(430) Definition Of Risk Categories For Urethral Stricture Recurrence After Free Graft Bulbar Urethroplasty Using Classification And Regression Tree Analysis

Dr Enzo Palminteri,Dr Stefano Toso,Dr Giovanni Ferrari,Dr Luciano Nava,Dr Nicolò M Buffi,Dr Manuela Pane,Dr Mirko Preto,Dr Andrea Gobbo
DOI: https://doi.org/10.1093/jsxmed/qdae041.059
2024-06-15
The Journal of Sexual Medicine
Abstract:Objectives This study aims to stratify patients undergoing buccal mucosa bulbar urethroplasty (BMU) for urethral stricture into risk groups (RGs) for stricture recurrence at 5y, providing clinicians with an easy-to-use tool useful in practice. Methods To define the RGs for stricture recurrence we used classification and regression tree (CART) analysis. In the CART we included the variables age, stricture length, previous dilatation/urethrotomy, and stricture etiology, that were further selected and split by the program to produce the optimal tree. The final leaves were used to define three RGs for stricture recurrence. To test the predictive accuracy of the RGs, we compared the accuracy in the logistic regression model with the RGs against the model with all the variables used to generate the CART. We repeated the same process in a second CART, also including the first postoperative uroflowmetry Qmax performed within two months after surgery. Results The analysis included 855 BMU, of those 93 experienced stricture recurrence. At logistic regression, the first CART-generated RGs were significant predictors of treatment failure (p<.001) and performed equally well when compared to the standard model with all variables, with an accuracy of 62%. The second CART-generated RGs were also significant predictors of failure (p<.001) and performed as the standard model with all variables, with an accuracy of 77%. The prediction accuracy was significantly higher in the RGs from the second CART (p<.001), which included the Qmax. Conclusions CART analysis assisted in creating three RGs with an increasing recurrence risk at 5y. At logistic regression, RGs performed equally well when compared to the relative standard models. Using RGs has the significant advantage of being easy to interpret and implement into clinical practice, and the stratification of patients can be used to tailor follow-up. Our analysis proved that the first postoperative uroflowmetry can significantly improve patient classification. Conflicts of Interest No conflict of interest to disclose.
urology & nephrology
What problem does this paper attempt to address?