A predictive model for the failure of the first placement of ureteral access sheath (12/14Fr): a multi-center retrospective study

Abstract

Background: Method for predicting failure of the first placement of ureteral access sheath (UAS) are currently lacking. Methods: All patients were enrolled in two large medical centers from 2015 to 2022. Univariate and multivariate logistic regression analysis were conducted to identify independent prognostic factors. Based on the results of the multivariate logistic regression models, a nomogram was developed. The consistency index, receiver operating characteristic curve (ROC), calibration curve, decision curve analysis, and kappa-test was used for validating efficiency of the prognostic nomogram. Results: 563 eligible patients were enrolled in the study and then randomly divided into a training cohort (n=299) and a validation cohort (n=213). Multivariate logistic regression analysis indicated that the cross-sectional area of the lower ureter, history of calculi expulsion from the urethra, and normal ureteral wall thickness of the lower ureter were independent risk factors associated with failure of the first placement of UAS. The C-index was 0.980 and 0.968 in the training and validation cohorts, respectively. The area under the curve (AUC) was 0.986 and 0.961 in the training and validation cohorts, respectively. Moreover, calibration curves and decision curve analysis demonstrated that the nomogram exhibited favorable predictive accuracy in the training cohort. Ultimately, the kappa test confirmed a significant agreement between predicted outcomes and actual outcomes in 51 patients (kappa value = 0.538, P < 0.001). Conclusion: Our study provided an individualized risk assessment for the initial placement of UAS in patients with renal or ureteral calculi using a prognostic nomogram. Additionally, based on this prediction, urologists can advise patients to choose between ureteral stenting or flexible ureteroscopy prior to surgery.