- Email: [email protected]
Pelvic floor muscle exercises: 5 Years later
ADULT UROLOGY
PELVIC FLOOR MUSCLE EXERCISES" 5 YEARS LATER HENDRIK CAMMU, M.D. MICHELLE VAN NYLEN, M.S.P.T.
From the Urogynecological Unit and Department of Physiotherapy, Academisch Ziekenhuis - Vrije Universiteit Brussel, Brussels, Belgium
ABSTRACT--Objectives. To determine the outcome of pelvic floor muscle exercises for genuine stress incontinence after 5 years. Methods. Questionnaires were sent to 48 women, mean age 57 years, with troublesome stress incontinence treated as outpatients by a skilled female physiotherapist to elucidate a self-assessment of therapy outcome and to determine patients' compliance concerning fulfillment of home exercises and attitude toward physiotherapy. Patients' self-assessment responses indicated cured, much improved, some improvement, or unchanged/worse and incidence of anti-incontinence surgery after physiotherapy. Results. The overall cure/much improvement rate for physiotherapy at the end of therapy was 54% and 5 years later it was 58% (confidence interval, 43 to 72); (P = 1.000, binomial test). Thirteen women (27%) underwent surgery. Seven unoperated women (I 5%) showed only some improvement or relapse and may have been undertreated. Severity of symptoms before therapy was an important factor in therapy outcome but not in therapy maintenance. Frequency of home practicing was comparable in those who had surgery afterward and those who had not. There was no clear linear relationship in long-term effect and frequency of home practicing. Severity of symptoms and behavioral changes bias this relationship. Physiotherapy was well tolerated, as 73% would still prefer it as first choice. Pelvic floor muscle exercises were recommended to friends or relatives by 77% of the patients. Conclusions. Once a certain level of incontinence is established with pelvic floor muscle exercises, that level is maintained over 5 years.
Urinary incontinence is defined as "the involuntary loss of urine which is objectively demonstrable and is a social or hygienic problem. ''1 Stress urinary incontinence denotes the loss of urine from the urethra synchronous with physical exertion and in the absence of a detrusor contraction. 1 Urinary i n c o n t i n e n c e affects 14% of women, of w h o m one third report that incontinence has an important limiting effect on daily social activities. 2 Treatment consists primarily in surgical correction of urethral sphincter incompetence with excellent results. Surgery, however, has its risks and is not indicated for all. In 1948, Kegel3 was the Submitted: April 28, 1994, accepted (with revisions): August 29, 1994
UROLOGY®/JANUARY1995 / VOLUME45, NUMBER1
first to describe the association of inadequate pelvic tone and strength with stress incontinence. By means of an intravaginal pneumatic device, a perineometer, he tried to restore pelvic floor muscular strength. During the past decade, there has been renewed interest in conservative management: more than two dozen articles concerning the effect of pelvic muscle exercises on incontinence have been published. 4 Although the cure rate thereafter is considerably lower (20% to 50%) than after surgery, the improvement rate is high, thereby reducing the need for surgery. 5 Physiotherapy is also effective in improving patients' well-being. 5 Therefore, pelvic muscle exercises do have a place in the treatment armamentarium~4 However, one of the many questions that remains unanswered concerns the
| 1S
long-term effects of pelvic floor muscle exercises. 6 This article aims at contributing an answer to this latter quegtion. MATERIAL AND METHODS From January 1, 1986, to June 30, 1987, we conducted a cohort study concerning the impact of physiotherapy on the treatment of genuine stress incontinence, s All 52 participating women had troublesome and urodynamically proven stress incontinence. Urodynamics consisted of transurethral continuous filling subtracted cystometry in the sitting position. Provocation tests for elucidating detrusor instability and sphincteric incompetence were performed. During these tests, direct visualization of urinary leakage in the absence of detrusor contraction was diagnostic of stress incontinence. Further inclusion criteria consisted of: having no type lII incontinence, 7 being mentally and physically fit, not being institutionalized, or being in the postpartum period. The patients had neither a concomitant urogynecologic disease warranting surgery, nor a contraindication for surgery. PF.twc P~YS~OTttERA1"Y
Pelvic physiotherapy consisted of 30-minute, twice weekly sessions under the direction of a skilled female physiotherapist in the outpatient clinic for 10 weeks. The patients were first instructed about pelvic anatomy. Pelvic muscle reeducation was carried out by means of vaginal palpation that served to enhance levator ani muscle awareness, muscle contraction strengthening, and to control correct performance of the exercises. The patient was instructed to retract the anus and vagina without using abdominal or gluteal muscles. Counterproductive Valsalva-type maneuvers were monitored by placing the therapist's hand on the lower abdomen and the patient's hands on the buttocks. Fast and slow twitch fibers were trained by alternating fast and firm with long-sustained and slow contractions. One series of contractions consisted of 10 brief, forceful contractions followed by 10 slow maximal contractions held for 5 to a maximum of 10 seconds. The patient was asked to perform, if possible, 10 series per session. Patients were also instructed to contract, to squeeze, voluntarily the pelvic floor muscles prior to sudden intra-abdominal pressure rises such as that occurring during a coughing spell. Home therapy consisted in practicing contractions as frequently as possible according to the patient's ability and using it in daily activities. The patients could at any time abandon the exercises and request surgery. 114
ASSESSMENT OF THERAI"r OUTCOME
