Learning curves of 13 surgeons for robot-assisted radical prostatectomy in a single institution: role of early case density
摘要
Robot-assisted radical prostatectomy (RARP) is widely used for localized prostate cancer, yet learning trajectories among multiple surgeons within the same institutional program are not well defined. In this single-center study, we described surgeon-specific learning curves for RARP and examined whether early case density was associated with achieving technical proficiency. We retrospectively reviewed 348 non-proctored RARP cases performed by 13 surgeons at a single tertiary center. For learning curve analysis, 279 standard procedures without extended pelvic lymph node dissection or planned nerve-sparing were included. Surgeon-specific logistic regression models used pneumoperitoneum time ≤ 230 min as the binary outcome and cumulative case number as the explanatory variable; proficiency was defined as the case number at which the predicted probability reached 0.8. Early case density was quantified as 5-case and 10-case densities (cases/day) and evaluated using linear regression. Logistic models could be estimated for 10 surgeons, and 7 (70%) reached the proficiency threshold. The case number at which proficiency was achieved ranged from 16 to 38 (median 26). Higher early case density was associated with a lower case number at proficiency; this association was significant for the 10-case density (P = 0.048) and showed a trend for the 5-case density (P = 0.054). In a single-institution RARP program with standardized technique and perioperative pathways, most surgeons achieved the predefined pneumoperitoneum-time–based proficiency threshold after several dozen cases. Higher early case density was linked to a lower case number at proficiency, suggesting that concentrating initial RARP exposure may support more efficient training and case allocation.