Conservative Therapy for Stress Incontinence
Tomas L. Griebling and Ingrid E. Nygaard
Table Of Contents
Tomas L. Griebling, MD
Ingrid E. Nygaard, MD
PELVIC FLOOR MUSCLE TRAINING
Surgery has long been considered the mainstay of treatment for urinary incontinence. However, nonsurgical treatment can improve symptoms for many women and should be undertaken before surgical treatment is considered. Although fewer women treated nonsurgically are completely cured of their incontinence compared with those who undergo operative therapy, the tradeoffs of decreased risk and morbidity result in a high satisfaction rate for many women. Many researchers are exploring various areas of nonsurgical treatment; in the future, clinicians probably will have more information about which modalities to use for specific patients.
Several lifestyle interventions are often recommended to prevent or treat stress incontinence. However, there is no scientific evidence to guide recommendations for any of these interventions. It is clear that obesity is a risk factor for stress incontinence1,2,3,4 and that weight loss in morbidly obese women alleviates stress incontinence.5,6 Recent evidence now suggests that weight loss also reduces urinary incontinence in moderately obese women. A cohort of 10 women with a mean body mass index (BMI) of 38.3 and mean weight of 106 kg decreased the mean number of weekly incontinence episodes from 13 to 8 per week. All six of the women who lost at least 5% of their weight decreased the number of incontinence episodes by at least 50%, compared with only one of the four women who lost less weight.7 In a larger randomized trial,8 47 women with stress incontinence and a BMI between 25 and 40 kg/m2 (mean BMI: 35.2, mean weight: 93 kg) were randomly assigned to either immediate weight loss with a very low-calorie liquid diet and exercise or delayed weight reduction after the 3-month study program. The women in the weight loss group experienced a 53% decrease in weekly incontinent episodes compared with 4% for the control women. Further work is ongoing to determine whether this effect holds over time.
Some studies have found an association between cigarette smoking and urinary incontinence,9 but it is unknown whether smoking cessation decreases urine leakage. A small number of studies have assessed the efficacy of caffeine restriction, fluid management, or dietary changes in the treatment of incontinence. In a cohort study in which 41 women worked with a nurse during home visits to reduce incontinence, women were advised to decrease caffeine and increase other fluid intake. There was a trend between the decrease in caffeine and fewer daytime incontinence episodes.10 In a randomized trial, adults that received education by continence advisors about reducing caffeine intake did so, and they also noticed improved urgency and frequency.11 One multivariate analysis1 found no association between coffee drinking or alcohol consumption and urinary incontinence.
One study12 found that 30% of women with stress incontinence reported straining at stool as a young adult compared with 4% of women without urogynecologic symptoms. While no rigorous research has assessed whether treating constipation affects incontinence, anecdotal evidence suggests that it does in certain instances. One study13 found that urine loss during provocative maneuvers such as coughing can be significantly decreased by crossing the legs or by crossing the legs and bending forward; whether this is a satisfactory long-term management strategy has not been addressed.
For individual patients, a voiding diary may provide important clues for managing women with incontinence. By reviewing this 3- to 7-day record of the patient's intake and output, the patient and clinician can discuss whether decreasing the fluid intake or increasing the voiding interval is appropriate. Although there is no role for decreasing fluid intake in women with normal urinary output, women who drink excessive volumes of liquid can benefit by restructuring their intake. Restricting fluids after 6:00 PM often results in fewer nocturnal voids and may be particularly helpful to those with restricted mobility. Certain professions are associated with more infrequent voiding than professions with easier bathroom access. When a woman becomes aware of the connection between her daily stress incontinence and concomitant very full bladder, she usually finds the time to void once in the middle of the working day. Of note, in a randomized trial, women with stress incontinence were assigned to one of three groups: increase fluid by 500 mL, maintain fluid intake at baseline level, or decrease fluid by 300 mL. The adherence to any of the fluid recommendations was poor and precluded the ability to draw conclusions. However, subjectively, women who reported fewer incontinence episodes felt that the most significant learning from the study was their recognition of the need to increase fluid intake.14
In some women, caffeinated and carbonated beverages contribute to incontinence or urinary frequency. Advise women who report a high caffeine intake (more than 6 cups per day) to slowly taper down to 1 to 2 cups per day. Making this change gradually over 3 to 4 weeks helps to minimize the caffeine withdrawal headaches experienced by many patients.
Although it is more widely known that certain medications may exacerbate urge incontinence, few clinicians are aware that alpha-adrenergic blocking medications (e.g., prazosin) can precipitate stress incontinence.15 Women whose incontinence is temporally correlated with such medications may present for evaluation and consideration of surgery; the leakage resolves when the antihypertensive medication is discontinued or changed to a different type. Similarly, Julian16 described a series of women undergoing treatment for “stress incontinence.” The cough stress test result was negative, and the women had excessive vaginal discharge caused by unopposed estrogen diagnosed. This highlights the importance of documenting stress incontinence before proceeding with surgical treatment.
