Cumulative summation analysis of learning curve for robot-assisted gastrectomy in gastric cancer

Abstract

Background The purpose of this study was to determine the learning curve for robot-assisted gastrectomy using the Cumulative Summation (CUSUM) technique. Methods Two series of consecutive robotic gastrectomy were retrospective analyzed. Patient demographics, surgical performance, and short-term outcomes were examined and data of operation time were abstracted for the learning curve analysis. Results Similar processes occurred in the two surgeons. Each of their learning curves of robotic gastrectomy was best modeled as a third-order polynomial, with equation CUSUMOT in minutes equal to 0.0495 case number 3-4.217 case number2 + 91.206 case number-100.11(R2 = 0.8731) for surgeon A and 0.0314 case number 3-2.4106 case number2 + 33.682 case number + 315.28(R2 = 0.8816) for surgeon B. They both included three unique phases: an initial phase, a well-developed phase, and a mastery phase after the accumulation of additional experience. Conclusion The CUSUM method is a useful tool for objective evaluation of practical skills for surgeons during the learning phase of robotic surgery training. The robotic gastrectomy is found to have a short learning curve for experienced laparoscopic surgeons and the popularity of this new technology won't reduce because of its difficulty to learn.