eMedicine Specialties > Obstetrics and Gynecology > Prolapse and Incontinence

Urinary Incontinence, Medical and Surgical Aspects: Treatment

Author: Michael O'Shaughnessy, MD, FACOG, Assistant Chief, Director of Urogynecology, Assistant Clinical Professor, Department of Obstetrics and Gynecology, University of California at San Francisco, UCSF Fresno University Medical Center
Contributor Information and Disclosures

Updated: Feb 9, 2007

Treatment

Medical Therapy

Medical therapy of stress urinary incontinence can include pharmacology therapy, behavioral therapy, biofeedback, pelvic floor conditioning, pelvic floor stimulation, bladder training, use of pessaries or occlusive devices, lifestyle modification, ISC, and treatment of medical comorbidities. Each of the therapies is discussed. In some instances, combinations of 2 or more therapies prove beneficial. Some types of incontinence (eg, DI) are more amenable to medical therapy than other types (eg, GSI). Some experts recommend a trial of medical therapy before considering surgical treatment. Others believe that if the incontinence is severe and correctable by surgical means, a trial of medical therapy is not mandatory and does not need to be performed if the informed patient chooses to proceed directly to surgery.

Pharmacologic therapy

Alpha-agonists, such as midodrine (Pro-Amatine) or pseudoephedrine (Sudafed), may improve symptoms of mild stress incontinence by increasing intrinsic urethral tone due to effects on the urethral sphincter. A slow-release form of pseudoephedrine can be used, allowing once- or twice-daily dosing. Imipramine, which has both alpha-agonist and anticholinergic effects, has been shown to have moderate efficacy (60-71%) in treating stress incontinence. Imipramine appears to work by increasing both urethral closure pressure and functional urethral length. High pretreatment urethral closure pressure has served as a predictor of success.

The recently approved serotonin/norepinephrine reuptake inhibitor duloxetine is the first drug developed and marketed specifically for stress incontinence. The mechanism of action is not known conclusively in humans; however, in animal models, duloxetine seems to increase pudendal motor nerve output via increased levels of serotonin and norephinephrine in the pudendal motor nucleus of the sacral spinal segments. As a result, urethral muscular tone and closure pressure is augmented.

A number of clinical trials have demonstrated the efficacy of duloxetine compared with placebo in the treatment of mild and moderate stress incontinence.

A recent, small, prospective, double-blind, randomized, placebo-controlled trial demonstrated modest efficacy in patients with severe stress urinary incontinence.

In this study, patients with pure, urodynamically confirmed stress incontinence who were awaiting surgery were treated with duloxetine for 8 weeks. All participants had, on average, 14 or more episodes of stress incontinence per week. Significant improvement was observed in the quality of life indices and in frequency of incontinence episodes and use of protective pads in the patients treated with duloxetine compared with placebo. All positive clinical responses were observed within 2 weeks after initiation of therapy–some as early as 5 days.

The most common side effect was nausea, which tended to decrease with continued use. Discontinuation of therapy was significantly more common in the treatment group with equal numbers of patients withdrawing because of nausea, vomiting, worsening of hypertension, and headache. Other common side effects included constipation and dry mouth. At the end of the 8-week trial, 20% of the treatment group were no longer interested in surgical therapy versus 0% in the placebo arm.

Another small study demonstrated similar results with 24% of the patients taking duloxetine declining their planned surgical therapy. Of note, 48% of the patients stopped the medication due to side effects at the 40 mg twice daily dose schedule used in this study (Duckett, 2006).

Estrogen therapy may have several positive effects in patients with stress incontinence who are estrogen deficient. Estrogen may increase the density of alpha-receptors in the urethra. In addition, the vascularity of the urethral mucosa is increased and the coaptive abilities of the urethral mucosa may be augmented. In theory, these effects would seem to translate into improved continence; however, several studies stand in opposition of these assumptions.

A number of small studies show oral estrogen therapy to be of no clinical benefit to women with stress incontinence or DI. In a subgroup analysis of postmenopausal women enrolled in the Heart and Estrogen/Progestin Replacement Study (HERS), worsening of incontinence occurred in 39% of patients in the hormone treatment group compared with 27% of patients in the placebo group (P=.001). In the Women's Health Initiative Study, women with baseline incontinence being treated with combined or unopposed estrogen oral therapy also showed exacerbation of symptoms significantly more often than women in the placebo group. In addition, women with no baseline incontinence in the hormone exposed groups developed symptoms more often than those in the placebo group.

