The evidence on pelvic floor muscle training (PFMT) in women with urinary incontinence

Abstract

Dear Editor, We are grateful for the opportunity to respond to the comments of Bary Berghmans and Kari Bø to our review 1. A systematic review has been defined as ‘an overview of primary studies that used explicit and reproducible methods,’ and correspondingly a meta-analysis as ‘a mathematical synthesis of the results of two or more primary studies that addressed the same hypothesis in the same way’ 2. Following these universally accepted definitions, we did a comprehensive and thorough systematic review: we used clearly stated definitions and search strategies, as well as inclusion and exclusion criteria, for the studies and data used in our review. We made very clear in our manuscript that we did not do a meta-analysis. Neither did the other systematic reviews, whose conclusions seem to appeal more to Berghmans and Bø 3-6, it is simply not possible due to poor quality and heterogeneity of studies. We do believe, however, that out extraction, calculation, and tabulation of cure rates (with confidence intervals) represents one of the more comprehensive and transparent reviews on the topic. Our review is no less systematic, and no more narrative, than the other reviews cited 3-6, or indeed the letter of Berghmans and Bø. The difference in the story-telling seems to be which aspects of the inadequate evidence one chooses to highlight. While Wilson et al. have been oft-quoted for concluding that ‘PFMT should be offered, as first-line treatment’ to women with UI, they were also careful to state that ‘it is disappointing that so few data were available’ and that they found ‘no differences in quality of life’ between PFMT and control groups 3. While the Cochrane-review of Dumoulin and Hay-Smith is contrasted by Berghmans and Bø as presenting a ‘totally different conclusion’ to our review, they did in fact note, as we did, that there is a lack of data on the long-term efficacy of PFMT, and only a modest immediate response 5. The fact that our discussion and conclusions are unappealing to Berghmans and Bø does not justify labeling our manuscript as ‘low-quality.’ We do accept criticism for references missed. Berghmans and Bø are kind enough to point out six studies we missed 7-12. None of our oversights were intentional. We should not be blamed for excluding a Spanish-language reference 9. Omitting the one-year follow-up of Mørkved does not alter conclusions 10. Berghmans and Bø take issue with our labeling of some large, well-designed randomized controlled trials (RCT) as ‘impressive’ despite the fact these studies only used ‘low-dose’ PFMT 13, 14. However, we contend that these studies reflect much better the realities of everyday clinical practice. Prescribing high-intensity long-term PFMT as utilized by Bø and others in research settings is simply not realistic for most of us. It seems Berghmans and Bø take issue mostly with the critical language we used, not so much with the facts we present. They claim we are biased and without consideration for a corresponding critical appraisal of drugs or surgery. How wrong they are! The starting point of our review was to apply the exact same critical standards we use to evaluate drug and surgery evidence. We are not ‘against’ PFMT, we prescribe it in our daily clinical practice, we have an excellent professional relationships with its’ practitioners, and after the (mostly) positive reception of our review, we have seriously considered contributing RCT-evidence to the research in the area. We do think there is a problem with labeling people as ‘for’ or ‘against’; with zealous advocacy of PFMT you risk achieving the opposite.