Establishment and validation of a risk prediction model for postpartum stress urinary incontinence based on pelvic floor ultrasound and clinical data

Abstract Introduction and hypothesis This study aimed to establish a risk prediction model for postpartum stress urinary incontinence (SUI) based on pelvic floor ultrasound measurement data and certain clinical data. Methods Singleton pregnant women aged ≥ 18 years who underwent delivery were selected. All participants were followed up to determine the symptoms of SUI, and pregnancy-related data were collected at the time of registration. Pelvic floor ultrasound was performed at 6–12 weeks postpartum to obtain ultrasonic measurement data. Logistic regression analysis was used to select predictors and establish a nomogram to predict the risk of postpartum SUI. Area under the ROC curve (AUC) values and calibration curves were used for discrimination and calibration, respectively. Finally, external verification of the model was carried out. Results A total of 255 participants were included in the analysis, comprising 105 in the postpartum SUI group and 150 in the non-SUI group. Logistic regression analysis identified age, parity, vaginal delivery, bladder neck descent (BND), and angle of internal urethral orifice funnel as risk factors for postpartum SUI (all P < 0.05). Conclusions We constructed a prediction model for postpartum SUI based on pelvic floor ultrasound measurement data and certain clinical data. In clinical practice, this convenient and reliable tool can provide a basis for formulation of treatment strategies for patients with postpartum SUI.