Both of these trials present level 1 evidence against oral hormone therapy to treat incontinence. No adequate studies of local estrogen therapy exist. Local urogenital treatment provides more rapid and reliable effects in treating genitourinary atrophy and deserve study as a preoperative adjunct.

Discontinuing alpha-blockers and ACE inhibitors may be beneficial in patients who are taking these agents and have mild GSI. Discontinue only if safe and if effective alternative medications can be substituted. Appropriate consultation is recommended before initiation of any medication changes.

Pelvic floor rehabilitation

Kegel exercises have been shown to improve the strength and tone of the muscles of the pelvic floor. During times of increased intra-abdominal pressure, tensing of these muscles tightens the connective tissue supports surrounding the urethra. Thus, pressure transmission to the urethra may increase, and the urethra compresses shut during times of increased stress. The exercises consist of voluntary contractions of the muscles of the pelvic floor. Because both fast-twitch and slow-twitch muscle fibers are found in the levator ani complex, both rapid contractions and slow contractions held for maximal duration should be performed to achieve the best possible results. If instructions for Kegel exercises are provided verbally or in written form alone, a significant percentage of individuals perform repetitive Valsalva maneuvers or gluteal contractions. Therefore, instructing the patient in the examining room by having her squeeze the examiner's intravaginal or intrarectal finger is important.

The patient can confirm that she is using the correct muscles at home by periodically performing the contractions during voiding with the goal of interrupting the urinary stream. The number of daily contractions needed to achieve optimal efficacy is unknown, but, at a minimum, probably 3-5 sets of 10 exercises are required. Approximately 6-12 weeks of exercises are required before improvement is noted, and 3-6 months are needed before maximal benefit is reached. Patients with severe neuromuscular damage to the pelvic floor may not be able to perform Kegel exercises even with proper instruction.

Vaginal cones are weighted devices designed to increase the strength of the pelvic floor muscles. Typically, the cones are retained for 15 minutes twice a day, and the weight of the cones is gradually increased. Although probably no more efficacious than properly performed Kegel exercises, this method may be preferred by some individuals.

Biofeedback in the form of visual or auditory signals may be an effective method of exercising the pelvic floor. The responses may provide incentive and confirmation of proper performance of the muscle contractions. Efficacy rates of 50-85% have been demonstrated. Electrical stimulation of the pelvic floor can be achieved with a vaginal or rectal probe. Low-frequency electrical pulses of 50-100 Hz are used. This method can be useful in treating individuals who desire therapy via pelvic floor rehabilitation but have difficulty performing pelvic muscle contractions on their own. Improvement rates of close to 90% have been reported.

Extracorporeal magnetic resonance therapy has been introduced recently into the therapeutic armamentarium for stress incontinence. The NeoControl unit (Neotonus, Marietta, Ga) recently was approved by the Food and Drug Administration (FDA) for this purpose. Resonating magnetic flux within a magnetic field induces electrical depolarization of targeted nerves and muscles. No probes are required. The patient simply sits on a chair containing the magnetic device. A small study achieved an improvement rate of 77% after 8 weeks of therapy, with 56% of patients being completely dry.

Prosthetic devices

The Impress Softpatch (UroMed Corporation, Needham, Mass) is an adhesive foam patch designed for a single use. A hydrogel adhesive anchors the patch over the urethral meatus. In one study, 52% of women with mild-to-moderate stress incontinence were dry, and 82% were improved with the use of this device. The Reliance Urinary Control Insert (UroMed Corporation, Needham, Mass) is a small catheterlike device that is inserted into the urethra. The balloon is inflated with air. The single-use device is removed before voiding by pulling on an attached thread. Approximately 80% of patients are dry with the use of this insert, with an additional 12% greatly improved. Complications such as migration into the bladder and UTI have been reported.

The FemAssist (Insight Medical Corporation, Boston, Mass) and CapSure Shield (Bard Urological, Covington, Ga) are silicon devices that seal the urethral meatus with an action similar to a suction cup. These devices can decrease significantly the urine loss in short-term pad test studies. In a more extended study of the device, only 2 of 31 women finished a 6-month trial. Reasons cited for drop out were lack of efficacy, poor adhesion, discomfort, and difficulty placing the device. This study points out that short-term results in highly motivated individuals may not be generalized to longer-term use under real-life circumstances.

