The Role of Home Trainers in Kegel's Exercise Program for the Treatment of Incontinence

by John D. Perry, PhD and Leslie Talcott Hullett, MS, RN

from Ostomy/Wound Management, The Journal for Extended Patient Care Management, Volume 30, September/October 1990

Reprinted by permission from Ostomy/Wound Management. Copyright 1990 by Health Management Publications, Inc.

In the past decade, many studies have demonstrated the effectiveness of biofeedback-assisted Kegel exercises in the treatment or even cure of fecal incontinence.[1, 2, 3] The earliest of these studies were based on the use of anorectal manometry, a mechanical pressure transducer system, to treat fecal incontinence. Recently, Burgio, et al. [4] have shown that the same techniques can be applied to the treatment of urinary incontinence with an average of 85% reduction in frequency of accidents (40% actually cured). In another study, biofeedback from a polygraph chart was found to be superior to mere verbal feedback from a therapist (76% improvement vs. only 51%). All of these studies were conducted at the National Institute on Aging laboratories in Baltimore, using an ano-rectal manometer, a complex polygraph laboratory instrument, and all of them reference the original Kegel exercise program as their primary model and inspiration.

These NIA-sponsored reports and their projects differ in several important aspects from the original research that they emulate. First and most significantly, they obtain considerably poorer results than those reported by Arnold Kegel. In an important 1950 chapter,6 Kegel claimed "complete relief of urinary stress incontinence" in 93% of some 300 patients treated with his perineometer device. He also observed that "preliminary returns of a statistical inquiry among private physicians using the perineometer" yielded a comparable figure of 91%.

The NIA-sponsored research also reports percentages in the 80s and 90s, but it is necessary to note that the data always represent an "average reduction in symptoms (accidents)" across all subjects; the percentage of patients actually "cured," or obtaining Kegel's "complete relief," is only 29% (weighted across all studies), significantly less than the research they are duplicating. Some critics have suggested in private that Kegel may have exaggerated his figures; however, we have found no evidence concerning that charge. A closer examination of the other differences offers a more plausible and less sinister explanation. The clue to the differences in results may be found in differences in philosophy and implementation. The NIA research differs from Kegel's in terms of l) the type of instrumentation, 2) the basic orientation (home vs. office), and 3) the patient work requirements.

Kegel's Perineometer in use. Thrice daily home practice sessions with this biofeedback device were the basis for Kegel's exercise program. © 1952 Ciba-Geigy Corporation. Reproduced with permission from Clinical Symposia by Frank H. Netter, MD. All rights Reserved.

The primary instrumentation involved in Kegel's exercise program was a home training device, the Kegel Perineometer, which each patient was required to obtain and use at home in three daily practice sessions. The device consisted of a small rubber cone that was inserted into the vaginal entrance, a length of rubber tubing, and a 2-inch diameter manometer (air pressure gauge) calibrated in millimeters of Mercury. Contractions of the pelvic muscles around the vagina depressed the walls of the cone and this activity was conveyed to and displayed on the hand-held gauge (see Figure 1). Because the cone was sensitive to insertion pressure as well, it was typically held in place by a band of gauze so as to avoid movement artifact.

The patient was introduced to the perineometer in the physician's office, and her first practice sessions were conducted in his presence so that he could ensure that the patient would be practicing correctly at home. The patient returned for weekly verification sessions during the course of therapy, but the therapy consisted of the at-home biofeedback sessions. Kegel even created and furnished an apparently effective self-charting progress report form that antedates techniques used in Behavior Modification programs by over 20 years. We shall refer to the office monitoring of home practice as the "Kegel Method." (See Figure 2)

Figure 2: The Home Practice Chart distributed with Kegel's 1948 Perineometer. Patients were expected to practice three times a day, starting at five minutes per session and increasing to 20 minutes per session during the fourth week. Adapted from a figure by Arnold Kegel in Am J Obst & Gynec 1948; 16(2) 244.

In 1985, the National Institute on Aging produced a documentary videotape that summarized the biofeedback and behavioral techniques that had been developed in their physiology laboratory.[7] The film was intended to provide instruction for non-physicians in the NIA's treatment method. Like the Kegel Method, this "NIA Method" involves the use of biofeedback, but only in the laboratory, not at home. The limitation apparently stems from the use of manometric instrumentation; the sensor is a series of balloons mounted on a shaft, which must be carefully inserted several inches into the rectum under direction of a physician. The balloons are connected by several pieces of tubing to a bank of mechanical to electrical transducers, which are in turn connected to a laboratory polygraph or multi-channel strip chart recorder. The patient, who must remain in the lithotomy position, watches the strip chart output, sometimes with the aid of an opaque projector. The apparatus costs about $25,000 and requires a medical laboratory environment.

