Introduction: Urethroplasty is generally regarded as the most effective treatment of urethral stricture. However, recurrences and complications can occur even in the most skilled hands and reliably predicting these outcomes remains elusive. Several existing classification systems (U-Score and LSE) exist but remain largely unvalidated. The purpose of this study is to examine stricture-specific factors associated with recurrence after anterior urethroplasty in order to better refine existing classification systems. Methods: A retrospective review was performed of men undergoing anterior urethroplasty at a single center from 2003-2021. Stricture-specific variables of location, etiology, length, number, prior urethroplasty and previous endoscopic treatment were identified. Additionally, U Scores (US) and LSE Scores were calculated. Success was defined as easy passage of a flexible cystoscope at routine follow-up with no change in urinary function thereafter. Complications were defined as a Clavien >1 complication during the 90-day perioperative time period. Associations between clinical variables and stricture recurrence were evaluated using univariable and multivariable Cox regression analysis. Variables independently associated with stricture recurrence were sub-stratified using Kaplan-Meier analysis and arranged into a survival-analysis based classification system. Receiver operator characteristic (ROC) analysis was performed on U-Score (US), LSE and refined classification system. Results: 1573 patients underwent anterior urethroplasty over the study period with a median patient age of 47 years (IQR 34-59) and stricture length of 4 cm (IQR 3-6). Urethroplasty success was 92.0% (1447) at a median follow-up of 90 months. On multivariable Cox regression, stricture length (Hazard Ratio 1.09, 95%CI 1.04-1.16, p=0.001), etiology (HR 1.16, 95%CI 1.06-1.28, p=0.002), revision urethroplasty (HR 1.56, 95%CI 1.07-2.28, p=0.02), stricture number (HR 2.34, 95%CI 1.01-7.43, p=0.05), and location (HR 1.32, 95%CI 1.04-1.68, p=0.02) were independently associated with stricture recurrence after anterior urethroplasty. On Kaplan-Meier analysis there was clustering within each independent variable which allowed revision to a stricture-specific classification termed “LERNS” (Length, Etiology, Revision, Number and Segment). On ROC analysis, area under the curve (AUC) for LERNS indicated excellent discrimination as a predictive tool for stricture recurrence (AUC 0.76, 95%CI 0.71-0.80, p<0.001). This was superior to both US (AUC 0.71, 95%CI 0.66-0.75, p<0.001) and LSE (AUC 0.69, 95%CI 0.64-0.74, p<0.001) confirmed with bootstrap analysis using 1000 replicates. On ROC analysis, both US (AUC 0.57, 95%CI 0.51-0.62, p=0.01) and LERNS (AUC 0.58, 95%CI 0.53-0.63, p=0.001) significantly predicting 90-day complications but with overall poor diagnostic discrimination. Conclusions: While increasing U-Score and LSE scale are both associated with stricture recurrence, modifying these systems as “LERNS” (Length, Etiology, Revision, Number, Segment) provides excellent discrimination when predicting stricture recurrence. However, accurately predicting 90-day complications will likely require further incorporation of patient-specific parameters. SOURCE OF Funding: None
.jpg "Keith Francis Rourke, MD photo")