RESUMO
OBJECTIVE: The reproducibility of high-volume published series of laparoscopic radical prostatectomy (LRP) is still debatable. Many questions about its implementation, safety, and number of procedures required to achieve competence and improvement of outcomes with the technique remain unclear, and a learning curve study is crucial to investigate the acceptable performance of this advanced, minimally invasive procedure. SUBJECTS AND METHODS: Between 2004 and 2011, 240 consecutive patients underwent an LRP performed by a single surgeon and were divided into the first, second, and third groups of 80 patients each. Perioperative and oncologic outcomes were compared across the groups to assess the impact of the learning curve for LRP. All surgical complications were classified using the Clavien-Dindo system (CDS). RESULTS: Mean (range) patient age was 61 (43-78) years. The mean (range) level of prostate-specific antigen was 6.47 (3-18) ng/mL. The mean (range) Gleason sum was 6 (5-9). There was a significant reduction in the mean operative time (P<.001), mean anastomosis time (P<.001), mean blood loss (P<.001), mean hospital stay (P<.001), and mean minor CDS complications (P<.01) among the three groups as the series progressed. The D'Amico tumor stage was an independent factor for positive surgical margin across the learning curve (P<.001). CONCLUSIONS: Our study demonstrated safety and low morbidity of the LRP technique since the beginning of a learning curve development, in which up to 80 cases were necessary to create a plateau to improve faster perioperative parameters, although, from the plateau created, it requires a very large number of surgeries for slightly better, additional overall benefits. In spite of its complexity and steep learning curve, new surgeons can be encouraged in the LRP technique with mentorship training without compromising overall outcomes, permitting the wide spread of an alternative minimally invasive procedure in low-volume centers.