The Introl Bladder Neck Support Prosthesis (UroMed Corporation, Needham, Mass) is an insertable vaginal device with arms that provide support on each side of the bladder neck. If properly fitted, the prosthesis can achieve cure rates of approximately 80% for stress incontinence. Incontinence pessaries also are available for use in mild stress incontinence. These ring- or dish-shaped devices usually are reinforced in the area that sits under the bladder neck/proximal urethra. A modicum of success has been achieved with these specially designed pessaries.

Behavioral approaches

Timed, frequent voiding can be used to minimize incontinence, especially if the bladder is kept empty before incontinence-producing activities. Symptoms of urgency and frequency can develop over time with this strategy due to decreased bladder capacity. Anticipatory pelvic floor contractions can be taught to patients to cut down on incontinence episodes. The patient is taught to perform a strong pelvic floor contraction just before anticipated episodes of increased intra-abdominal pressure, such as a cough or a sneeze. Modifying activities occasionally can be a solution to incontinence-related specific activities. For example, if a woman experiences incontinence only during high-impact aerobics, substitution of another fitness activity, such as swimming, may solve the incontinence problem.

Treatment of medical comorbidities

As previously discussed, disease treatment, including weight loss, smoking cessation, control of asthma, and chronic constipation, may minimize incontinence episodes. A few studies show significant decreases in urinary incontinence in small cohorts of women experiencing weight loss. For the most part, these women underwent obesity surgery. In a study of 138 patients who lost 50% or more of their excess body weight, stress incontinence dropped from 61% to 11%. In another smaller study of 12 women with morbid obesity, 9 had relief of stress incontinence after obesity surgery. Presumably, nonsurgical methods of weight loss could lead to similar benefits. Weight loss as an adjunct to incontinence surgery has not been studied specifically but often is recommended.

Medical management of detrusor overactivity

Management options include pharmacologic, behavioral, and electrical stimulation. These options may be tried sequentially, but combined approaches often work best. Idiopathic DI (ie, nonneuropathic) is more amenable to medical approaches than detrusor hyperreflexia (ie, neuropathic). The latter often may require a more invasive surgical approach. Clinically significant improvement in DI can be expected in approximately 80% of the cases using one or more of the described medical approaches.

Anticholinergic agents

The clinical and urodynamic effects of blocking cholinergic receptors in the bladder are to increase bladder capacity, increase the volume threshold for initiation of an involuntary contraction, and decrease the strength of involuntary contractions. Propantheline bromide is an anticholinergic agent that has been used to treat DI. Propantheline commonly is prescribed in dosages of 15-30 mg every 4-6 hours. Because gastrointestinal absorption is poor, many recommended that propantheline be taken on an empty stomach. Typical anticholinergic adverse effects can be expected, including dry mouth, constipation, dry eyes, blurred vision, orthostatic hypotension, and increased heart rate. This agent probably should be avoided by patients with heart disease and closed-angle glaucoma. Improvement rates in various studies generally have been approximately 50%. Propantheline is no longer considered a first-line drug for DI due to relatively poor efficacy and a high incidence of adverse effects.

Tolterodine (Detrol) is a potent antimuscarinic agent for treating DI. In animal models, the drug has shown selectivity for the urinary tract over the salivary glands. Tolterodine has performed well in clinical trials, showing comparable efficacy to oxybutynin with lower discontinuance rates. The dosage range is 1-2 mg twice daily. Two small studies examining the use of transdermal scopolamine in the treatment of DI recently were reported. The results of these studies are conflicting in terms of both efficacy and tolerability of adverse effects.

Trospium (Sanctura) elicits antispasmodic and antimuscarinic effects. It acts by antagonizing acetylcholine effect on muscarinic receptors. Parasympathetic effect reduces smooth muscle tone in the bladder. Trospium is indicated to treat symptoms of overactive bladder (eg, urinary incontinence, urgency, frequency). The typical dose is 20 mg PO bid taken on an empty stomach at least 1 h before meals. The dose is reduced with CrCl <30 mL/min to 20 mg PO hs. Elderly individuals (ie, >75 y) may require a dose reduction (eg, 20 mg PO qd) to avoid adverse effects. Mild anticholinergic effects (eg, dry mouth, constipation, dry eyes, blurred vision) may occur.

Solifenacin (VESIcare) is a competitive muscarinic receptor antagonist that causes anticholinergic effects and inhibits bladder smooth muscle contraction. The initial dose is 5 mg PO qd, which may be increased to 10 mg/d if tolerated and warranted. Precautions include renal or hepatic impairment (do not exceed 5 mg with CrCl <30 mL/min or moderate hepatic impairment [Child-Pugh class B]), controlled narrow-angle glaucoma, history of prolonged QT interval, bladder outflow obstruction, or decreased GI motility. The tablet must be swallowed whole (not crushed) with liquid.