Therapy outcome was originally assessed by means of a urinary diary recording the number of leakages and pad changes as well as by a selfmade postal questionnaire. 5 Five years after the end of therap); 52 women who had taken part in the initial investigation received another quest i o n n a i r e - w i t h a self-addressed stamped envel o p e - c o m p r i s i n g mostly the same questions. Patients could evaluate their actual condition compared with that before physiotherapy: cured (no leakage under any circumstance), much improved (occasional wetness), some improvement (reduced incontinence), and relapsed (at pretherapy level)/worse. Patients were asked to report any urogynecologic operations during the followup period. Information was obtained about the frequency of pad use and about the actual frequency of home practicing (daily/several times per week/weekl); or less/never). The questionnaire also served to measure the patients' satisfaction: "If you had to do it again, would you have physiotherapy first, and if necessary, followed by surgery?" and "Did you recommend pelvic muscle exercises to friends or relatives?" STATISTICAL ANALYSIS
Statistical tests were carried out two-sided at the 5% level of significance. The relationship between outcome of physiotherapy and the patient characteristics was investigated by means of the Spearman rank correlation coefficient for continuous variables and by means of the chi-square test for discrete variables. The change in response to therapy after 5 years was investigated using the binomial test (comparing the number of patients with worse response to the number of patients with better response). The frequency of patients recommending physiotherapy to relatives was compared for patients who underwent surgery and for patients who exclusively received physiotherapy by using the Fisher exact test.
RESULTS Of the 52 women who had taken part in our initial investigation, s 48 women (92%) responded to our 5-year follow-up questionnaire. These 48 patients, with a mean age of 57 years (range, 37 to 77), formed the study group. Patient characteristics before therapy are shown in Table I. The subjective outcome is shown in Table II. Of 26 patients who believed they had been cured or much improved at the end of therapy, 21 (81%) were still cured or m u c h improved 5 years later. UROLOGY®/JANUARY1995 / VOLUME45, NUMBER 1
Patient characteristics before physiotherapy *
TABLE I. Characteristic
Mean (SD)
Age (yr] Parity (no.) Weight index (kg/m 2) Duration of symptoms (yr) Pads used per week (no.) Menopause (%] Urgency symptom (%) Previous anti-incontinence surgery (%)
52.4 2.2 24.6 4.4 19.5 26 25 7
(11] (1) (S) (4.6) (12.5) (54) (52) (15)
There was rio influence of these characteristics in the 48 paticnts on the outcome of physiotherapy except for pads used per"week, where Spearman rank correlation coefficient was 0.48, P 0.001. =
Eighteen w o m e n showed some improvement at the end of therapy and 6 of them were operated on for insufficient i m p r o v e m e n t or relapse. However, 7 women (39%) found after 5 years that they were now cured or m u c h improved. The 4 patients who showed no improvement at the end of therapy had been operated on. In summary, the overall cure/much improvement rate for physiotherapy at the end of therapy was 54% and 5 years later it was 58% (confidence interval, 43 to 72) (Table II). Comparison of therapy outcome at the end of therapy and 5 years later shows that the effect was maintained (P = 1.000 binomial test). Thirteen w o m e n (27%) u n d e r w e n t surgery: 9 within th e year following physiotherapy, 1 after 16 months and 1 after 4 years. Two w o m e n - - 1 inadequately improved and 1 r e l a p s e d - - w e r e recruited at the 5-year follow-up investigation and operated on afterward. The principal reason for surgery was incontinence in 10 cases and prolapse with incontinence in 3 cases. Seven women (15%),
TABLE II.
not operated on, found themselves 5 years later at the level of some improvement or relapse and may be considered as undertreate~ (Table II). Pad usage per week was 19.5 (SD 12.5) before treatment, 9.5 (11.2) at the end of therapy, and 5:1 (4.3) 5 years later. If those 35 w o m e n who only received physiotherapy are taken into consideration, then pad usage was 17.8 (13.1) before therapy, 5.8 (3.9) at therapy end, and 4.2 (3.9) 5 years later. Severity of symptoms--as expressed by means of the number of pads used daily--was an important factor in therapy outcomes (Table I) but not in therapy maintenance: patients who, before therapy used 2 pads per day or less were cured/much improved in 67% of the cases at therapy end and in 70% 5 years later. However, when initially more than 2 pads per day were used, the cure/much improvement rate was only 42% and it was still 42% 5 years later. Physiotherapy contributed to a reduction in pad usage: of the 35 unoperated patients, 2 (6%) did not use pads before therapy; this number increased to 16 (46%) at therapy end and increased further to 23 (66%) 5 years later. Frequency of h o m e practicing is shown in Table III. Those w o m e n who finally had surgery had practiced as frequently as the others. Once operated on, the need for exercising clearly decreased. However, of the 6 unoperated w o m e n with an impaired level of continence at follow-up (Table II), only 1 practiced regularly, 4 sometimes, and 1 never. Thirty-three women agreed to be seen at the outpatient clinic. Together, we went over the answers to the questionnaire and we assessed the pelvic floor. Correct performance of a good active voluntary muscle contraction was witnessed in 19
Subjective outcome of physiotherapy Improvement
Cured
Much
Some
End of therapy ]2 14 18 Surgery* 0 1 •6 S-year follow-up of the unoperated women (n - 37) Cured 7 3 2 Much improved 3 8 5 Some improvement 2 (1 t) 0 4 Relapse 0 2 (! t) 1 Surgery* 1 2 6
Unchanged
Total
4 4
48 !1
.. .. .. .. 4
12 16 6 3 !3
*Of the 48 women, number who had surgery from therapy end to the 5-year follow-up for inadequate improvement or relapse. rWomen who had surgery after the 5-year follow-up recruitment. ~Total number of women who had surgery to date.