The rectal insertion is quite uncomfortable for the patient. In general, most patients endure the treatment only three or four sessions. In addition, when the patient is being treated for urge incontinence, the bladder is infused with sterile water through a catheter for direct feedback of bladder contractions; and 2 or 3% of these infusions result in bladder infections. In three or four laboratory biofeedback sessions, the patient must learn as much as possible about his or her internal muscles. Patients are, of course, encouraged to practice the same exercises on a daily basis at home, but they are not given any biofeedback device.

Kegel's patients practiced between 90 and 360 sessions with the biofeedback device (for one to four months participation, respectively), whereas the NIA patients had only 1 to 5 biofeedback sessions all together. This represents the third major difference between these two approaches: the amount of exercise prescribed for the patient. Kegel recommended 300 instrumented contractions per day for therapy. These were to be supplemented by 150 non-instrumented contractions per day after the therapy was well established and were to continue throughout life. In contrast, the NIA protocol uses only one-sixth as much practice, 51 per day (in three groups of 17) in the urinary incontinence research; all of them are done without instrumentation.

The rationale for the drastic reduction in exercise requirements has never been described in the literature but may be related to the nature of participation. All of Kegel's subjects were patients who were paying for their therapy (through insurance or otherwise), whereas all of the NIA subjects were receiving "free experimental" therapy in a government-sponsored research program. This may have affected the level of expected commitment in the later group. The greater amount of exercise prescribed may account for the better results obtained by Kegel. Only one study has compared exercise levels; Rudinger showed that a group doing 300 contractions per day did considerably better than a control group doing only 100 contractions per day. [8]

One element both programs have in common is their use of air pressure operated sensors to infer muscle activity on the basis of physical displacement of the vaginal or rectal space. Beginning in the 1970s, however, researchers have been making direct muscle assessments through newly developed electromyographic (EMG) amplifiers and feedback systems. These have the advantage of being considerably more sensitive than manometric devices; they are capable of detecting muscle activity at trace level (i.e., below the level that actually results in a muscle contraction). At first, external "patch" electrodes were used; but in 1975, an intra-vaginal (or rectal) electrode assembly, the EMG Perineometer, was introduced, allowing more accurate detection of pelvic floor muscles. In 1982, a portable home trainer, the Personal Perineometer™ (see Figure 3) was designed for use with it.

Figure 3 A contemporary EMG home trainer.

The home trainer was followed in 1983 by an office instrument, the Clinical Perineometer™, which performs some of the same functions as the polygraph, but in a transportable table-top instrument. The new system combines the office instrumentation of the NIA method with the portable home trainers used by Arnold Kegel. The first project to use this new system was sponsored by the New Jersey Department of Health; nurse practitioners obtained a 78% symptom reduction rate among 54 seniors at a walk-in health facility.[9] Most of the participants used the Personal Perineometer home trainer but usually only for the first one or two weeks. Although only 37% were "cured" at the end of treatment, this number actually increased to 59% cured after an additional year of unsupervised practice. In a related project, using the same Clinical Perineometer but without the home trainers, a group of 25 home-bound seniors obtained an overall 70% reduction in accidents.[10]

The same combination of office instrumentation and partial use of home trainers was employed in a study as part of an on-going Continence Clinic at Presbyterian Hospital in Philadelphia. LaRiccia and Chapman [11] obtained an overall 87% symptom improvement rate using the same EMG instruments plus a newly-developed, computerized biofeedback system, The Computerized Perineometer TM, which gave vivid graphic displays of patient muscle activity on a video monitor. (Note 2 )

(Note 2) The Computerized Perineometer , Model CP-300 from PerryMeter Systems, Division of Biotechnologies, Inc., Strafford, PA. 242 Old Eagle School Road, Strafford, PA 19087,1-800-537-3779.

