Objectives: To assess the impact of the learning curve (LC) on perioperative and long-term functional outcomes of a consecutive single-centre series of robot-assisted radical cystectomy with Padua intracorporeal orthotopic neobladder.
Patients and methods: Patients treated between 2013 and 2022 were included, with ≥1 year of follow-up. The entire cohort was divided in tertiles. Categorical and continuous variables were compared. Joinpoint regression analysis was used to identify significant changes over the decade in linear slope of the 1-year day- and night-time continence. Uni- and multivariable Cox regression analyses identified predictors of day- and night-time continence recovery. Day-time continence was defined as 'totally dry' (no pads), night-time continence as pad wetness ≤50 mL (one safety pad).
Results: Overall, 200 patients were included. The mean hospital stay (P = 0.002) and 30-day complications (P = 0.04) significantly reduced over time; the LC significantly impacted on Trifecta achievement (P < 0.001). The 1-year day- and night-time continence probabilities displayed a significant improving trend (day-time continence annual average percentage change [AAPC] 11.45%, P < 0.001; night-time continence AAPC 10.05%, P = 0.009). The LC was an independent predictor of day- (hazard ratio [HR] 1.008; P < 0.001) and night-time continence (HR 1.004; P = 0.03) over time.
Conclusion: Patients at the beginning of the LC had significantly longer hospitalisations, more postoperative complications, and lower Trifecta rates. At the 10-year analyses, we observed a significant improving trend for both the 1-year day- and night-time continence probabilities, highlighting the crucial role of the LC. However, we are unable to assess the case volume needed to achieve a plateau in terms of day- and night-time continence rates.
Keywords: bladder cancer; functional outcomes; intracorporeal orthotopic neobladder; learning curve; robotic cystectomy; urinary continence.
© 2024 BJU International.