The learning curve for reducing complications of robotic-assisted laparoscopic radical prostatectomy by a single surgeon

OBJECTIVE To analyse the learning curve for reducing complications of robotic-assisted laparoscopic radical prostatectomy (RALP) performed by a single surgeon in Taiwan. PATIENTS AND METHODS Complication rates were prospectively assessed in 200 consecutive patients undergoing RALP (Group I: cases 1-50; Group II: cases 51-100; Group III: cases 101-150 and Group IV: cases 151-200). Complications were classified using the Clavien system: grade I: deviation normal postoperative course without treatment; grade II: drug or bedside treatment; grade III: endoscopic or surgical intervention; grade IV: life-threatening problem; and grade V: death. Operative parameters and peri-operative complications were evaluated, including operative and console time, blood loss and transfusion rate, Gleason scores, positive surgical margin (PSM) rate, specimen volume, tumour size, tumour percentage, node positive rate and intra- and postoperative complications. RESULTS RALP console time was gradually lowered from Group I to Group IV (P < 0.05). Significantly less blood loss occurred after every 50 cases of RALP (Group I 275 mL, Group II 179 mL, Group III 145 mL, Group IV 102 mL, P < 0.001). Blood transfusion incidence was 8%, 4%, 2% and 0% in Groups I, II, III and IV, respectively. Complication rates were 18%, 12%, 18% and 0% in Groups I, II, III and IV, respectively. Major complications (grade III-IV) were 6%, 2%, 4% and 0% in Groups I, II, III and IV, respectively. Bowel injury occurred in three cases (Group II: 1; Group III: 2); one received intra-operative repair without sequelae and two received a transient colostomy and later colostomy closure. CONCLUSIONS The learning curve for every 50 cases of RALP showed significantly less blood loss and blood transfusion rate. The learning curve for significantly decreasing complications is 150 cases.