An acute urinary tract infection can cause transient stress incontinence, urgency, and urge incontinence. Some strains of Escherichia coli can inhibit the periurethral alpha-adrenergic fibers; incontinence resolves once the infection is treated.17
Some women report that their stress incontinence is related only to certain provocations. Coughing is a common precipitating event. Occasionally, nonsmoking women with chronic coughing may have complete resolution of the incontinence after treatment for sinusitis, allergic rhinitis, or other treatable conditions. Although the patient would still have stress incontinence if she coughed vigorously, treatment of the cough can make her condition less symptomatic, rendering surgery for the incontinence unnecessary.
Scheduling regimens, including bladder training, are frequently used to manage urge incontinence18,19 but may also play a role in stress incontinence. After a 6-week course of formal bladder training, Fantl and coworkers20 reported more than a 50% reduction in incontinence episodes and amount of fluid lost on pad testing in women with stress incontinence and those with detrusor instability. In this program, the voiding interval is initially set between 30 and 60 minutes. A reasonable goal, achieved over 6 weeks, is a 2- to 3-hour interval between voids. The woman is encouraged to void on schedule, even if the desire to void is not experienced. Relaxation and distraction techniques are used to suppress urge sensations. The woman meets weekly with her health care provider, reviews the bladder diaries, and receives further instruction and reinforcement. When the number of incontinent episodes and untimed voids are decreased, the voiding interval can be increased by 15 to 30 minutes. Bladder training does not impose fluid modifications.
The urethra and bladder contain a rich supply of estrogen receptors,21 and it is therefore biologically feasible that estrogen replacement can affect postmenopausal urogenital symptoms. Estrogen promotes vaginal cellular maturation22,23,24 and maintains a more beneficial vaginal flora. Clinically, estrogen improves symptoms of atrophy such as vaginal dryness, irritation, and burning.25,26 Whether it also improves urinary incontinence is controversial. Earlier nonrandomized studies with subjective outcome measures27,28,29 reported often dramatic improvements in symptoms of urge and stress incontinence. However, randomized studies specifically examining stress incontinence report mixed results. Neither Samsioe30 nor Cardozo31 found a difference in stress incontinence symptoms in patients randomly assigned to receive estrogen or placebo. Wilson and coworkers32 reported greater subjective improvement in stress-incontinent women using estrogen than those using placebo, but no objective difference was found in urethral pressure profile or volume of urine lost. In a randomized, double-blind, placebo-controlled study, Walter and associates33 observed cure or improvement in incontinence symptoms in 43% of women administered estriol, compared with no improvement in the placebo group. Significant improvement was seen in 1-hour pad test results, although not in the number of leakage episodes per 24 hours.
Fantl and colleagues34 examined the efficacy of estrogen replacement therapy in 83 hypoestrogenic women with urinary incontinence. In this randomized controlled trial, there was no difference between the placebo and treatment groups in the number of incontinent episodes, volume of urine loss, standardized quality of life outcomes, or subjective improvement. Although of unclear significance at this point, results from a few large multivariate analyses counterintuitively suggest that women using estrogen replacement are more likely to have urinary incontinence.35,36
A recent large prospective randomized trial of estrogen replacement therapy versus placebo in women with heart disease37 provided a look at the effect of estrogen plus progesterone on urinary incontinence. Of 1525 women with incontinence at baseline, a similar proportion of women improved after 4 years: 21% of those using estrogen therapy and 26% of those receiving placebo. However, significantly more women receiving estrogen plus progesterone worsened than did women receiving placebo (39% versus 27%, respectively).
It is possible that estrogen administered locally, without additional progesterone, may have different effects. To date, there are no methodologically sound studies that carefully assess this question.
Of interest, there is emerging evidence that selective estrogen receptor modulators (SERMs) may impact pelvic floor dysfunction in varying ways. In 6926 women randomly assigned to either raloxifene or placebo, women receiving the SERM were less likely to undergo surgery for prolapse during the 3-year study period than women receiving placebo (.7% vs. 1.5%).38
In contrast, an investigational study of levormeloxifene, a different type of SERM, halted the participation of the 2924 women after 10 months because of a marked increase in urinary incontinence (17% incidence in the study group compared with 4% in the placebo group) and pelvic organ prolapse (7% in the study group versus 2% in the placebo group).39
Alpha-Adrenergic Stimulating Agents
The bladder neck and proximal urethra in women contain predominantly alpha-adrenergic receptors. Stimulation of these receptors produces smooth muscle contraction,40 resulting in increased bladder outlet resistance. Some women with stress incontinence improve after treatment with alpha-adrenergic-stimulating drugs. Most studies found improvement only in women with mild41,42,43 but not severe44 stress incontinence.
Estrogen and alpha-sympathetic agonist medications may have an additive effect in the treatment of stress incontinence. Studies of combined therapy45 generally show an improvement in symptoms; however, the objective effects on urethral pressure are less consistent. Ahlstrom and colleagues46 treated 29 women with stress incontinence with estriol alone or with combined estriol and phenylpropanolamine. Those receiving the combined therapy had a better clinical response and more significant improvements in maximum urethral closure pressure. However, other studies have found no significant differences in objective results in women using combined therapy.
Based on these results and the low risk of short-term estrogen use, it seems reasonable to recommend a 6- to 8-week course of estrogen therapy in women with vaginal mucosal atrophy and urinary incontinence. If clear-cut improvement is not reported after this trial period and the woman has no other desires to use estrogen replacement therapy, treatment can be discontinued. There is no information about whether estrogen therapy improves the effectiveness of physiotherapy in the treatment of stress incontinence.
