To develop a predictive model based on sphincter- and prostate-related multiparametric MRI (mpMRI) anatomical parameters to guide the preoperative assessment of immediate UC recovery following RS-RARP.

A retrospective cohort of 356 prostate cancer patients who underwent RS-RARP was analyzed. Patients were divided into immediate UC and non-immediate UC groups. Independent predictors for immediate UC recovery were identified through univariate analysis and least absolute shrinkage and selection operator regression. A nomogram was constructed using multivariate logistic regression. Receiver operating characteristic (ROC) curve analysis, Hosmer–Lemeshow test, and decision curve analysis were performed to evaluate nomogram performance.

Immediate UC recovery was observed in 84.55% (301/356) of patients. Multivariate analysis revealed that neurovascular bundle preservation (OR=3.37, P<0.001), Lee type D (OR=4.99, P<0.001), membranous urethral length (MUL) (OR=1.28, P=0.019), and urethral sphincter length (USL) (OR=1.51, P=0.002) were independent risk factors, while intravesical prostatic protrusion length (IPPL) (OR=0.49, P<0.001) and prostate volume (PV) (OR=0.92, P=0.005) were protective. The nomogram integrating MUL, USL, IPPL, and PV demonstrated superior predictive accuracy (AUC=0.89, 95% CI: 0.76-0.85), outperforming individual parameters (DeLong test P<0.05). The Hosmer–Lemeshow test confirmed the goodness-of-fit of the nomogram, and decision curve analysis demonstrated significantly improved net benefits across a range of reasonable threshold probabilities.

This study presents the first mpMRI-based predictive model for immediate UC recovery after RS-RARP, highlighting the synergistic value of membranous urethral morphology and prostatic spatial configuration. The nomogram provides a quantitative tool for preoperative risk stratification and personalized surgical planning, potentially reducing the need for secondary interventions.