 months in the initial group and 9 months in the expert group (P = 0.002). Preoperative features overlapped between the two groups. Shorter console time (140 vs 120 min; P = 0.001) and a trend towards lower complications rates (13 vs 5.5%; P = 0.038) were observed in the expert group. The relationship between surgical experience and console time, immediate urinary continence recovery and Clavien-Dindo grade ≥2 complications was linear, without reaching a plateau, after 50 cases. Conversely, a non-linear relationship was observed between surgical experience and positive surgical margins (PSMs). CONCLUSIONS: In this first report of a multicentre experience of RsRARP during the learning curve, we found that console time, immediate urinary continence recovery and postoperative complications are optimal from the beginning and further quickly improve during the learning process, while PSM rates did not clearly improve over the first 50 cases.