Unveiling predictors of positive surgical margins in Retzius-sparing and anterograde robot-assisted radical prostatectomy: insights from a dual-centre learning curve experience
摘要
Robot-assisted radical prostatectomy (RARP) is a key procedure in urologic surgical training. Both the learning curve and surgical approach may influence the risk of positive surgical margins (PSMs), but evidence from trainee-performed RARP remains limited. This study aimed to identify independent predictors of PSMs in two trainee cohorts using either the Retzius-sparing or anterograde technique under structured mentorship. Two prospectively maintained institutional databases were reviewed for RARP cases performed by five trainees per centre during their first 50 procedures. The patients who underwent pelvic lymph node dissection or had incomplete data were excluded. Cases were stratified by technique: Retzius-sparing (n = 261) and anterograde (n = 276). Baseline and pathological features were compared. Logistic regression was used to identify independent predictors of PSMs for each group. Compared to the Retzius-sparing group, the anterograde cohort had larger prostate volumes (48 vs 42 mL, p = 0.002), higher biopsy core positivity, and more frequent extrafascial dissections (each p < 0.05). In the Retzius-sparing group, performing at least 20 procedures (OR 0.25, p = 0.03) and longer training duration (OR 0.90 per month, p = 0.04) were associated with lower PSMs risk. In the anterograde group, pathological stage ≥ pT3 (OR 3.74, p = 0.02) and prostate volume < 40 mL (OR 0.29, p = 0.04) independently predicted PSMs. Independent predictors of PSMs differed by surgical approach. Surgeon experience and procedure volume reduced PSM risk in Retzius-sparing RARP, whilst pathological stage and prostate size were key factors in the anterograde group. These results may guide trainee selection and case assignment during early robotic training.