The same computer system and EMG sensors were employed in a federally funded project at SUNY Buffalo, but without using any home trainers. Burns [12] obtained a 75% symptom reduction rate for patients trained in this "NIA Method;" a control group given extensive verbal instruction in Kegel exercises achieved only a 50% reduction in symptoms. Even more importantly, at six months follow-up, the verbal-instruction-only group had regressed to be statistically indistinguishable from the control group, whereas the (office) instrument-trained group remained significantly better.[13] (At the 1988 NIH Consensus Development Conference, Burns mentioned a 20% cure rate, which many commentators erroneously compared with the 75 to 85% symptom reduction rate of the other studies.)[14]

Similar results were obtained by Wells [15] and colleagues at the University of Michigan. In a large study involving nearly 400 women with stress and urge incontinence, an overall symptom-reduction rate of 77% was obtained, without using any home trainers. Wells explained in a speech about the project that they rejected using home trainers because they feared "sexual arousal" among their patients [l6] (Kegel had acknowledged this concern, but he stated that it "had no basis in fact in normal women.''[l7] ) It appears, therefore, that incontinence symptom reductions of between 70 and 90% are obtainable by biofeedback methods and that the higher figures seem to be associated with at least limited or occasional use of home training devices.

Only two small studies have directly compared "home-trainer" groups and control groups without devices. Waletzky and Weiss [l8] found that a group of 12 sexual dysfunction patients who used EMG home trainers had a 76% increase in pelvic muscle strength, whereas a control group without home trainers showed only a 24% increase. Taylor and Henderson,l9 in a pilot project involving matched groups of four women each compared to the use of the vaginal sensor and EMG biofeedback system with the same vaginal sensor used only as a resistive (or isometric exercise) device. They found that the group with home trainers achieved 100% control over their incontinence, whereas the resistive device and a no-device control group each obtained a 67% reduction in symptoms.

Dougherty, et al.[20] found that 16 patients using a resistive device (a soft rubber cylinder for isometric exercise) that provided no biofeedback actually did worse than 16 patients who practiced Kegel exercises on an empty vagina; the resistive device group was not significantly different from the control group, which did no exercises at all. These results are unexpected, since Arnold Kegel had also described some of the value of his Kegel Perineometer as a resistive device. Although Kegel never used the term biofeedback (which first appeared in the late 1960s), it may be that the biofeedback provided by his perineometer was more significant than its isometric qualities.

Only one clinical study [2l] has used both office instrumentation and home trainers with all patients. This study reported the highest improvement of any to date: 100% improvement in symptom reduction and elimination of urinary incontinence in a population of 31 patients. Six months later, the same project reported a 99.95% improvement rate (and 98% cured) for a new total of 56 patients.[22] The National Institute on Aging recently gave this project staff a Small Business Innovation Research Grant, [23] which resulted in the development of a microprocessor-controlled "Kegel Counter™." Like the cardiac Holter™ Monitor, the device records in its memory a summary of the patient's at-home practice activity as well as providing immediate visual biofeedback to the patient.

Table 1 summarizes the outcomes of projects using office instruments only (the NIA Method), home trainers only (the Kegel Method), and both types. Projects that utilize home trainers (as Kegel first proposed and practiced) are shown to have higher success rates and cure rates than projects that rely only on a few office training sessions.

It seems most likely that the better outcome advantage of the regular use of home trainers lies in the fact that such patients are forced into compliance with their homework instructions by the presence or absence of the physical device. The most recent study of pelvic floor exercise compliance, without the use of home biofeedback trainers, found that their patients completed only 15% of their assigned exercise.[24] Like other forms of personal discipline, it is easy to overlook doing exercises when their performance cannot even be detected by an outside observer. But when the exercise program requires unpacking and setting up a device, inserting a sensor, exercising, removing the sensor and washing it in the sink, it is decisively either done or neglected.

The widespread use of the Personal Perineometer EMG home training device was greatly enhanced in April 1990, when the Health Care Financing Administration issued a new code number (E0746) for Electromyography (EMG) biofeedback device. Subsequently, the Pennsylvania Medicare administrator published reimbursement schedules which include $670.00 for purchase of a new instrument and $75.00 per month for rental.[25] The code is applicable to EMG instruments used for muscle rehabilitation, but not for general relaxation training. Many private insurance companies had already covered the cost of the device, although some pay only for purchase, while others only pay for rentals. Even before Medicare began covering these devices, many patients elected to pay out-of-pocket, because the full price was still less than one year's cost of their incontinence care products, which are still non-reimbursable. (Note 3)

(Note 3) The Personal Perineometer lists for $995.00 (patient retail) and is sold to therapists and DMEs at $699.00 wholesale. It is available for direct patient rental at $150.00 per month from PerryMeter Systems, 242 Old Eagle School Road, Strafford, PA 19087,1-800 XXX-XXXX.

