April 12-13, 2000
REPORT OF SUB-COMMITTEE COMPOSED OF:
Alan Wein, MD (Committee Chair)
Jerry Blaivis, MD
Roger R. Dmochowski, MD
Edward McGuire, MD
Dianne Newman, RNC, MSN, CRNP, SAAN
Peggy Norton, M.D
Christopher Payne, MD
Nancy Reilly, RN, MSN, CURN, CRNP
The American Urological Association based its position statement upon literature review of evidentiary support for specific therapies for treatment of urinary incontinence, including behavioral modification, biofeedback and electrical stimulation. Supporting documents are herein enclosed, including:
The preceding materials were reviewed in light of the efficacy of the previously mentioned treatments from the standpoint of adequacy of evidence supporting the utilization of the materials and also the size (magnitude) of the health effect of these interventions.
Summary of Recommendations:
With respect to biofeedback, there is much confusion regarding use of this term when in effect, most investigators are really referring to biofeedback assisted behavioral modification. This makes it extremely difficult to evaluate the effectiveness/noneffectiveness of biofeedback as it is classically defined. Although there does not appear to be an overwhelming consensus based upon objective data for the use of biofeedback in the treatment of urge and/or stress incontinence, the committee would like to recommend a program of behavioral modification as effective in the treatment of urge and/or stress incontinence. Behavioral modification includes toileting assistance, bladder retraining (educational instruction in volitional changes in voiding habit with verbal prompting and specific scheduling), pelvic floor muscle exercises (Kegel type), and volume of oral intake modification. Further, on the basis of expert opinion based somewhat on results and literature, the committee would like to go on record as recommending integral biofeedback within such a program of behavioral modification. Treatment sessions are recommended no more than once weekly, not to exceed a total of ten sessions in one year. The selection of frequency and total number of treatments is obviously arbitrary. The committee recommended that further research in this area be done, specifically to include: 1) trials of behavioral modification alone versus behavioral modification with biofeedback assistance; and 2) a comparison of the various biofeedback vaginal, rectal pressures, EMG monitoring, including frequency and an ideal number of sessions, 3) the utilization of the aformentioned techniques coupled with chronic maintenance biofeedback in the home setting.
Based on the committee's literature review, we believe that biofeedback represents an approach that would produce results that would approximate a value between a Level 3 (as effective but with advantages) and a Level 4 (as effective but with no advantages). This recommendation is based on HCFA's seven categories of effectiveness as compared to standard behavioral and voiding retraining therapies.
Based upon all available evidence and expert opinion, the committee does not feel there is strong consensus in the effectiveness of vaginal, suprapubic and/or anal electrical stimulation for urge, stress or mixed incontinence. The committee does urge further randomized control trials be done, looking at all objective and subjective parameters related to incontinence and such trials should include a comparison of electrical stimulation to behavioral modification programs within various populations (men, women, the elderly, children and the neurogenic population).
Given the above considerations, isolated electrical stimulation achieves a rating of Level 5 (less effective but with advantages); however, some data do exist to support a Level 4 (as effective but with no advantages).
The committee does not feel that the recommendations for the use of biofeedback or functional electrical stimulation for treatment of urinary incontinence put forth by the American Urogynecologic Society are supported by the references which they site.
The opinion regarding neuromodulation will be the subject of a second opinion paper for future meetings.
Summary of Evidence
The committee heavily relied on references 1,2,3 in establishing its position statement. The most current meta analytic review of literature specific to conservative therapy for urinary incontinence (UI) is that provided in reference 3, which adds further data retrieval to that contained in reference 2. Strict methodologic analysis in these references make clear that many articles which evaluate specific interventions often evaluate these interventions within the context of a physiotherapeutic program. The importance of the program cannot be understated and the elements involved provide a background for the more involved interventions of biofeedback and electrical stimulation.
Berghmans (ref 3) identified fifteen RCT (randomized controlled trial) articles which had supportive methodology of reasonable ("moderate") quality to evaluate specific interventions. Eight of those articles have sufficient quality to identify the effect of interventions. The general conclusion was that bladder retraining was more effective than no treatment. Studies evaluating electrical stimulation demonstrated substantial variability in stimulation type and parameters. Berghmans concluded that insufficient evidence existed to state that active electrical stimulation was more effective than sham electrical stimulation treatments of the pelvic floor. Insufficient evidence existed to support biofeedback assisted pelvic floor exercises against pelvic floor exercises alone in treating UI. However, all the included studies did report some favorable results for these various interventions in the treatment of UI. These positive results were limited in applicability by the methodologic weaknesses on the specific articles.
Weatherall (ref 8) embarked on a similar methodologic analysis of the literature (however using a more rigorous analysis of number of included patients and variability of treatment outcomes) specifically evaluating the hypothesis that biofeedback was no more effective than pelvic floor muscle exercises for the treatment of UI in women. The author identified five RCTs, of which two had significant analytic or outcome measure flaws which excluded their utilization. The three remaining articles demonstrated a trend in favor of biofeedback efficacy over that achieved with pelvic floor muscle exercises alone. Weatherall concluded that prior trials lacked statistical power to detect moderate trends of efficacy between therapy types (and that this also explained the different conclusion obtained in contrast to that of Berghmans).
Wyman (ref 9) extensively evaluated bladder training alone versus pelvic muscle training with biofeedback assistance versus a combination arm utilizing both approaches for acute and intermediate time frame (three month) effect. Although initially the combination arm had the most benefit, at three months, all three arms showed similar results. The authors concluded that all three arms showed benefit, but that the most important component of therapy was "a structured intervention program with education, counseling, and frequent patient contact."
The salubrious effects of behavioral therapy were also noted by McDowell et al (ref 7) in a group of home-bound elderly. In RCT format two groups (biofeedback assisted pelvic floor muscle exercises versus a control group receiving reinforcement only) were treated with crossover of the control group at the end of the initial study phase. Substantial efficacy was noted in the active arm of the trial, which was repeated during the crossover arm of the study. The only limitation to response was cognitive inability to comply with exercise protocol. Therefore this type of therapy would appear to have applicability in the Medicare aged population.
Finally, the World Health Organization statement on pelvic floor therapies attempted to exhaustively review the world's literature on behavioral therapies for incontinence. That committee concluded that there was anecdotal evidence to support biofeedback as a separate and unique component to pelvic floor physiotherapies in select groups (i.e. patients unable to perform voluntary pelvic floor muscle exercises). However, they stressed a structured program with pelvic floor exercises in addition to structured voiding programs as an efficacious (65 - 70 % cure or improvement rate) and reasonable treatment protocol.
Electrical stimulation was not felt to have sufficient validation to determine it's efficacy and the committee agreed that more research was needed. Additionally, that committee did not add electrical stimulation to their algorithm for therapy of urinary incontinence.
What is clear from this truncated literature review is that more trials with defined and accepted methodologies and outcome parameters are necessary to fully understand the role and overall effect of biofeedback and electrical stimulation in the treatment of UI.