Imipramine hydrochloride, an antidepressant with alpha-adrenergic agonist and anticholinergic effects, appears to improve symptoms in some women with stress urinary incontinence. In a nonrandomized trial in which 30 women received 75 mg of imipramine hydrochloride daily for 4 weeks, 71% stated that they were continent after treatment with imipramine but 9% had no improvement. The maximum urethral closure pressure and functional urethral length increased significantly after treatment. However, the closure pressure also increased markedly in women who remained incontinent.34 Prospective, randomized, controlled trials to verify this high success rate are not available. Anecdotal clinical evidence indicates that the success rate is lower.
Theoretically, beta-adrenergic blocking agents may be expected to potentiate an alpha-adrenergic effect, thereby increasing resistance in the urethra. Kaisary47 reported success in treating stress-incontinent patients with the beta-adrenergic blocking agent propranolol. Although such treatment has been suggested as alternative drug therapy in patients with sphincteric incompetence and hypertension, no reports of such efficacy have appeared. Other reports have found no significant changes in urethral pressure profile measurements in normal women after beta-adrenergic blockade.
Serotonin Reuptake Inhibitors
While the medications described previously have not had a major treatment role in the treatment of stress incontinence because of limited effect and unacceptable side effects, new serotonin and norepinephrine reuptake inhibitors show promise. There are data to suggest that serotonergic activity may be involved in stress and urge forms of incontinence. A recent study of an animal model revealed that rats treated with clomipramine to induce a state of depression demonstrated changes related to overactive bladder, including alterations in voiding frequency and unstable bladder contractions.48 These authors found that administration of fluoxetine reversed these changes. This suggests that treatment with selective serotonin reuptake inhibitors may be of potential benefit in women with mixed stress and urge incontinence. There is evidence that depression and urinary incontinence are associated49; this type of therapy with SSRIs might prove promising for these combined conditions in humans.
A recent large, randomized, placebo-controlled trial assessed the impact of duloxetine on stress incontinence in women.50 Five-hundred fifty-three women between the ages of 18 and 65 years were randomly assigned to placebo or to three different doses of medication. After 12 weeks, the median incontinent episode frequency decreased in all the groups, with the greatest reduction in women using the largest dose of 80 mg per day. In this group, the median incontinence frequency decreased by 64%, compared with 41% in women using placebo. Five percent of women using placebo discontinued treatment, compared with 15% of those receiving the highest dose of medication.
Pessaries, until recently considered a relic of the past, have experienced resurgence as physicians and patients become more conscious of the long-term risks and failure rate of surgical procedures. During the past decade, more types of vaginal devices and urethral devices have become available. Some patients choose to wear a pessary or device as the final therapy for their pelvic floor dysfunction, others use such a device to temporize before considering surgery, and many wear such a device only when undertaking an activity that results in urine loss, such as exercise.
In our experience and that of others,51 adverse outcomes from wearing a pessary, such as vaginal abrasions or ulcerations, are rare if a patient is able to insert and remove the pessary on her own and remove it at least once per week overnight. Table 1 shows the types of vaginal devices we use to treat stress incontinence. Others are in development and may be released in the next few years.
Before a pessary fitting session, we pretreat all postmenopausal women who have vaginal atrophy with estrogen cream. We instruct women to come to the fitting session with a moderately full bladder, which allows testing of the efficacy of the pessary. Women are instructed to mimic vigorous activity in the clinic area (e.g., brisk walking, jumping jacks, straining) to ensure that the device stops leakage and is comfortable. After the patient demonstrates the ability to void with the pessary in situ, she is sent home with the best fit. We find that the best fit in the office fails to prevent incontinence for approximately one fourth of women, who then return for further fitting. When the best type and fit is determined, we teach women to insert and remove the device. Because they generally remove the pessary at least once per week, they rarely encounter excessive or malodorous vaginal discharge and have little use for creams other than estrogen.
Two newer devices for stress incontinence include the urethral insert (or plug) and the urethral patch. The United States Food and Drug Administration (FDA) has approved the Reliance Urinary Control insert (UroMed, Needham, MA, USA) and the Femsoft insert (Rochester Medical, Stewartville, MN, USA). A sterile insert is placed into the urethra by the patient and removed before a void, after which a new sterile insert is reinserted. Such inserts are appropriate for women with relatively pure stress incontinence, no history of recurrent urinary tract infections, and no serious contraindications to bacteriuria (e.g., artificial heart valves). Although many women do experience bacteriuria and transient hematuria secondary to insert use, this is often not clinically significant. However, because the inserts are not completely without risk, their use should be supervised by a physician familiar with the expected findings and management of problems. Cost of these types of devices may also prevent some patients from selecting this type of therapy. The urethral inserts are sterile devices that are not intended for reuse. To date, there has been no analysis on the cost-effectiveness of these treatments.
The urethral patch that fits over the meatus (Impress Softpatch; UroMed, Needham, MA, USA) appears to be appropriate for women to try without physician guidance. A new patch must be reapplied after each void. Preliminary study of this modality found that the patch dislodged during detrusor contractions; this finding alleviates concerns that a woman with detrusor instability who wears the patch may be at risk for ureteral reflux because of outlet obstruction.