UROLOGY®/JANUARY1995
/ VOLUME 45, NUMBER 1
1 15
TABLE III.
Home practicing of pelvic floor exercises Total No.
Home Practicing of Patients (n = 48) Patients Only physiotherapy 35 Physiotherapy followed by surgery Before surgery 13 After surgery 13
Daily/Several Times per Week
Once a Week or Less
18 (51%)
I ] (31%)
Never 6 (17%)
7 (54%) 2 (I 5%)
4 (Sl %) 3 (23%)
2 (I 5%) 8 (62%)
patients, of whom 4 had had surgery and 9 practiced regularly or daily. Eight women had a correct but a weak contraction, of whom 4 practiced on a regular basis and 2 had had surgery. Six women were unable to perform a correct contraction, 5 had had surgery and 1 trained very occasionally. "If you had to do it again, would you have physiotherapy first a n d - - i f necessary--followed by surgery?" Overall, 73% agreed with this statement: 31 (89%) of those not operated on agreed, 2 had no opinion, and 2 would never undergo surgery despite the level of incontinence. Of those operated on, 4 (31%) still would first prefer physiotherapy, 9 would now have preferred surgery without delay. Pelvic muscle exercises were recommended to friends or relatives by 37 women (77%): 31 (89%) who only had exercises and 6 (46%) who had surgery (P = 0.004, Fisher exact test). A historical control group comprising 18 women (mean age, 54 years, and 18.4 mean pad use/week) who wanted surgery, but where the intervention was postponed for a mean of 8.2 months, was designed. Of these, 16 were finally operated on, 2 still wanted to wait because some improvement was achieved without therapy. COMMENT Substantial improvement or cure of urinary incontinence can be achieved with pelvic floor exercises. 4 Our study of 48 patients suggests that this effect, once established, is maintained for 5 years. This corroborates the data reported in the literature conceming a follow-up period of more than 1 year. 8-11 Benvenuti et al. 8 studied 22 patients for 12 to 36 months (median duration, 21 months) and noticed that 82% had maintained or improved urinary control attained at the end of physiotherapy. Ferguson et al. 9 had good results in 19 women questioned 12 to 24 months poststudy. Klarskov et al. 10 examined 10 women, who only had physiotherapy, after 4 to 8 years (median, 6 years) and noted that 7 were stable or had improved when compared with 1 year after physiotherapy.
1 16
Recently, however, Hahn et al. 11 examined 152 women who had participated in a pelvic floor exercise program, 2 to 7 (mean, 4.3 years) years post-training. In their series, only 41% were still cured or improved whereas 34% had relapsed or deteriorated and 25% were operated on. They mentioned that this could be due to an unsatisfactory patients' training frequency, ll Indeed, some authors agreed that continued home training was mandatory to have a long-lasting effect. 8,9 Benvenuti et al. s found that of the 4 women with worsened condition, 2 had stopped exercising and 2 continued irregularly. Overall, 77% continued practicing. Ferguson et al. 9 noticed that, after 12 to 24 months, 9 (47%) women still exercised, and 10 discontinued home training and had symptoms similar to those before therapy. Klarskov et al. 1° reported that 59% of patients still used the physiotherapy program at least once weekly but no association of long-term home practicing with symptoms was made. Hahn et al., 11 analyzing the relationship between the frequency of training and the degree of improvement came to "confusing" findings : on one hand, they found that improved/cured women trained daily, whereas other improved/cured w o m e n trained once a week, yet some women who trained several times per day were no better or even had worsened the incontinence. The authors attributed this conflicting data to a different way of defining "training" by different patients a n d to the fact that patients probably did not include functional (voluntary contraction) training. 11 Similar to Hahn et a l . l l we found no linear relationship between frequency of home practicing and long-term results. This may be caused by the fact that degree of incontinence and behavioral changes bias this relationship) For instance, some women with mild incontinence who considered that they were cured, did not practice at all. They only used the active voluntary muscle contraction when necessary and were satisfied. Others, merely with an important degree of incontinence, tended to practice more frequently to maintain the incontinence at an acceptable level. When this could
UROLOGY ® ~JANUARY1995 / VO/_UM~45, NUMBER1
