RESUMO
OBJECTIVE: Study of the learning curve of robot-assisted radical prostatectomy, evaluating intraoperative difficulties and postoperative complications according to Clavien-Dindo classification. PATIENTS AND METHODS: Retrospective study of our first 157 consecutive patients treated with robot-assisted prostatectomy for localized prostate cancer between September 2011 and December 2014. Comparison of learning for each group of 50 procedures and then comparison between patients operated on by a pair of two seniors specially trained for robotic surgery and patients operated on by one mixed pair including a surgeon junior coached by one senior of the first group. RESULTS: Only postoperative complications decreased significantly from the 51st patient (P=0.04). The curves showing the evolution of the operative time decreased with a parallel trend between the two pairs, but with more variability in the mixed pair. There was no significant difference in terms of intraoperative difficulties (P=0.59), nor postoperative complications (P=0.56) mainly of grade 2. The blood loss, transfusion rate, duration of hospitalization and readmission rates did not differ. Lymph node dissection did not affect outcomes. For oncological results, the overall rate of positive surgical margins (R+) was 30.6 % in the initial pair against 24.2 % in the mixed group with no significant difference. Nevertheless, the subpopulation study objectified a R+ rate of 12.86 % for pT2 against 42.85 % for pT3. CONCLUSION: The early involvement of a junior surgeon who did not receive specific training, but benefiting from the guidance of a senior surgeon, did not compromise the results while allowing a faster learning curve with a rate of operative complications close to the one observed by the senior pair. LEVEL OF EVIDENCE: 4.