Two reusable external suction caps that fit over the urethral meatus have received FDA approval (Fem Assist; Insight Medical, Bolton, MA, USA and CapSure; Bard, Covington, GA, USA). This type of patch is reapplied after voids and reused for up to 1 week. Versi and coworkers52 found that 44% of women using this type of device were dry on a 48-hour pad test. At the time of this writing, neither device is currently available on the market, but either may be available again in the near future.
|PELVIC FLOOR MUSCLE TRAINING|
Pelvic floor muscle exercises have been one of the traditional nonoperative treatments used for management of female stress urinary incontinence. First described by Kegel in 1948, the exercises are designed to strengthen the musculature of the pelvic floor to provide better support for the bladder and urethra.53 Kegel described concomitant use of a pneumatic device, the perineometer, to measure the strength of the pelvic contractions during the exercises. This was essentially an early type of biofeedback therapy.
Although the exercises are designed to improve muscle strength of the pelvic floor, the exact physiologic basis for symptom improvement is not completely understood. Bernstein described an increase in the thickness of the pelvic floor musculature as measured by perineal ultrasound in a group of 40 stress-incontinent women treated with pelvic floor muscle exercises.54 She also observed a significant improvement in subjective and objective incontinence symptoms in this series. However, the increase in muscle thickness itself was not statistically correlated with the subjective improvement or demonstrable decrease in leakage. A recent study of pelvic floor electromyography (EMG) revealed that women with stress incontinence have impairment in vaginal EMG compared with continent controls.55 In this study, 60 women with incontinence were compared with 28 healthy control subjects. Fifty of the incontinent women completed the study and were evaluated. Overall, the vaginal EMG function improved successively over the 4-month treatment period for the incontinent patients. This was associated with improvement in clinical symptoms. This supports the contention that pelvic floor muscle exercises help to strengthen muscular function.
Despite widespread advocacy for pelvic floor muscle exercises in the medical literature and the lay press, success rates reported in clinical trials have varied widely. A number of different clinical parameters have been examined to assess their influence on the ultimate success rates for pelvic floor muscle exercise programs. Pre-intervention incontinence severity has been examined as a predictor of successful outcome. Elia and Bergman examined the ability of pretreatment urodynamic parameters to predict continence outcomes.56 They found that the urethral closure pressure significantly improved after pelvic floor exercise therapy in those patients who reported symptomatic improvements. They also found that 86% of the women with mild incontinence experienced significant subjective improvements, compared with only 13% of the women with severe incontinence. However, this finding has not been consistent across studies. Bø and Larsen found that patients with more severe incontinence demonstrated a greater degree of response to therapy and continence improvement than patients with less severe pretreatment incontinence.57
Training women how to perform pelvic floor exercises may also improve continence outcomes after pregnancy and delivery. In a recent trial of 268 primigravida women, Reilly and colleagues randomized women to monthly physiotherapy appointments beginning at 20 weeks' gestation until delivery versus control.58 The control subjects received verbal instruction in pelvic floor exercise, but no physiotherapy sessions. They found that fewer women in the formal physiotherapy group reported postpartum incontinence (19.2% compared with 32.7%; RR: .59; 95% CI: .37 to .92). They concluded that antenatal pelvic floor muscle exercise improved outcomes, although it did not necessarily improve pelvic floor strength.
Motivation and proper technical education on the part of the clinician teaching the exercises and by the patient herself may affect outcomes. The patient must be able to perform the exercises correctly for them to be expected to work. Kegel himself reported that 30% of women could not correctly contract their pelvic floor muscles on the first attempt.59 Bump and coworkers examined the pelvic floor muscle contraction efficacy in a group of 47 women provided with standard verbal and written instructions.60 Using urethral pressure profile measurements, they found that 49% of the women were subsequently able to correctly perform an ideal contraction. However, 25% of the patients actually contracted the pelvic muscles in a manner that could possibly increase their incontinence. They concluded that simple verbal or written instructions were not an effective educational technique to ensure that patients could adequately contract the pelvic floor muscles. Most clinicians advocate teaching pelvic floor muscle exercises during pelvic examination or urodynamic assessment, when more directed instruction can be provided. This also permits the clinician to determine if the exercises are being performed correctly from an anatomic perspective. Despite intensive therapy, it is still possible that women may not be able to correctly contract the pelvic floor muscles. In one study, Bø and coworkers found that even after 6 months of training, four (7.6%) of 52 women could not correctly perform the muscle contractions.61 In some cases, the patients were actually performing a Valsalva maneuver, which would increase urinary leakage.
After patients learn how to perform the exercises correctly, compliance becomes an important issue. Intuitively, repeated training sessions or other types of motivational or educational reinforcements would be thought to increase the likelihood of success. However, studies that have examined the effectiveness of adjunct training methods designed to reinforce the initial education have yielded conflicting results. Nygaard and associates tested the usefulness of an audio tape to help coach patients at home.62 After initial training in the clinical setting, patients were randomized to perform the exercises at home with no additional training reinforcements or with an audio tape designed to coach the patient in correct exercise techniques. At the end of the 3-month trial, 44% of all patients reported at least a 50% reduction in incontinent episodes, and 56% of those patients who completed the entire 3-month course of therapy reported this level of improvement. These investigators identified significant improvements in patients with stress incontinence and in patients with urge and mixed urinary incontinence. However, there was no statistically significant difference in continence outcomes or treatment compliance between the group who exercised with the audio tape and those who exercised independently. They concluded that the reinforcement provided by the audio tape did not influence treatment outcomes or patient dropout rates.
