Background: Periurethral or transurethral injection of bulking agents is a surgical procedure most often used for the treatment of stress urinary incontinence a common, troublesome symptom amongst adult women.
Objectives: To assess the effects of periurethral/transurethral injection therapy in the treatment of urinary incontinence in women.
Search strategy: We searched the Cochrane Incontinence Group Specialised Trials Register (28 February 2007), MEDLINE (January 1996 to March 2007, PREMEDLINE (7 February 2007) and the reference lists of relevant articles.
Selection criteria: All randomised or quasi-randomised controlled trials of treatment for urinary incontinence, in which at least one management arm involved periurethral/transurethral injection therapy.
Data collection and analysis: Two reviewers independently assessed methodological quality of each study using explicit criteria. Data extraction was undertaken independently using a standard form and clarification concerning possible unreported data sought directly from the investigators.
Main results: We identified twelve trials including 1318 women that met the inclusion criteria. The limited data available were not suitable for meta-analysis. Injection of autologous fat was compared to placebo in a study of 68 women which was terminated early because of safety concerns. No differences in subjective or objective outcome were found in the two groups. No studies were found comparing injection therapy with conservative treatment. Two studies that compared injection with surgical management found significantly better objective outcome in the surgical group. Eight studies compared different agents - all results had wide confidence intervals. Silicone particles, calcium hydroxylapatite, ethylene vinyl alcohol and carbon spheres gave improvements equivalent to collagen. Porcine dermal implant gave improvements comparable to silicone at six months. A comparison of periurethral and transurethral methods of delivery of the bulking agent found similar outcome but a higher rate of early complications in the periurethral group.
Authors' conclusions: Despite five additional trials, this updated review is still an unsatisfactory basis for practice. The trials were small and generally of moderate quality. The only evidence of benefit comes for within-group short-term changes following injection. The finding that placebo saline injection was followed by a similar symptomatic improvement questions the mechanism of any effects. There were no trials in comparison with pelvic floor muscle training -the obvious non-surgical comparator. Greater symptomatic improvement was observed after surgery, although these advantages need to be set against likely higher risks. No clear-cut conclusions could be drawn from trials comparing alternative agents; one small trial suggests that periurethral injection may carry more risks than transurethral injection. The single trial of autologous fat provides a reminder that periurethral injections can occasionally cause serious side-effects. Pending further evidence, injection therapy may represent a useful option for short-term symptomatic relief amongst selected women with co-morbidity that precludes anaesthesia - two or three injections are likely to be required to achieve a satisfactory result.