부산대학교 도서관

Background: The utilisation of robot-assisted radical cystectomy with intracorporeal reconstruction (iRARC) has increased in recent years. Little is known about the length of the learning curve (LC) for this procedure. Objective: To study the length of the LC for iRARC in terms of 90-d major complications (MC90; Clavien-Dindo grade ≥3), 90-d overall complications (OC90, Clavien-Dindo grades 1–5), operating time (OT), estimated blood loss (EBL), and length of hospital stay (LOS). Design, setting, and participants: This was a retrospective analysis of all consecutive iRARC cases from nine European high-volume hospitals with ≥100 cases. All patients had bladder cancer for which iRARC was performed, with an ileal conduit or neobladder as the urinary diversion. Outcome measurements and statistical analysis: Outcome parameters used as a proxy for LC length were the number of consecutive cases needed to reach a plateau level in two-piece mixed-effects models for MC90, OC90, OT, EBL, and LOS. Results and limitations: A total of 2186 patients undergoing iRARC between 2003 and 2018were included. The plateau levels for MC90 and OC90 were reached after 137 cases (95% confidence interval [CI] 80–193) and 97 cases (95% CI 41–154), respectively. The mean MC90 rate at the plateau was 14% (95% CI 7–21%). The plateau level was reached after 75 cases (95% CI 65–86) for OT, 88 cases (95% CI 70–106) for EBL, and 198 cases (95% CI 130–266) for LOS. A major limitation of the study is the difference in the balance of urinary diversion types between centres. Conclusions: This multicentre retrospective analysis for the iRARC LC among nine European centres showed that 137 consecutive cases were needed to reach a stable MC90 rate. Patient summary: We carried out a multicentre analysis of the surgical learning curve for robot-assisted removal of the bladder and bladder reconstruction in patients with bladder cancer. We found that 137 consecutive cases were needed to reach a stable rate of serious complications.