Another anticholinergic agent is the extended-release product darifenacin (Enablex). It has high affinity for M₃ receptors involved in bladder and GI smooth muscle contraction, saliva production, and iris sphincter function. The initial dose is 7.5 mg PO qd. After 2 wk the dose may be increased to 15 mg/d based on response. Do not exceed 7.5 mg/d if moderate hepatic impairment (Child-Pugh class B) is present or coadministered with potent CYP-450 3A4 inhibitors.

A prospective, randomized, placebo-controlled, double-blind study demonstrated efficacy of this medication with regard to reductions in incontinence episodes, decreases in frequency and urgency, and improved bladder capacity. Additive toxicity may occur if administered with other anticholinergics (eg, antihistamines). Coadministration with CYP-2D6 substrates that have a narrow therapeutic index (eg, flecainide, thioridazine, TCA [imipramine]) may cause toxicity of these other 2D6 substrates. Coadministration may also increase midazolam or digoxin levels.

In a study of 188 healthy volunteers receiving therapeutic (15 mg) and supratherapeutic doses, no prolongation of the corrected QT interval could be documented (Serra, 2005). Increasingly, prolongation of the QT interval has been recognized as a potential problem in antimuscarinic drugs as well as medications of many different classes. Individuals with pharmacologic prolongation of the QT interval may be at increased risk for potentially fatal polymorphic ventricular tachyarrhythmia. Additional risk factors for this problem include female gender, advanced age, hypokalemia, and polypharmacy. No direct studies compare the incidence of prolonged QT intervals or clinically concerning tachyarrhythmias among commonly prescribed antimuscarinic agents. More commonly observed adverse effects include xerostomia, constipation, and blurred vision. Darifenacin must be swallowed whole; do not chew, divide, or crush.

Tricyclic antidepressants

These drugs have complicated direct and indirect effects on the lower genitourinary tract. They possess both a central and peripheral anticholinergic effect. Tricyclic antidepressants also are alpha-adrenergic agonists and central sedatives. The resultant clinical effect is bladder muscle relaxation and increased urethral sphincter tone. The pharmacological effects make these drugs good choices for mixed incontinence, nocturia, and nocturnal enuresis. Imipramine is the most widely used tricyclic for urologic indications. The dosage range is 25 mg 2-4 times per day.

In addition to anticholinergic adverse effects, serious allergic reactions have been reported, although rarely. Cardiotoxicity rarely is problematic at the low doses used for treatment of DI. Central effects, such as sedation and tremor, may be troublesome to some patients. On occasion, this author has found prescribing imipramine at bedtime and a musculotropic agent in the daytime to be useful. Studies have demonstrated an approximate cure rate of 60% in individuals with DI.

Musculotropic relaxants

Musculotropic relaxants depress smooth muscle activity directly but at a site distal to the cholinergic receptor. Relaxants also may work, in part due to anticholinergic and local anesthetic properties at the level of the bladder. Oxybutynin is the prototype drug in this class. The typical dosage is 5 mg 2-4 times per day. Adverse effects are related mostly to the anticholinergic effects. Lower dosages, such as 2.5 mg 2-3 times a day, may be more appropriate for elderly patients.

Good-to-excellent results have been obtained in clinical trials, with improvement rates ranging from 61-86%. Oxybutynin is available in syrup and extended release formulations. The extended release form is dosed 5-15 mg once daily and is of comparable efficacy to the parent drug. One study used specially prepared oxybutynin suppositories in patients who were intolerant of anticholinergic adverse effects when taking the oral form. Most of these patients were elderly. Adverse effects via this route of administration were less, but the overall symptomatic improvement rate was only 48%.

Flavoxate hydrochloride is a direct smooth muscle relaxant with very weak anticholinergic properties. Few adverse effects are associated with its administration, but efficacy has been questionable. An observational study of flavoxate use in clinical practice described good results in decreasing daytime and nighttime urgency and the number of voids, but urge incontinence was not examined. The usual dosage is 100-200 mg 3-4 times per day.

Dicyclomine hydrochloride is a smooth muscle relaxant that has been used most commonly to treat irritable bowel syndrome. Moderate efficacy has been reported with a dosage of 20 mg taken orally 3 times daily. Adverse effects mostly are anticholinergic.

