Abstract
BACKGROUND: Urinary incontinence is a disease that can be easily neglected but affects women's health, with stress urinary incontinence (SUI) being the most common type. We analyzed multiple parameters, including medical history, physical examination, surface electromyography, and pelvic floor ultrasound findings, to identify factors associated with SUI and its severity and integrated these factors into predictive models. METHODS: In this retrospective study, a consecutive series of women admitted to Chenghua District Maternal and Child Health Hospital (Chengdu, China) between November 2022 and February 2024 were stratified according to whether they were diagnosed with SUI or not (with 249 cases in each group). Patients with SUI were further stratified by the Ingelman-Sundberg hierarchical method and 1-hour urinary pad test. Results from medical history, physical examination, surface electromyography, and pelvic floor ultrasonography were compared between the different groups, and multivariate regression models were derived to predict the risk of SUI among all study participants or to distinguish the severity of the condition in the participants. Diagnostic performance of the models was assessed in terms of the area under the receiver operating characteristic curve (AUC). According to the same inclusion and exclusion criteria, 74 patients with non-SUI and 46 patients with SUI from the same medical center between April and June 2025 were recruited into an external validation set. RESULTS: We constructed a model that incorporated age, gravida, parity, bladder neck position, levator hiatus area, bladder neck descent, body mass index, delivery mode, pelvic floor muscle strength, point anterior wall point A (Aa) position, phasic contractions, tonic contractions, variability, urethral rotation angle, and posterior urethrovesical angle to predict SUI, which achieved an AUC of 0.861 [95% confidence interval (CI): 0.829-0.893]. Another model, which incorporated age, gravida, parity, bladder neck position, levator hiatus area, and bladder neck descent, could distinguish the severity of SUI, with an AUC of 0.783 (95% CI: 0.742-0.822). CONCLUSIONS: The combined model is reasonably reliable in predicting the risk of SUI but slightly less reliable for distinguishing its severity.