not be achieved, they requested surgery. Several w o m e n who were seen at the outpatient clinic told us that frequency of home practicing varied in time according to their needs, that they practice up to the level of their own satisfaction. This latter is mainly but not solely a result of improved continence but also related to the women's active contribution in the treatment and a regained selfconfidence. 5 However, we believe that regular h o m e training is important: the 6 w o m e n who were not operated on and had an impaired continence practiced very irregularly; furthermore, after surgery both contraction strength and home practicing diminished (Table III). There was a gradual reduction in the use of pads over the years. Questioned about this, several w o m e n stated that, although much improved or cured, it took them some time to stop using pads. After 5 years, some of the pads were used probably only as a way of precaution. It is known that pelvic floor exercises, similar to surgery, improves abdominal pressure transmission during stress, 12,13 possibly due to an active levator contraction that causes ventralization and uplift of the urethrovesical junction. 12 A major weakness of this study is the lack of urodynamic data. However, there is a good correlation between subjective complaints and objective findings, such as pressure to transmission ratio or pad testing. 13,14 Another shortcoming of the study is that our follow-up results are based on the answers of a mail-in survey and are, therefore, necessarily of a subjective nature. It is known that the subjective outcome is mostly better than the objective one. This may be due to the fact that assessment in terms of success depends on the interpretation of the interviewer at the follow-up conversation. To minimize interpretation biases both for the patient and the interviewer and to overcome recall biases, we asked some simple and straightforward and well-defined questions and we used a questionnaire that had to be sent back to the physician and not to the physiotherapist. It would have been even better if a completely neutral third party, not involved in the treatment of the patients concerned, had collated the information. However, it is not known whether a response of 92% would then have been achieved. We conclude that pelvic floor exercises have a long-lasting effect, but an active cooperation and easy contact between urogynecologist, physiotherapist, and patient is mandatory. Surgery or repeat physiotherapy sessions must be easily available to avoid dissatisfied, undertreated patients. Physiotherapy should be individualized. We do
UROLOGY®/JANUARY1995 / VOLUIvlE45, NUMBERI
not forget that an important contribution to success lies in the professionalism and dedication of the therapist, for intense training with the therapist improves the resuhs. 15 Hendrik Cammu, M.D. Department of Obstetrics, Gynecology, and Andrology A.Z.-V.U.B. Laarbeeklaan 101 1090 Brussels, Belgium ACKNOWLEDGMENT.To J. J. Amy for reviewing the manuscript and M.P. Derde for helping in interpretation of data. REFERENCES 1. Abrams P, Blaivas JG, Stanton SL, and Andersen JT: The standardization of terminology of lower urinary tract function recommended by the International Continence Society. Int Urogynecol J 1: 45-58, 1990. 2. Brocklehurst JC: Urinary incontinence in the communityanalysis of a MOR! poll. Br Med J 306: 832-834, 1993. 3. Kegel AH: Progressive resistance exercise in the functional restoration of the perineal muscles. Am J Obstet Gynecol 56: 238-248, 1948. 4. Wall LL, and Davidson TG: The role of muscular reeducation by physical therapy in the treatment of genuine stress urinary incontinence. Obstet Gynecol Surv 47: 322-331, 1992. 5. Cammu H, Van Nylen M, Derde MP, De Bruyne R, and Amy JJ: Pelvic physiotherapy in genuine stress incontinence. Urology 38: 332-337,1991. 6. Pierson CA: Strengthening the pelvic floor with Kegel exercises. Int UrogynecolJ 1: 59-61,1990. 7. McGuire EJ, Lytton B, Pepe V, and Kohorn EI: Stress urinary incontinence. Obstet Gynecol 47: 255-264, 1976. 8. Benvenuti F, Caputo GM, Bandinelli S, Mayer F, Biagini G, and Sommavilla A: Reeducative treatment of female genuine stress incontinence. Am J Phys Med 66: 155-168, 1987. 9. Ferguson KL, McKey PL, Bishop KR, Kloen P, Verheul JB, and Dougherty MC: Stress urinary incontinence: effect of pelvic muscle exercise. Obstet Gynecol 75: 671-675, 1990. 10. Klarskov P, Nielsen KK, KTomann-Andersen B, and Maegaard E: Long-term results of pelvic floor training and surgery for female genuine stress incontinence. Int Urogynecol J 2: 132-135, 1991. 11. Hahn I, Milsom I, Fall M, and Ekelund P: Long-term results of pelvic floor training in female stress urinary incontinence. BrJ Urol 72: 421-427, 1993. 12. Scht~ssler B, Von Obernitz N, Frimberger J, ROmisch M, and Hesse U: Analysis of successful treatment of SUI by pelvic floor reeducation. A urodynamic and radiological study. Neurourol Urodynam 9: 433-434, 1990. 13. Elia G, and Bergman A: Pelvic muscle exercises: when do they work? Obstet Gynecol 81: 283-286, 1993. 14. Mouritsen L, Frimodt-Moller C, and Moller M: Longterm effect of pelvic floor exercises on female urinary incontinence. BrJ Urol 68: 32-37, 1991. 15. Bo K, Hagen RH, Kvarstein B, Jorgensen J, and Larsen S: Pelvic floor muscle exercise for the treatment of female stress incontinence: III. Effects Of two different degrees of pelvic floor muscle exercises. Neurourol Urodynam 9: 489-502, 1990.
