Meta-analysis of four different surgical treatments for stress urinary incontinence.

INTRODUCTION: Multiple surgical approaches for the correction of stress urinary incontinence (SUI) have been described. In an attempt to determine which procedure may be superior, we performed a meta-analysis comparing the needle suspensions (i.e. Gittes, Stamey), Raz, Burch/MMK colposuspension, and urethral sling. METHODS: The most complete 87 of 310 articles relating to the surgical therapy for SUI published between 1982 and 1995 were selected by Medline search for review. These articles incorporated patients operated on from 1965 to 1995. Fixed effects Baysian and variance weighted models were applied to derive estimates and confidence intervals. Bonferroni multiple comparison procedures wee also incorporated. RESULTS: The mean reported success rates were 81% for the needle suspensions, 84% for the Raz, 85% for the colposuspension, and 85% for the sling. There was a significant difference between the sling and needle suspension as well as between the colposuspension and needle suspension (P<0.05). There was no correlation between length of follow-up and the success rate (P>.05). CONCLUSION: There is currently a general impression among urologists that the sling has the best success rates for the treatment of SUI and that the Raz and colposuspension are clinically superior to the needle suspension. Recent literature also indicates a significant increase in the failure rate over time. Our meta-analysis of the best 25% of SUI papers does not support either view entirely. This may be due to publication bias and incomplete reporting by the authors. Prior to developing ever "improved" procedures, urologist must start to report their results in a consistent fashion to allow meaningful conclusions concerning the risks and benefits of our existing procedures.