Biofeedback and Physiotherapy Versus Physiotherapy Alone in the Treatment of Genuine Stress Urinary Incontinence



Biofeedback is a method of pelvic floor rehabilitation using a surface electrode inserted into the vagina and a catheter in the rectum. Forty women with genuine urinary stress incontinence were randomized to compare the efficacy of physiotherapy and physiother- apy in combination with biofeedback. The effect of the treatment was determined by a standardized pad-weigh- ing test. Long-term status was determined using a questionnaire after 2-3 years. Thirty-four women com- pleted the treatment. The study showed a statistically significant better improvement in the biofeedback group. The long-term effect in the biofeedback group seemed better and the patients were more motivated for training afterwards.
Keywords: Biofeedback; Stress urinary incontinence; Vaginal surface electrode
Urinary incontinence severely affects the quality of life for many women. Genuine stress incontinence (GSI) is the most common cause of urinary incontinence, and is associated with congenital weakness of the connective tissue, pregnancy, vaginal delivery, the menopause and exacerbation by previous surgery for stress incontin- ence. The role of active exercise for the pelvic floor muscu- lature has been known Since 1948, when Kegel des- cribed such exercises using a perineometer to determine the strength of the contractions [1]. Pelvic floor exer-
Correspondence and offprint requests to: K. Glavind, Department of Gynecology and Obstetrics, Aalborg Sygehus, 9000 Aalborg, Den- mark.
cises are accepted as first-line treatment for GSI: numerous studies have shown the effect of physiother- apy [2-5], but fully satisfactory results have never been achieved, perhaps because many women are not aware of how to contract the pelvic floor muscles or how to avoid a simultaneous rise in intra-abdominal pressure. Biofeedback is a method which teaches this [6]. We present a prospective randomized study comparing the efficacy of physiotherapy alone with physiotherapy in combination with biofeedback. Long-term status was measured after 2-3 years by a questionnaire.

Materials and Methods :

Forty women with GSI were entered into the study. All patients had a thorough history taken, stating incontin- ence when coughing, laughing, lifting and during physi- cal exercise. A gynecological examination with a stress test was performed. The patients ,were assessed by digital palpation as to their ability to contract the pelvic floor musculature, but we did not distinguish between patients with a weak contraction or those with no contraction at all. Both groups were included in the study. Uroflowmetry and sitting medium-fill water cys- tometry was performed to exclude infravesical obstruction and detrusor instability. Cystometry was performed with the Dantec Etude (Skovlunde, Den- mark), using coughing as the provocative maneuver to elicit detrusor instability. Patients with detrusor instability were excluded from the study. All patients included had normal cystometrograms. Urinary incon- tinence was verified by a positive 1-hour pad-weighing test (>2 g) with a bladder volume of three-quarters of the cystometric capacity [7]. All patients leaked urine while coughing and jumping. Patients with severe
۳۴۰ K. Glavind et al.
incontinence were not excluded, but were excluded if previous surgery for urinary incontinence had been performed. All patients who fulfilled the criteria were asked to enter the study. After giving informed consent the women were ran- domized using sealed envelopes to receive treatment with physiotherapy alone or in combination with bio- feedback. The control group was treated according to our standard procedure, with physiotherapy 2-3 times with individual instruction. The biofeedback group received the same physiotherapy treatment plus bio- feedback treatment four times. Biofeedback was performed with a vaginal surface electrode (Dantec 21L20, Skovlunde, Denmark) inserted in the vagina approximately 3 cm from the introitus (Fig. 1) [8]. A rectal catheter was placed approximately 10 cm into the rectum. The patient was instructed in contraction of the pelvic floor and how to avoid intra-abdominal pressure rises as she continuously observed the recording of the vaginal EMG and the intra-abdominal pressure [6]. Each contraction was sustained for 5-10 seconds, the exercises were per- formed supine, sitting and standing up, 10 times in each position. Figure 2 shows the vaginal EMG and the rectal pressure in a patient during biofeedback exercise. Biofeedback was performed weekly, with a total of four lessons per patient. Physiotherapy was performed with 2-3 lessons per patient. The patients received individual tuition, with verbal instruction in the exer- cises and presentation of anatomical charts of the pelvic floor muscles, and instructed to perform the same exercises at home at least three times a day, and as often as possible. Study outcome was measured by performing a new standardized pad-weighing test 1 and 3 months after the
Fig. 2. Vaginal EMG and rectal pressure in a patient during biofeed- back exercise.
treatment and comparing the values to baseline values. The patient was considered cured when the pad test showed a result of 0 or 1 g. After a median 2 89 years a questionnaire was sent to each patient concerning the current degree of incontin- ence and the amount of pelvic floor exercises being presently performed.

Statistical Analysis :
Descriptive statistics include the median, quantiles and confidence limits of the median. The results of the pad- weighing tests were analysed as an ANOVA with repeated measurements by means of PROC MIXED of SAS [9]. In order to approximate data to a normal distribution a logarithmic transformation with a cor- rection factor of 0.25 (y = ln) (pad-weighing test in g + 0.25 g) was used. Absolute changes in logarithmic transformed data correspond to relative changes in untransformed data. To compare outcome in relation to the ICS classification we used the Mann-Whitney rank sum test, and to compare outcome in terms of cured/not cured we applied Fisher’s exact test. The study was accepted by the local ethical com- mittee.
Fig. 1. A vaginal surface electrode (Dantec 2tL20).

Results :
Forty women entered the study. Six women failed to complete the treatment period, 1 in the biofeedback plus physiotherapy group (biofeedback group) and 5

K. Glavind, S. B. N0hr and S. Walter Departments of Gynecology and Obstetrics and Urology, Aalborg Sygehus, Aalborg, Denmark