Long-term results of surgery for urethral stricture: a statistical analysis

We assessed the long-term results of surgery for urethral stricture by statistical methods using guidelines for survival analysis.Of the 267 patients who underwent surgery for urethral strictures between March 1993 and December 1999, 238 (89%) were followed prospectively. The primary end point was success rate of urethroplasty. Secondary end points were variables that might influence success rates. Survival data were analyzed by the Kaplan-Meier method and Cox regression using guidelines for survival analysis.The success rate of all urethroplasties was 82% (95% CI 66 to 89) at 7.5 years. A 6.5-year success rate of 86% (95% CI 70 to 95) was achieved by flap, of 86% (95% CI 66 to 100) by anastomotic, of 79% (95% CI 64 to 94) by mesh graft and of 56% (95% CI 20 to 100) by graft urethroplasty. No significant difference was found among the surgical procedures using the log rank (p = 0.269) or Gehan-Wilcoxon (p = 0.259) test. Multivariate analysis showed an increased risk of failed urethroplasty in patients previously treated with urethral stents (HR 3.69, 95% CI 1.27 to 10.8) and by 2 or more urethrotomies (HR 2.25, 95% CI 1.05 to 4.8).Using standardized success criteria and statistical methods that applied guidelines for survival analysis conclusions were drawn. Excellent results are achievable by all forms of urethroplasty, justifying a more liberal indication for urethral reconstruction.