Stress urinary incontinence after vaginal prolapse repair: development and internal validation of a prediction model with and without the stress test

Abstract

Objective: To develop a prediction model for stress urinary incontinence (SUI) after vaginal prolapse repair (postoperative stress urinary incontinence [POSUI]) and assess the value of a preoperative stress test. Patients and methods: Secondary analysis of two trials in which women were randomised for prolapse repair with or without a midurethral sling (MUS). The trials included women with (CUPIDO-1, n = 134) and without (CUPIDO-2, n = 225) coexisting SUI. POSUI was defined as bothersome SUI one year after surgery and/or treatment of SUI in the first postoperative year. Logistic regression analysis was used to define a reference model, which was extended with the preoperative stress test. The stress test was performed with and without reduction of the prolapse. Missing values were imputed 20 times, with bootstrap resampling for internal validation of discriminatory ability. Results: Three hundred fifty-six women could be included. POSUI occurred in 17% of the women (n = 61). The reference model included age (<55 years), point Ba of the pelvic organ prolapse quantification system (<−1), vaginal parity (≤3), subjective urinary incontinence, and MUS. The stress test had an odds ratio of 2.4 (95% confidence interval [CI], 1.2-4.6) in the extended model, which increased the optimism-corrected area under the receiver-operating curve from 0.74 to 0.76. The stress test was especially valuable in women with a 10% to 30% POSUI risk, where a stress test substantially impacted the POSUI risk. In more than 50% of the women, the stress test had no additional value in predicting POSUI. Conclusion: A preoperative stress test is not valuable for women at low risk of SUI after vaginal prolapse repair.