A cross-sectional study of 129 women who were prescribed pelvic floor muscle exercise as a treatment for incontinence identified several predictors of continued compliance.63 Subjects were asked to complete a questionnaire that was developed based on behavioral theories, focus group interviews, and published literature regarding adherence to treatment regimens. A regression analysis of the data revealed that the amount of urinary loss per episode and an individual patient's perception of her ability to perform the exercises were the most important predicators of continued exercise.
The overall reported efficacy of pelvic floor muscle exercises is highly variable. A recent Cochrane database analysis of 43 clinical trials yielded divergent results.64 Most of the trials were quite small, and many had limited follow-up. Patient discontinuation of therapy was also variable. Loss to follow-up ranged from none (reported in 10 trials) or less than 10% (in 7 trials) up to a maximum of 41% in some studies. Overall, these authors concluded that, based on the available literature reviewed, pelvic floor exercises were better than placebo or no therapy for women with stress or mixed incontinence. They also found that intensive forms of training tended to yield better overall results than standard training techniques. The data were too limited to permit conclusions about the effect of combining pelvic floor muscle exercises with other forms of therapy. They concluded that large, high-quality, randomized trials were necessary to determine the overall efficacy of pelvic floor exercise in women with urinary incontinence.
A number of studies have examined the long-term efficacy of pelvic floor muscle exercises for treatment of urinary incontinence. Hahn and coworkers examined 170 women who underwent a training program for pelvic floor muscle exercises.65 At the end of the training period (mean: 4.7 months), subjective cure was reported by 23% of patients, 48% felt they were improved, and 29% reported they were unchanged. A follow-up survey was performed using a mailed questionnaire. The mean follow-up was 4.3 years (range: 2 to 7 years), with a response rate of 89.4%. Thirty-eight of the 152 patients who completed the survey had undergone surgery for their urinary incontinence. Of the remaining 114 patients, 11% reported they were still cured, and 44% remained improved. However, only 15% of the patients reported that they continued to exercise regularly four- to six-times daily as initially instructed, and 6% had stopped exercises completely. These investigators concluded that continued therapeutic training might help to improve long-term efficacy. In a similar study of 76 incontinent women initially treated with pelvic floor muscle exercise training, Mouritsen and associates reported that at 1-year follow-up, 30% were subjectively cured and 17% were improved.66 They also found that patients with adequate estrogen status and those who were not overweight had better overall outcomes.
Estrogen replacement has been tested in combination with pelvic floor muscle exercise. Ishiko and colleagues randomized 66 women to pelvic floor exercises alone or in combination with estrogen replacement using estriol 1 mg/day.67 They demonstrated improvement in both groups, although the treatment effect was more pronounced for the group that also received hormone replacement. The effect was sustained for 18 months in subjects with mild incontinence and 12 months for those with moderate incontinence. However, the findings are limited by the short duration of follow-up and small patient numbers.
A variety of other techniques and devices have been devised to supplement pelvic floor muscle training exercises, including weighted vaginal cones, biofeedback therapy training, and electrical stimulation. Each of these treatment adjuncts are considered separately.
The idea of using weighted vaginal cones to test pelvic floor muscle contraction strength and augment exercises was first introduced by Plevnik in 1985.68 The original description included a set of nine cones that ranged from 20 to 100 grams. The cones are placed intravaginally, with the tapered end directed toward the introitus. The patient attempts to retain the cones in the vagina when she is up and active. Theoretically, the weighted cone acts to provide a constant stimulus for proprioceptive feedback. The patient is instructed to actively contract her pelvic floor muscles when she feels the cone slipping out of the vagina. This provides a form of biofeedback in an attempt to improve pelvic floor muscle exercise efficacy.
Vaginal cones have also been used for diagnostic purposes. In a comparison of continent and incontinent women, Hahn and coworkers found that continent women were better able to retain heavier vaginal cones.69 Although some incontinent women were able to retain heavy cones, radiographs revealed that the cones often were incorrectly positioned in a transverse orientation below the level of the pelvic floor muscles. In a review of the literature, Bø found generally poor correlation between ability to hold cones and pelvic floor muscle contractile strength.70 Although cone retention tests have been described as a theoretical option, no well-designed trials have been performed to validate cone retention as a method of clinical evaluation.
Few randomized trials of cone therapy have been reported. Cammu and Vannylen randomized 30 patients to cone therapy and 30 patients to pelvic floor muscle exercises without cones.71 However, 47% of the patients in the cone group discontinued treatment before the completion of the 12-week treatment program. All of the remaining patients in the cone group chose to discontinue therapy at the end of the 12 weeks. The investigators compared outcomes in three groups: those who completed cone therapy, those who discontinued cone therapy, and those who completed the exercise program without cones. There was no statistically significant difference in continence outcomes identified between any of the groups. These investigators concluded that the high discontinuation rate and lack of significant improvement compared with standard pelvic floor muscle exercise programs does not justify the use of weighted vaginal cones. A more recent study randomized women to pelvic floor exercises alone or in combination with either vaginal cones or pressure biofeedback.72 These authors found that patients in all three treatment groups demonstrated improvement in urinary incontinence symptoms. However, there was no significant difference between the groups.
