Retropubic and transobturator midurethral slings: a decision analysis to compare outcomes including efficacy and complications

The objective of this paper is to compare retropubic (RP) and transobturator (TO) midurethral slings using decision analysis techniques. A decision analysis was constructed including efficacy and complication data. Probability of complication-free surgery and overall utility were analyzed using two models: ALL (all 42 trials) and RCT (seven randomized controlled trials with higher quality data, but missing data on some complications). Surgery was complication-free more frequently with TO approach with 83.7% vs. 55.7% (ALL) and 70.9% vs. 62.8% (RCT). One-year overall utility favored TO in the ALL model (0.943 vs. 0.895). Conversely, the RCT model favored RP (0.936 vs. 0.910). These differences were both less than published minimally important differences (MID) for utilities. Multiple one-way sensitivity analyses confirmed robustness of results. The difference between the two surgeries in both utility-based models was less than the MID. Therefore, the separate approaches are highly comparable with similar overall utility. Complications are more frequent with the retropubic approach.