Abstract
OBJECTIVE: To develop a risk stratification model for predicting urinary incontinence following radical prostatectomy (RP) in high-risk prostate cancer patients based on their clinicopathological characteristics. METHODS: A retrospective analysis was conducted on 520 prostate cancer patients who underwent RP between January 2016 and January 2024. Baseline characteristics, pathological data, laboratory parameters, and surgery-related factors were collected. Urinary continence status at 1, 3, and 6 months postoperatively was assessed. Multivariate logistic regression analyses were performed to identify independent risk factors, and nomograms were constructed to predict urinary incontinence risk at each time point. RESULTS: Urinary incontinence rates at 1, 3, and 6 months postoperatively were 92.88%, 69.62%, and 23.65%, respectively. At 1 month, a higher Gleason score (OR=2.178, P=0.003) was a risk factor, while robot-assisted surgery was protective (OR=0.289, P=0.003). At 3 months, higher Gleason score (OR=1.565, P=0.004) increased risk, whereas lower BMI (<25 kg/m(2)) (OR=0.448, P=0.005) and longer preoperative membranous urethral length (≥14 mm) (OR=2.368, P<0.001) were protective. At 6 months, shorter membranous urethral length (<14 mm) (OR=3.622, P<0.001), neoadjuvant hormone therapy (OR=5.783, P<0.001), and higher Gleason score (OR=2.824, P<0.001) were risk factors, while lower BMI (OR=0.317, P<0.001), smaller prostate volume (<40 mL) (OR=0.591, P=0.044), and lower CONUT score (<4) (OR=0.372, P<0.001) were protective. The nomograms showed good predictive performance, with AUCs of 0.679 at 3 months and 0.818 at 6 months. CONCLUSIONS: The developed nomograms effectively stratify the risk of urinary incontinence following RP in high-risk patients, facilitating individualized perioperative management and rehabilitation strategies.