A recent Cochrane database analysis examined the use of vaginal cones from a meta-analysis perspective.73 This review of 15 studies included a total of 1126 patients, 466 of whom received vaginal cones. The data revealed that cones were better than no active treatment (failure to cure incontinence RR: .74; 95% CI: .59 to .93), but not significantly better than pelvic floor muscle exercises alone (RR: 1.09; 95% CI: .86 to 1.38) or electrostimulation (RR: 1.0; 95% CI: ,89 to 1.13).
Many of these small published studies on the use of vaginal cones have methodologic flaws that prevent adequate comparative interpretation for measured outcomes. In particular, the paucity of randomized trials and the relatively high discontinuation rate of patients treated with vaginal cones make this a less appealing treatment option. Well-designed randomized studies are needed to answer some of the fundamental questions about the ultimate usefulness of vaginal cones in the treatment of female stress urinary incontinence. In particular, there is a specific need for good long-term data on the efficacy of vaginal cone therapy. The data available to date suggest that the primary benefit of cones may be to help motivate or remind patients to continue to perform pelvic floor exercises.
Biofeedback therapy uses monitoring devices that provide visual or auditory cues to the patient during pelvic floor muscle exercises. This helps to indicate that the patient is contracting the correct muscle and provides a measure of the strength of contractions. The concept of biofeedback is based on the belief that the added stimulus of the feedback information helps improve motor skill development.74 Most programs involve multiple formal training sessions with the patient using the device during each session to monitor progress. The devices typically use urethral, vaginal, or rectal probes that are connected to monitors to provide information directly back to the patient. Workman and colleagues have also validated the quality of information provided by surface electromyographic patch electrodes for this type of biofeedback therapy.75 Burgio and her colleagues76 found that patients who received formal biofeedback improved the selective control and contraction strength of their perineal muscles more than patients who only received verbal feedback.
Studies comparing pelvic floor muscle exercise training with and without the biofeedback component have yielded mixed results. Berghmans and coworkers randomized 40 patients to pelvic floor muscle exercises alone or with biofeedback.77 All patients completed 12 sessions, with three sessions per week. Patients were followed-up with subjective survey reports, voiding diary, and pad tests. There was significant improvement compared with baseline in both groups. However, these investigators observed that improvements occurred earlier in the course of therapy in the biofeedback group. Glavind and coworkers randomized 40 patients in a similar fashion and found that patients who underwent biofeedback therapy demonstrated significantly greater improvement in both objective symptoms, as measured by pad tests and subjective symptoms, compared with those who underwent exercises alone.78 The effect appeared sustained at 2 to 3 years based on survey data. They concluded that biofeedback training may have added to the motivation to continue the exercises after the conclusion of the formal training sessions. Sherman and associates similarly reported that most patients in their series did not desire additional therapy after the conclusion of a biofeedback training program.79
Biofeedback has also been reported as a successful option for treating patients with urge or mixed incontinence. Stein and coworkers reported that biofeedback was successful in treating 36% of their patients with stress incontinence and 43% of their urge incontinence patients.80 They also found a significant reduction in daytime urinary frequency and nocturia in the urge incontinence group. Susset and associates reported favorable outcomes in a group of 15 women with mixed stress and urge incontinence.81 After 6 weeks of biofeedback therapy, 12 of the 15 patients were subjectively cured, and 13 were demonstrated to be dry on pad tests.
Several studies have demonstrated efficacy of biofeedback therapy in the management of urinary incontinence in elderly women. The technique helps patients to specifically identify which muscles should be contracted during pelvic floor muscle exercises.82 Burgio and coworkers demonstrated treatment efficacy in elderly patients with stress and urge incontinence.83 Overall, patients experienced an 82% to 94% reduction in incontinence episodes after a mean of 3.5 treatment sessions. However, in a randomized trial in which 222 women between ages 55 and 92 years were randomly assigned to receive 8 weeks of biofeedback-assisted behavioral training, 8 weeks of behavioral training with only verbal feedback based on vaginal palpation, or 8 weeks of self-administered behavioral treatment using a self-help booklet, there were similar reductions in incontinence episodes in all groups, and similar improvements in quality of life in all groups. However, the patients' perceptions of treatment were best in the two active intervention groups.84 Most studies have examined biofeedback therapy as a center-based treatment. Stoddart reported the usefulness of a portable device for home use.85 This was essentially a perineometer, which permitted the patient to monitor the strength of pelvic contractions during exercises at home. A variety of more sophisticated portable devices have been developed for home use. We have found this approach to biofeedback therapy particularly useful for patients who have are unable to contract the pelvic muscles effectively at their clinic visit and who live a great distance from our facility. Patients complete their initial training session at the clinic. Subsequent sessions are completed by the patient at home. Most of these devices permit the patient to store data that can be downloaded and reviewed at her follow-up appointments. This permits the patient to perform regular biofeedback sessions at home and still review the efficacy of muscle contractions and overall progress with her clinician.