1 17
PELVIC FLOOR MUSCLE EXERCISES" 5 YEARS LATER HENDRIK CAMMU, M.D. MICHELLE VAN NYLEN, M.S.P.T.
From the Urogynecological Unit and Department of Physiotherapy, Academisch Ziekenhuis - Vrije Universiteit Brussel, Brussels, Belgium
ABSTRACT--Objectives. To determine the outcome of pelvic floor muscle exercises for genuine stress incontinence after 5 years. Methods. Questionnaires were sent to 48 women, mean age 57 years, with troublesome stress incontinence treated as outpatients by a skilled female physiotherapist to elucidate a self-assessment of therapy outcome and to determine patients' compliance concerning fulfillment of home exercises and attitude toward physiotherapy. Patients' self-assessment responses indicated cured, much improved, some improvement, or unchanged/worse and incidence of anti-incontinence surgery after physiotherapy. Results. The overall cure/much improvement rate for physiotherapy at the end of therapy was 54% and 5 years later it was 58% (confidence interval, 43 to 72); (P = 1.000, binomial test). Thirteen women (27%) underwent surgery. Seven unoperated women (I 5%) showed only some improvement or relapse and may have been undertreated. Severity of symptoms before therapy was an important factor in therapy outcome but not in therapy maintenance. Frequency of home practicing was comparable in those who had surgery afterward and those who had not. There was no clear linear relationship in long-term effect and frequency of home practicing. Severity of symptoms and behavioral changes bias this relationship. Physiotherapy was well tolerated, as 73% would still prefer it as first choice. Pelvic floor muscle exercises were recommended to friends or relatives by 77% of the patients. Conclusions. Once a certain level of incontinence is established with pelvic floor muscle exercises, that level is maintained over 5 years.
Urinary incontinence is defined as "the involuntary loss of urine which is objectively demonstrable and is a social or hygienic problem. ''1 Stress urinary incontinence denotes the loss of urine from the urethra synchronous with physical exertion and in the absence of a detrusor contraction. 1 Urinary i n c o n t i n e n c e affects 14% of women, of w h o m one third report that incontinence has an important limiting effect on daily social activities. 2 Treatment consists primarily in surgical correction of urethral sphincter incompetence with excellent results. Surgery, however, has its risks and is not indicated for all. In 1948, Kegel3 was the Submitted: April 28, 1994, accepted (with revisions): August 29, 1994
UROLOGY®/JANUARY1995 / VOLUME45, NUMBER1
first to describe the association of inadequate pelvic tone and strength with stress incontinence. By means of an intravaginal pneumatic device, a perineometer, he tried to restore pelvic floor muscular strength. During the past decade, there has been renewed interest in conservative management: more than two dozen articles concerning the effect of pelvic muscle exercises on incontinence have been published. 4 Although the cure rate thereafter is considerably lower (20% to 50%) than after surgery, the improvement rate is high, thereby reducing the need for surgery. 5 Physiotherapy is also effective in improving patients' well-being. 5 Therefore, pelvic muscle exercises do have a place in the treatment armamentarium~4 However, one of the many questions that remains unanswered concerns the
| 1S
long-term effects of pelvic floor muscle exercises. 6 This article aims at contributing an answer to this latter quegtion. MATERIAL AND METHODS From January 1, 1986, to June 30, 1987, we conducted a cohort study concerning the impact of physiotherapy on the treatment of genuine stress incontinence, s All 52 participating women had troublesome and urodynamically proven stress incontinence. Urodynamics consisted of transurethral continuous filling subtracted cystometry in the sitting position. Provocation tests for elucidating detrusor instability and sphincteric incompetence were performed. During these tests, direct visualization of urinary leakage in the absence of detrusor contraction was diagnostic of stress incontinence. Further inclusion criteria consisted of: having no type lII incontinence, 7 being mentally and physically fit, not being institutionalized, or being in the postpartum period. The patients had neither a concomitant urogynecologic disease warranting surgery, nor a contraindication for surgery. PF.twc P~YS~OTttERA1"Y
Pelvic physiotherapy consisted of 30-minute, twice weekly sessions under the direction of a skilled female physiotherapist in the outpatient clinic for 10 weeks. The patients were first instructed about pelvic anatomy. Pelvic muscle reeducation was carried out by means of vaginal palpation that served to enhance levator ani muscle awareness, muscle contraction strengthening, and to control correct performance of the exercises. The patient was instructed to retract the anus and vagina without using abdominal or gluteal muscles. Counterproductive Valsalva-type maneuvers were monitored by placing the therapist's hand on the lower abdomen and the patient's hands on the buttocks. Fast and slow twitch fibers were trained by alternating fast and firm with long-sustained and slow contractions. One series of contractions consisted of 10 brief, forceful contractions followed by 10 slow maximal contractions held for 5 to a maximum of 10 seconds. The patient was asked to perform, if possible, 10 series per session. Patients were also instructed to contract, to squeeze, voluntarily the pelvic floor muscles prior to sudden intra-abdominal pressure rises such as that occurring during a coughing spell. Home therapy consisted in practicing contractions as frequently as possible according to the patient's ability and using it in daily activities. The patients could at any time abandon the exercises and request surgery. 114
ASSESSMENT OF THERAI"r OUTCOME
