External validation of de novo stress urinary incontinence prediction model after vaginal prolapse surgery.

Abstract

Stress urinary incontinence (SUI) may appear after the correction of pelvic organ prolapse (POP). The aim of this study was to externally validate a described predictive model for de novo SUI and to assess its clinical performance when used as a diagnostic test. This was a retrospective descriptive study on a cohort of consecutive women treated in our institution. The main outcome used to validate the model was the presence of objective or subjective SUI 1year after surgery. A receiver operating characteristic curve was generated from our population to evaluate the predictive accuracy and to compare it with the original model. A cutoff point of ≥50% was used to evaluate its clinical performance as a diagnostic test. Of the full cohort, 169 women were suitable for analysis. The rate of de novo SUI was 11.8%. The predictive accuracy of the model in our population was similar to the original [area under the curve (AUC) = 0.69; 95% confidence interval (CI) = 0.58-0.80). However, its performance measures when evaluated as a diagnostic test were low: positive likelihood ratio = 2.71 and negative likelihood ratio = 0.86. Only 15 women presented a positive test result. External validation of the model found a global predictive accuracy similar to that of the original model. Despite the study being underpowered to give firm conclusions, the test did not show a good clinical performance when applied to our population with low de novo SUI prevalence. A larger sample size is needed to validate the model conclusively.