The efficacy and usefulness of biofeedback therapy remains controversial. Other studies have failed to identify a significant difference between patients treated with biofeedback and those treated with pelvic floor muscle exercises alone.61,86,87 Most studies have involved small numbers of patients who have been followed-up for a limited period of time. Hay-Smith and associates conducted a meta-analysis of 10 randomized trials that compared biofeedback assisted pelvic muscle training with pelvic muscle training alone. There was no statistically significant difference between the groups for rates of self reported cure or improvement, the number of daily leakage episodes, results of pad testing, or measures of muscle activity.64 Larger prospective studies with longer follow-up will help to sort out these issues.
Electrical stimulation is another modality that is often used to supplement pelvic floor muscle exercises in an attempt to strengthen these muscles in women with urinary incontinence. The treatment has been used for stress, urge, and mixed incontinence. Theoretically, electrical stimulation may help to restore the normal reflex physiology in the muscles of the pelvic floor. It is also hypothesized that, over time, electrical stimulation can induce a conversion of fast-twitch to slow-twitch fibers, which may improve resting tone of the pelvic floor musculature and the endurance level to perform effective contractions.88 A wide variety of treatment regimens have been tried with alterations in the duration, frequency, and intensity of stimulation. Short-term maximal stimulations appear to produce this effect most efficiently, particularly for patients with stress incontinence.
The principles of electrostimulation therapy assume that patients possess an intact perineal innervation that can allow them to respond to this form of treatment. It is the stimulation of the pelvic floor afferent fibers that induce the contractions.
However, there are still questions about the exact physiologic effects of electrical stimulation of the pelvic floor. A recent study of the urodynamic changes induced during electrical stimulation revealed that insertion of the vaginal probe may actually induce some physiologic changes.89 In this study, 30 women underwent urodynamics in three phases: a baseline study, a study with a vaginal probe inserted but without electrical stimulation, and a study with electrical stimulation. There was no effect on cystometry with the vaginal probe compared with baseline. However, there were improvements in the urethral functional profile length (FPL), maximum urethral closure pressure (MUCP), and the area of the resting urethral pressure profile (UPP) after insertion of the vaginal probe. There were no significant differences in results between studies with the probe in place but with and without electrical stimulation. This finding suggests that insertion of the probe itself may induce some type of physiologic change that affects these urodynamic parameters.
Although electrical stimulation may enhance the contractions, the ability to perform voluntary contractions is still critically important. Bø and Talseth examined the changes in urethral pressure produced during voluntary contractions and contractions caused by electrical stimulation alone.90 They found that the voluntary contractions actually increased urethral pressure to a significantly greater degree than vaginal electrical stimulations. They did find a moderate level of pain associated with the electrical stimulation, which could influence a patient's compliance behavior.
Studies of the efficacy of electrical stimulation have yielded conflicting data. In a randomized, controlled, double-blind study of 121 incontinent women, Brubaker and colleagues91 found that, compared with a sham probe, electrical stimulation did improve outcomes in patients with urge incontinence, although not in patients with stress incontinence. Luber and Wolde-Tsadik also failed to find a significant improvement in patients treated with electrostimulation for stress incontinence compared with sham controls.92
In contrast, Sand and associates reported a significant improvement in patients using electrostimulation for treatment of genuine stress incontinence compared with sham controls.93 Subjective and objective measures of incontinence improved significantly in this population after 15 weeks of therapy. Similarly, Yamanishi and coworkers reported that 60% of patients in the electrostimulation group reported good impressions of therapy compared with only 8% of sham control patients.94
A recent prospective study of a large cohort of women in Norway revealed that home-based electrical stimulation may be effective for the treatment of stress urinary incontinence.95 In this study, 3198 women were treated with electrical stimulation using a home-managed device. They performed pelvic floor exercises in combination with electrical stimulation at home. The overall self-reported improvement was 61%, with 29% of women reporting cure or much improvement. A survey of the treating physicians revealed their assessment of the patient outcomes to include 33% cured or much improved with 55% improved overall. Of the total, 44% of patients reported improvement of incontinence severity using a validated severity scale. In a separate report of patient compliance, this group reported that only 12% of patients discontinued therapy, and 79% would recommend it to others.96 They concluded that home-based electrical stimulation was a feasible option for women with stress urinary incontinence.
All of the adjuvant therapies used to augment pelvic floor muscle exercise require a strong commitment on the part of the patient. Susset and coworkers examined predictive variables of success in a single-arm trial of electrostimulation therapy for urinary incontinence.97 They concluded that patient compliance is the strongest predictor of continued success. An underlying theme in almost all of the available research on these modalities is that patient compliance is critical for any measure of improvement. In this way, the continence outcomes of pelvic floor muscle exercises are similar to the results of all other types of physical exercise programs.
Reimbursement has been an issue for many of the nonsurgical treatment options for the management of stress urinary incontinence. In October 2000, the Centers for Medicare and Medicaid Services (CMS) issued a national coverage decision regarding the use of biofeedback and electrical stimulation for the management of urinary incontinence.98 This policy decision was formulated on the basis of an expert review of available data by a panel of physicians involved in the care of incontinent patients. This decision should help to provide treatment for a larger segment of patients who otherwise would not have had access to these forms of therapy.
