Nonsurgical treatment of stress urinary incontinence

Authors


Sir,

I read with interest the review of systematic reviews of nonsurgical treatment of stress urinary incontinence by Latthe et al.1 They sought to assess the quality of these reviews by applying the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system. By using this system to provide rigorous and objective assessment of the evidence, they then argue this is useful in generating recommendations. For example, this study generates results that concur with National Institute of Clinical Evidence recommendations, namely that a trial of supervised pelvic floor muscle training (PFMT) of at least 3-month duration should be offered as first-line treatment to women with stress urinary incontinence.

What has failed to be assessed in this GRADE study is the long-term outcome in women with stress urinary incontinence who have nonsurgical treatments. There is a perception among practising clinicians that many women treated nonsurgically will show an initial improvement in their symptoms but with the passage of time will re-present with a recurrence of their stress urinary incontinence. A surgical mode of treatment is then likely to be offered. One could argue that for many women, a trial of supervised PFMT is a diversion on the route of getting more definitive surgical treatment.

The authors conclude that the GRADE system appropriately balances the need for simplicity with the need for a full and transparent consideration of all important issues. Hence, it should be used for generating recommendations for clinical practice in urogynaecology. In this case, the GRADE system has failed to take into account the important issue of permanence of nonsurgical treatment of stress urinary incontinence. Until the important issue of permanence (or not) of nonsurgical treatment of stress urinary incontinence is satisfactorily addressed, research tools such as the GRADE system might produce results and generate recommendations that ultimately prove to be wrong.

Ancillary