Therapy outcome was originally assessed by means of a urinary diary recording the number of leakages and pad changes as well as by a selfmade postal questionnaire. 5 Five years after the end of therap); 52 women who had taken part in the initial investigation received another quest i o n n a i r e - w i t h a self-addressed stamped envel o p e - c o m p r i s i n g mostly the same questions. Patients could evaluate their actual condition compared with that before physiotherapy: cured (no leakage under any circumstance), much improved (occasional wetness), some improvement (reduced incontinence), and relapsed (at pretherapy level)/worse. Patients were asked to report any urogynecologic operations during the followup period. Information was obtained about the frequency of pad use and about the actual frequency of home practicing (daily/several times per week/weekl); or less/never). The questionnaire also served to measure the patients' satisfaction: "If you had to do it again, would you have physiotherapy first, and if necessary, followed by surgery?" and "Did you recommend pelvic muscle exercises to friends or relatives?" STATISTICAL ANALYSIS
Statistical tests were carried out two-sided at the 5% level of significance. The relationship between outcome of physiotherapy and the patient characteristics was investigated by means of the Spearman rank correlation coefficient for continuous variables and by means of the chi-square test for discrete variables. The change in response to therapy after 5 years was investigated using the binomial test (comparing the number of patients with worse response to the number of patients with better response). The frequency of patients recommending physiotherapy to relatives was compared for patients who underwent surgery and for patients who exclusively received physiotherapy by using the Fisher exact test.
RESULTS Of the 52 women who had taken part in our initial investigation, s 48 women (92%) responded to our 5-year follow-up questionnaire. These 48 patients, with a mean age of 57 years (range, 37 to 77), formed the study group. Patient characteristics before therapy are shown in Table I. The subjective outcome is shown in Table II. Of 26 patients who believed they had been cured or much improved at the end of therapy, 21 (81%) were still cured or m u c h improved 5 years later. UROLOGY®/JANUARY1995 / VOLUME45, NUMBER 1
Patient characteristics before physiotherapy *
TABLE I. Characteristic
Mean (SD)
Age (yr] Parity (no.) Weight index (kg/m 2) Duration of symptoms (yr) Pads used per week (no.) Menopause (%] Urgency symptom (%) Previous anti-incontinence surgery (%)
52.4 2.2 24.6 4.4 19.5 26 25 7
(11] (1) (S) (4.6) (12.5) (54) (52) (15)
There was rio influence of these characteristics in the 48 paticnts on the outcome of physiotherapy except for pads used per"week, where Spearman rank correlation coefficient was 0.48, P 0.001. =
Eighteen w o m e n showed some improvement at the end of therapy and 6 of them were operated on for insufficient i m p r o v e m e n t or relapse. However, 7 women (39%) found after 5 years that they were now cured or m u c h improved. The 4 patients who showed no improvement at the end of therapy had been operated on. In summary, the overall cure/much improvement rate for physiotherapy at the end of therapy was 54% and 5 years later it was 58% (confidence interval, 43 to 72) (Table II). Comparison of therapy outcome at the end of therapy and 5 years later shows that the effect was maintained (P = 1.000 binomial test). Thirteen w o m e n (27%) u n d e r w e n t surgery: 9 within th e year following physiotherapy, 1 after 16 months and 1 after 4 years. Two w o m e n - - 1 inadequately improved and 1 r e l a p s e d - - w e r e recruited at the 5-year follow-up investigation and operated on afterward. The principal reason for surgery was incontinence in 10 cases and prolapse with incontinence in 3 cases. Seven women (15%),
TABLE II.
not operated on, found themselves 5 years later at the level of some improvement or relapse and may be considered as undertreate~ (Table II). Pad usage per week was 19.5 (SD 12.5) before treatment, 9.5 (11.2) at the end of therapy, and 5:1 (4.3) 5 years later. If those 35 w o m e n who only received physiotherapy are taken into consideration, then pad usage was 17.8 (13.1) before therapy, 5.8 (3.9) at therapy end, and 4.2 (3.9) 5 years later. Severity of symptoms--as expressed by means of the number of pads used daily--was an important factor in therapy outcomes (Table I) but not in therapy maintenance: patients who, before therapy used 2 pads per day or less were cured/much improved in 67% of the cases at therapy end and in 70% 5 years later. However, when initially more than 2 pads per day were used, the cure/much improvement rate was only 42% and it was still 42% 5 years later. Physiotherapy contributed to a reduction in pad usage: of the 35 unoperated patients, 2 (6%) did not use pads before therapy; this number increased to 16 (46%) at therapy end and increased further to 23 (66%) 5 years later. Frequency of h o m e practicing is shown in Table III. Those w o m e n who finally had surgery had practiced as frequently as the others. Once operated on, the need for exercising clearly decreased. However, of the 6 unoperated w o m e n with an impaired level of continence at follow-up (Table II), only 1 practiced regularly, 4 sometimes, and 1 never. Thirty-three women agreed to be seen at the outpatient clinic. Together, we went over the answers to the questionnaire and we assessed the pelvic floor. Correct performance of a good active voluntary muscle contraction was witnessed in 19
Subjective outcome of physiotherapy Improvement
Cured
Much
Some
End of therapy ]2 14 18 Surgery* 0 1 •6 S-year follow-up of the unoperated women (n - 37) Cured 7 3 2 Much improved 3 8 5 Some improvement 2 (1 t) 0 4 Relapse 0 2 (! t) 1 Surgery* 1 2 6
Unchanged
Total
4 4
48 !1
.. .. .. .. 4
12 16 6 3 !3
*Of the 48 women, number who had surgery from therapy end to the 5-year follow-up for inadequate improvement or relapse. rWomen who had surgery after the 5-year follow-up recruitment. ~Total number of women who had surgery to date.