Unfortunately, no treatment exists that can cure all women with stress incontinence. Although conservative management cures fewer women than surgery does, the decreased risk and expense and reasonable improvement rates associated with conservative therapy combine to make this approach attractive as a first treatment option for many women.
8. Subak LL, Vittinghoff E, Saxton J, et al: The effect of weight reduction of urinary incontinence: A randomized controlled trial. Presented at Annual Urogynecologic Society Meeting, San Francisco 2002
9. Bump RC, McClish DM: Cigarette smoking and pure genuine stress incontinence of urine: A comparison of risk factors and determinants between smokers and nonsmokers. Am J Obstet Gynecol 170:579, 1994
17. Nergardh A, Boreus LO, Holme T: The inhibitory effect of coli-endotoxin on alpha-adrenergic receptor functions in the lower urinary tract: An in vitro study in cats. Scand J Urol Nephrol 11:219, 1977
33. Walter S, Kjaergaard B, Lose G, et al: Stress urinary incontinence in postmenopausal women treated with oral estrogen (estriol) and alpha-adrenoreceptor-stimulating agent (phenylpropanolamine): A randomized double blind placebo-controlled study. Int Urogynecol J 1:74, 1990
39. Goldstein SR, Nanavati N: Adverse events that are associated with the selective estrogen receptor modulator levomeloxifene in an aborted phase III osteoporosis treatment study. Am J Obstet Gynecol 187:521-527, 2002
41. Beisland HO, Gossberg E, Moer A, et al: Urethral sphincteric insufficiency in postmenopausal females: Treatment with phenylpropanolamine and estriol separately and in combination. A urodynamic and clinical evaluation Urol Int 39:211, 1984
46. Ahlström K, Sandahl B, Sjöberg B, et al: Effect of combined treatment with phenylpropanolamine and estriol, compared with estriol treatment alone, in postmenopausal women with stress urinary incontinence. Gynecol Obstet Invest 30:37, 1990
54. Bernstein IT: The pelvic floor muscles: Muscle thickness in healthy and urinary incontinent women measured by perineal ultrasonography with reference to the effect of pelvic floor training: Estrogen receptor studies. Neurourol Urodyn 16:237, 1997
55. Gunnarsson M, Teleman P, Mattiasson A, et al: Effects of pelvic floor exercises in middle aged women with a history of naïve urinary incontinence: A population based study. Eur Urol 41:556-561, 2002
58. Reilly ET, Freeman RM, Waterfield MR, et al: Prevention of postpartum stress incontinence in primigravidae with increased bladder neck mobility: A randomised controlled trial of antenatal pelvic floor exercises. Br J Obstet Gynecol 109:68-76, 2002
67. Ishiko O, Hirai K, Sumi T, et al: Hormone replacement therapy plus pelvic floor muscle exercise for postmenopausal stress incontinence. A randomized, controlled trial J Reprod Med 43:213-220, 2001
69. Hahn I, Milsom I, Ohlsson BL, et al: Comparative assessment of pelvic floor function using vaginal cones, vaginal digital palpation and vaginal pressure measurements. Gynecol Obstet Invest 41:269, 1996
75. Workman DE, Cassisi JE, Dougherty MC: Validation of surface EMG as a measure of intravaginal and intra-abdominal activity: Implications for biofeedback-assisted Kegel exercises. Psychophysiology 30:120, 1993
82. Burns PA, Pranikoff K, Nochajski TH, et al: A comparison of effectiveness of biofeedback and pelvic muscle exercise treatment of stress incontinence in older community-dwelling women. J Gerontol 48:M167, 1993
84. Burgio KL, Goode PS, Locher JL, et al: Behavioral training with and without biofeedback in the treatment of urge incontinence in older women: A randomized controlled trial. JAMA 288:2293-2299, 2002
93. Sand PK, Richardson DA, Staskin DR, et al: Pelvic floor electrical stimulation in the treatment of genuine stress incontinence: A multicenter, placebo-controlled trial. Am J Obstet Gynecol 173:72, 1995
94. Yamanishi T, Yasuda K, Sakakibara R, et al: Pelvic floor electrical stimulation in the treatment of stress incontinence: An investigational study and a placebo controlled double-blind trial. J Urol 158:2171, 1997
95. Indrekvam S, Sandvik H, Hunskaar S: A Norwegian national cohort of 3198 women treated with home-managed electrical stimulation for urinary incontinence: effectiveness and treatment results. Scand J Urol Nephrol 35:32-39, 2001
96. Indrekvam S, Hunskaar S: Side effects, feasibility, and adherence to treatment during home-managed electrical stimulation for urinary incontinence: A Norwegian national cohort of 3,198 women. Neurourol Urodyn 21:546-552, 2002
97. Susset J, Galea G, Manbeck K, et al: A predictive score index for the outcome of associated biofeedback and vaginal electrical stimulation in the treatment of female incontinence. J Urol 153:1461, 1995
98. Thompson DL: The national coverage decision for reimbursement for biofeedback and pelvic floor electrical stimulation for treatment of urinary incontinence. J Wound Ostomy Continence Nurs 29:11-19, 2002