Step-specific temporal patterns in robot-assisted sacrocolpopexy: an exploratory CUSUM analysis

Learning curves for robot-assisted sacrocolpopexy (RSC) are commonly evaluated using total operative or console time, although RSC consists of multiple technically distinct steps. We explored step-specific temporal efficiency patterns using cumulative sum (CUSUM) analysis. We retrospectively analyzed the first 29 consecutive RSC procedures performed by a single surgeon with approximately 100 prior robotic procedures and experience in robotic surgical instruction. Seven procedural steps were timed using prespecified video landmarks. CUSUM curves and Spearman correlations with bootstrap 95% confidence intervals were used to describe temporal efficiency patterns. The console-time CUSUM peak was used only for a descriptive post hoc early-versus-late illustration and was not treated as an independently validated learning transition. Median operative and console times were 183 and 139 min, respectively. Neither total operative time nor total console time showed a significant monotonic relationship with case sequence, although both mean-centered CUSUM curves reached their maximum positive deviation at case 14. Step-specific temporal patterns varied substantially. Hysterectomy, anterior and posterior mesh suturing, and peritoneal closure showed negative correlations with case sequence, whereas bladder dissection showed the opposite trend and a later CUSUM peak. Early-versus-late comparisons were descriptive and did not independently validate a transition point. RSC demonstrated heterogeneous step-specific temporal efficiency patterns that were not captured by total console time alone. Because the global time trends were flat, the phase cutpoint was data-derived, and technical quality and safety were not assessed, these findings should be considered exploratory and hypothesis-generating rather than evidence of proficiency or training effectiveness.