Characterization of a learning curve for robotic cystectomy with intracorporeal urinary diversion at two institutions using the cumulative sum (CUSUM) method

Introduction: Robotic cystectomy with intracorporeal urinary diversion (RCID) is a technically challenging procedure. It is understood that this approach has a learning curve; however, limited studies have characterized this learning curve. The cumulative sum (CUSUM) method plots the learning curve. The aim of this study was to use the CUSUM approach to investigate the number of cases required to reach a consistent, desired performance level for RCID. Materials and methods: Retrospective study of the first 27 and 28 RCID cases performed by two new fellowship trained faculty at two separate institutions from November 2014 to January 2018. Total operating time was calculated and the CUSUM method was used to describe the learning curve, the number of cases needed for a consistent performance level. Results: Twenty-seven and 28 patients were reviewed from two institutions (A and B), with 8 and 7 females, 19 and 21 males and an average age of 66.7 and 67.6 years, respectively. Twelve and ten cases, respectively, had final pathology of stage T3 bladder cancer or higher. The CUSUM curve demonstrated a learning curve of 10 and 11 cases, respectively, when the curve transitioned from steady improvement in OR times (upward slope of curve) to a relative steady state of OR times (plateau of curve). The average lymph node yield, rate of ureteral stricture, and positive margins were also examined with no learning curve noted. Conclusion: In RCID, approximately 10 cases were required by robotically trained new faculty to reach a steady-state level of performance.