UROLOGY®/JANUARY1995
/ VOLUME 45, NUMBER 1
1 15
TABLE III.
Home practicing of pelvic floor exercises Total No.
Home Practicing of Patients (n = 48) Patients Only physiotherapy 35 Physiotherapy followed by surgery Before surgery 13 After surgery 13
Daily/Several Times per Week
Once a Week or Less
18 (51%)
I ] (31%)
Never 6 (17%)
7 (54%) 2 (I 5%)
4 (Sl %) 3 (23%)
2 (I 5%) 8 (62%)
patients, of whom 4 had had surgery and 9 practiced regularly or daily. Eight women had a correct but a weak contraction, of whom 4 practiced on a regular basis and 2 had had surgery. Six women were unable to perform a correct contraction, 5 had had surgery and 1 trained very occasionally. "If you had to do it again, would you have physiotherapy first a n d - - i f necessary--followed by surgery?" Overall, 73% agreed with this statement: 31 (89%) of those not operated on agreed, 2 had no opinion, and 2 would never undergo surgery despite the level of incontinence. Of those operated on, 4 (31%) still would first prefer physiotherapy, 9 would now have preferred surgery without delay. Pelvic muscle exercises were recommended to friends or relatives by 37 women (77%): 31 (89%) who only had exercises and 6 (46%) who had surgery (P = 0.004, Fisher exact test). A historical control group comprising 18 women (mean age, 54 years, and 18.4 mean pad use/week) who wanted surgery, but where the intervention was postponed for a mean of 8.2 months, was designed. Of these, 16 were finally operated on, 2 still wanted to wait because some improvement was achieved without therapy. COMMENT Substantial improvement or cure of urinary incontinence can be achieved with pelvic floor exercises. 4 Our study of 48 patients suggests that this effect, once established, is maintained for 5 years. This corroborates the data reported in the literature conceming a follow-up period of more than 1 year. 8-11 Benvenuti et al. 8 studied 22 patients for 12 to 36 months (median duration, 21 months) and noticed that 82% had maintained or improved urinary control attained at the end of physiotherapy. Ferguson et al. 9 had good results in 19 women questioned 12 to 24 months poststudy. Klarskov et al. 10 examined 10 women, who only had physiotherapy, after 4 to 8 years (median, 6 years) and noted that 7 were stable or had improved when compared with 1 year after physiotherapy.
1 16
Recently, however, Hahn et al. 11 examined 152 women who had participated in a pelvic floor exercise program, 2 to 7 (mean, 4.3 years) years post-training. In their series, only 41% were still cured or improved whereas 34% had relapsed or deteriorated and 25% were operated on. They mentioned that this could be due to an unsatisfactory patients' training frequency, ll Indeed, some authors agreed that continued home training was mandatory to have a long-lasting effect. 8,9 Benvenuti et al. s found that of the 4 women with worsened condition, 2 had stopped exercising and 2 continued irregularly. Overall, 77% continued practicing. Ferguson et al. 9 noticed that, after 12 to 24 months, 9 (47%) women still exercised, and 10 discontinued home training and had symptoms similar to those before therapy. Klarskov et al. 1° reported that 59% of patients still used the physiotherapy program at least once weekly but no association of long-term home practicing with symptoms was made. Hahn et al., 11 analyzing the relationship between the frequency of training and the degree of improvement came to "confusing" findings : on one hand, they found that improved/cured women trained daily, whereas other improved/cured w o m e n trained once a week, yet some women who trained several times per day were no better or even had worsened the incontinence. The authors attributed this conflicting data to a different way of defining "training" by different patients a n d to the fact that patients probably did not include functional (voluntary contraction) training. 11 Similar to Hahn et a l . l l we found no linear relationship between frequency of home practicing and long-term results. This may be caused by the fact that degree of incontinence and behavioral changes bias this relationship) For instance, some women with mild incontinence who considered that they were cured, did not practice at all. They only used the active voluntary muscle contraction when necessary and were satisfied. Others, merely with an important degree of incontinence, tended to practice more frequently to maintain the incontinence at an acceptable level. When this could
UROLOGY ® ~JANUARY1995 / VO/_UM~45, NUMBER1
