Application of CUSUM analysis in assessing learning curves in robot-assisted sacrocolpopexy performed by experienced gynecologist.

Abstract

Background: The aim of this study was to assess the learning curve of robotic-assisted sacrocolpopexy by applying CUSUM analysis based on operation time, complication rate and conversion rate to open laparotomy.

Methods: A retrospective study was conducted with 50 consecutive robotic-assisted sacrocolpopexy surgeries performed from June 2018 and June 2023 by a single experienced gynecologist. Baseline patient demographics, intraoperative parameters and postoperative outcomes were collected. Cumulative sum (CUSUM) of robotic-assisted sacrocolpopexy operation time was analyzed to determine breakpoints between learning phases using piecewise linear regression. This allowed the detection of subtle shifts in surgical parameters and ultimately surgeon proficiency and competency. Continuous variables, such as age, length of hospitalization and op time, were reported as mean (standard deviation). One-way analysis of variance was employed to compare continuous variables. Categorical variables were expressed as percentages and analyzed using the chi-square test.

Results: The regression identified breakpoints at case 8.47 (95% CI 8.0, 9.0) and case 34.41 (95% CI 32.7, 36.1), with an R² value of 0.87, which agrees with that of the second-order polynomial equation. The breakpoints were rounded to the next whole number at case 9 and 35. The Learning, Proficiency, and Competency phases consisted of 9, 26, and 15 cases, respectively in this consecutive series. This suggests that the surgeon achieved proficiency after the first 9 cases and competency after 35 cases. There were no intraoperative nor short-term post-operative complications during the span of this study. Furthermore, there were no conversions to open laparotomy. CUSUM analysis based on complication and conversion rate, therefore, was not available.

Conclusion: According to CUSUM analysis, surgical proficiency of robotic-assisted sacrocolpopexy was attained after the first 9 cases, and stabilization of operation time was achieved after 35 cases. This statistical tool has proven to be useful in objectively assessing learning curves for new surgical techniques, and the transition from laparoscopic sacrocolpopexy to robotic-assisted sacrocolpopexy seems achievable. This, however, may vary with each surgeon's manual dexterity and experience level. Further investigation with several surgeons and institutions is needed to define a more accurate and generalized learning curve of robotic-assisted sacrocolpopexy.