References

1. Engel BT, Nikoomanest P, Schuster MM: Operant conditioning of rectosphincteric responses in the treatment of fecal Incontinence. N Engl J Med 1974; 209:646-9. 2. Wald A: Biofeedback therapy for fecal incontinence. Annals of Internal Medicine 1981; 95:146-149.

3. Whitehead WE, Burgio KL, Engel BT: Biofeedback treatment of fecal incontinence in geriatric patients. Journal of American Geriatric Society 1985;33:320-324.

4. Burgio KL, Whitehead WE, Engel BT: Urinary incontinence in the elderly: bladder-sphincter biofeedback and toileting skills training. Annals of Internal Medicine 1985;104:507-515.

5. Burgio KL, Robinson JC, Engel BT: The role of biofeedback in Kegel exercise training for stress urinary incontinence. Am J Obstet Gynecol 1986;154:58-64.

6. Kegel A: Active exercise of the Pubococcygeus Muscle, in Progress in Gynecology. New York, Grune & Stratton, 1950), 778-792.

7. Engel BT, Whitehead WE, Burgio KL: Behavioral treatment of urinary incontinence: biofeedback of the bladder and sphincter muscles videotape. National Institutes of Health, Bethesda, MD, 1985.

8. Rudinger EA: Evaluation of a new electronic pelvic exercise and therapy. 26th Annual Meeting, Armed Forces District, American College of Obst. & Gynec, New Orleans, 1977.

9. Baigis-Smith J, Smith D, Rose M, Kaschak D: Managing urinary incontinence in community-residing elderly persons. The Gerontologist 1989;29(2):229-233.

10. Smith D, Smith J, Rose M: Control of urinary incontinence in the acutely ill home patient. Proceedings of the International Continence Society in Neurology and Urodynamics 7:4, 1987.

11. LaRiccia P, Chapman E: Report of the Incontinence Treatment Center. Seminar at Presbyterian Hospital, Philadelphia, January 14,1988.

12. Burns P, Pranikoff K, Reis JS, and Levy KJ: Effectiveness of biofeedback therapy of stress incontinent females. Proceedings of the International Continence Society in Neurology and Urodynamics, 7:4, p. 280, 1988.

13. Burns P, Pranikoff K, Nochajski TJ, Levy KJ: Effectiveness of Biofeedback Therapy for Stress Incontinence. Presented at the American Urological Society, Dallas, May 11, 1989.

14. Burns P: During Question and Answer period, at the NIH Consensus Development Conference on Urinary Incontinence in Adults, Bethesda, MD, October 6, 1988. This same 20% figure was also misquoted in Chalker, R & Whitemore, KE, Overcoming Bladder Disorders (New York, Harper and Row), 1990, p. 71.

15. Wells T, Brink C, Diokno A, Gillis G: Pelvic muscle exercises for stress urinary incontinence in elderly women. Paper presented at Gerontological Society of America, November 1987.

16. Wells T: Guest lecture read before the University of Pennsylvania School of Nursing, October 20,1987.

17. Kegel A: Progress in Gynecology, 1950, 786.

18. Waletsky LR, Weiss S: Early intervention in female sexual dysfunction: An Integrated Approach. Proc Banff Conf Behav Med, 1986.

19. Taylor K, Henderson J: Effects of biofeedback on simple stress urinary incontinence in older women. J of Geron Nursing 1986.

20. Dougherty MC, Abrams RM, Batich CD, Bishop KR, Gimptty P: Effect of exercise on the circumvaginal muscles (CVM). Proceedings of the International Continence Society in Neurology and Urodynamics, 6:4, 1987.

21. Perry JD, Hullett LT, Bollinger JR: Urinary incontinence treated by EMG biofeedback method. Gerontological Society of America, November 1987.

22. Perry JD, Hullett LT, Bollinger JR: EMG Biofeedback treatment of incontinence. California Biofeedback, Summer 1988; 79,18-19.

23. lR43-AG06755-01.

24. Ramsey IN, Thou MA: Randomized, double blind, placebo controlled trial of pelvic floor exercises in the treatment of genuine stress incontinence. Proceedings of the International Continence Society, Neurology and Urodynamics, 1990; 9:(4):398-399.

25. Estimated to average $931.00 per year at the NIH's Consensus Development Conference on Urinary Incontinence in Adults, Bethesda, 1988.

Table 1 Research Outcomes Based on Instrumentation Employed

NR = Not Reported; PP = Paying Patients (All others were subsidized research subjects.)