not be achieved, they requested surgery. Several w o m e n who were seen at the outpatient clinic told us that frequency of home practicing varied in time according to their needs, that they practice up to the level of their own satisfaction. This latter is mainly but not solely a result of improved continence but also related to the women's active contribution in the treatment and a regained selfconfidence. 5 However, we believe that regular h o m e training is important: the 6 w o m e n who were not operated on and had an impaired continence practiced very irregularly; furthermore, after surgery both contraction strength and home practicing diminished (Table III). There was a gradual reduction in the use of pads over the years. Questioned about this, several w o m e n stated that, although much improved or cured, it took them some time to stop using pads. After 5 years, some of the pads were used probably only as a way of precaution. It is known that pelvic floor exercises, similar to surgery, improves abdominal pressure transmission during stress, 12,13 possibly due to an active levator contraction that causes ventralization and uplift of the urethrovesical junction. 12 A major weakness of this study is the lack of urodynamic data. However, there is a good correlation between subjective complaints and objective findings, such as pressure to transmission ratio or pad testing. 13,14 Another shortcoming of the study is that our follow-up results are based on the answers of a mail-in survey and are, therefore, necessarily of a subjective nature. It is known that the subjective outcome is mostly better than the objective one. This may be due to the fact that assessment in terms of success depends on the interpretation of the interviewer at the follow-up conversation. To minimize interpretation biases both for the patient and the interviewer and to overcome recall biases, we asked some simple and straightforward and well-defined questions and we used a questionnaire that had to be sent back to the physician and not to the physiotherapist. It would have been even better if a completely neutral third party, not involved in the treatment of the patients concerned, had collated the information. However, it is not known whether a response of 92% would then have been achieved. We conclude that pelvic floor exercises have a long-lasting effect, but an active cooperation and easy contact between urogynecologist, physiotherapist, and patient is mandatory. Surgery or repeat physiotherapy sessions must be easily available to avoid dissatisfied, undertreated patients. Physiotherapy should be individualized. We do
UROLOGY®/JANUARY1995 / VOLUIvlE45, NUMBERI
not forget that an important contribution to success lies in the professionalism and dedication of the therapist, for intense training with the therapist improves the resuhs. 15 Hendrik Cammu, M.D. Department of Obstetrics, Gynecology, and Andrology A.Z.-V.U.B. Laarbeeklaan 101 1090 Brussels, Belgium ACKNOWLEDGMENT.To J. J. Amy for reviewing the manuscript and M.P. Derde for helping in interpretation of data. REFERENCES 1. Abrams P, Blaivas JG, Stanton SL, and Andersen JT: The standardization of terminology of lower urinary tract function recommended by the International Continence Society. Int Urogynecol J 1: 45-58, 1990. 2. Brocklehurst JC: Urinary incontinence in the communityanalysis of a MOR! poll. Br Med J 306: 832-834, 1993. 3. Kegel AH: Progressive resistance exercise in the functional restoration of the perineal muscles. Am J Obstet Gynecol 56: 238-248, 1948. 4. Wall LL, and Davidson TG: The role of muscular reeducation by physical therapy in the treatment of genuine stress urinary incontinence. Obstet Gynecol Surv 47: 322-331, 1992. 5. Cammu H, Van Nylen M, Derde MP, De Bruyne R, and Amy JJ: Pelvic physiotherapy in genuine stress incontinence. Urology 38: 332-337,1991. 6. Pierson CA: Strengthening the pelvic floor with Kegel exercises. Int UrogynecolJ 1: 59-61,1990. 7. McGuire EJ, Lytton B, Pepe V, and Kohorn EI: Stress urinary incontinence. Obstet Gynecol 47: 255-264, 1976. 8. Benvenuti F, Caputo GM, Bandinelli S, Mayer F, Biagini G, and Sommavilla A: Reeducative treatment of female genuine stress incontinence. Am J Phys Med 66: 155-168, 1987. 9. Ferguson KL, McKey PL, Bishop KR, Kloen P, Verheul JB, and Dougherty MC: Stress urinary incontinence: effect of pelvic muscle exercise. Obstet Gynecol 75: 671-675, 1990. 10. Klarskov P, Nielsen KK, KTomann-Andersen B, and Maegaard E: Long-term results of pelvic floor training and surgery for female genuine stress incontinence. Int Urogynecol J 2: 132-135, 1991. 11. Hahn I, Milsom I, Fall M, and Ekelund P: Long-term results of pelvic floor training in female stress urinary incontinence. BrJ Urol 72: 421-427, 1993. 12. Scht~ssler B, Von Obernitz N, Frimberger J, ROmisch M, and Hesse U: Analysis of successful treatment of SUI by pelvic floor reeducation. A urodynamic and radiological study. Neurourol Urodynam 9: 433-434, 1990. 13. Elia G, and Bergman A: Pelvic muscle exercises: when do they work? Obstet Gynecol 81: 283-286, 1993. 14. Mouritsen L, Frimodt-Moller C, and Moller M: Longterm effect of pelvic floor exercises on female urinary incontinence. BrJ Urol 68: 32-37, 1991. 15. Bo K, Hagen RH, Kvarstein B, Jorgensen J, and Larsen S: Pelvic floor muscle exercise for the treatment of female stress incontinence: III. Effects Of two different degrees of pelvic floor muscle exercises. Neurourol Urodynam 9: 489-502, 1990.
1 17