Study design, materials and methods
Data from three prospective multicenter cohort studies conducted in the Netherlands between 2006 and 2021 were used for the development of this prediction model. All studies included women who underwent vaginal POP surgery with anterior colporrhaphy without the use of mesh. Composite outcome was defined as anatomical recurrence beyond the hymen and/or subjective recurrence and/or retreatment (surgery or pessary) within one year postoperatively. Candidate predictors were selected based on previously identified risk factors from a validated model for anatomical recurrence, additional risk factors reported in the literature, and variables that showed univariable associations with recurrence in the present dataset. Multivariable logistic regression with backward elimination was applied, and internal validation was performed by bootstrapping across imputed datasets to assess discrimination and calibration.
Results
A total of 533 women were included. The final model comprised five predictors: preoperative Pelvic Organ Prolapse Quantification (POPQ) stage 3 or 4 in any compartment, the levator hiatal area during Valsalva on three-dimensional (3D) ultrasound, the patient-reported preoperative burden score for prolapse, active employment status, and the Manchester Fothergill procedure (protective). After internal validation, the optimism-corrected Nagelkerke R2 was 0.12, Brier score 0.18, and area under the receiver operating characteristic curve (AUC) 68.2%, with excellent calibration. Proposed clinical thresholds of 30% and 70% recurrence risk yielded a negative predictive value of 81.4% and a positive predictive value of 91.4%, supporting use in shared decision-making.
Interpretation of results
This study presents a simple and clinically applicable model for predicting composite POP recurrence after native tissue repair, incorporating both clinical and patient-reported factors. The model demonstrated fair discrimination and excellent calibration, offering a valuable tool for individualized risk assessment and preoperative counseling. Before clinical implementation, external validation is essential.