Calcium channel blockers

Terodiline was once a very popular drug for the treatment of DI in Europe but has since been withdrawn from the market due to a potential for serious adverse cardiac effects. In a small study, verapamil was no more effective than the placebo and less effective than oxybutynin. Verapamil combined with oxybutynin was more effective than oxybutynin alone. A small study showed magnesium hydroxide beneficial for some patients with sensory urgency and DI. The presumed mechanism of action is through calcium antagonism. More work is needed before this treatment is recommended.

Potassium channel openers

This class of drugs relaxes smooth muscle by increasing potassium efflux, with resultant membrane depolarization. Researchers who found supersensitivity of the detrusor muscle to depolarizing stimuli, such as potassium, in individuals with urge incontinence provides the theoretical basis for the use of these agents in patients with DI. One problem in the development of potassium channel openers for use in bladder disorders has been the lack of organ specificity. Overall progress toward the development of a viable clinical formulation has been disappointing.

Prostaglandin inhibitors

Prostaglandins may have an excitatory role in bladder contractility. Prostaglandin inhibitors, in theory, may block bladder contractility. Clinical trials with agents, such as indomethacin, have shown mixed and generally not impressive results. One research group reported evidence of the role of a relative prostacyclin deficiency in the promotion of bladder contractions. Pharmacotherapy to increase the ratio of prostacyclin to other prostaglandins has not been investigated to date.

Beta-adrenergic agonists

These agents relax beta-adrenergic receptors that contain smooth muscle. Such receptors exist in the human bladder. Studies of terbutaline and clenbuterol have yielded mixed results. The role of these drugs as adjuncts to other pharmacologic therapies has not been explored.

The agent 1-desamino-8-D-arginine vasopressin (DDAVP) has been used in children with nocturnal enuresis with good results. The hormone causes water to be reabsorbed from the renal collecting system. Reduction in nighttime urine production may be beneficial in patients with DI and a significant degree of nocturia. Caution is needed when using this drug in elderly patients. Do not use in patients with significant CHF and in children younger than 5 years (eg, water intoxication).

Estrogen

Sensory-urgency symptoms in postmenopausal females, which are common in patients with GSI and DI, often respond to estrogen therapy. No studies demonstrate the efficacy of estrogen in the treatment of DI.

Intravesical pharmacotherapy

Intravesical treatment of DI and hyperreflexia with capsaicin, the main pungent ingredient of hot peppers, has recently been evaluated. Improvement rates of 40-100% have been reported. Observation has ranged from 1-60 months. In the largest of these series, 44% of patients with detrusor hyperreflexia secondary to MS were dry. Positive findings from an ice water test indicating bladder hypersensitivity have been suggested as a method of selecting patients for capsaicin therapy. The proposed mechanism is through desensitization of capsaicin-sensitive unmyelinated afferents. Neuronal damage through osmotic swelling also may occur. One percent intravesical lidocaine is administered 5-15 minutes before capsaicin is administered. Approximately 50-100 mL of 1-2 mmol capsaicin is mixed in 30% ethanol with saline. The solution is left in the bladder for approximately 30 minutes.

Adverse effects can include transient worsening of irritative symptoms or incontinence, perineal pain, a burning sensation, hematuria, and UTI. Administration in the office or hospital, continuous blood pressure monitoring, and the ability to treat acute hypertension are recommended in patients with spinal cord injuries due to the rare possibility of exacerbation of autonomic dysreflexia. A small urethral catheter and balloon occlusion of the vesical neck are used to minimize spillage and leakage. At times, administration of capsaicin is best accomplished under general anesthesia.

Resiniferatoxin, a naturally occurring pungent substance from the Euphorbia resinifera plant, has been shown to have very potent capsaicinlike activity. This substance has been used successfully to treat DI and hyperreflexia. In one small study, some patients who failed capsaicin therapy responded to resiniferatoxin. More research is underway to clarify the role of these therapies in the treatment of urge incontinence disorders, sensory urgency, and interstitial cystitis.

Intravesical oxybutynin has been used in patients who are nonresponsive to the oral form or have severe adverse effects. The medication is self-administered following clean catheterization. This therapy has been shown to be safe and efficacious. Studies have shown that tissue and plasma concentration of the drug are higher after intravesical administration than after oral administration. Despite higher plasma levels, adverse effects appear to be minimal. This finding suggests that a hepatic metabolite may be responsible for many of the adverse effects observed after oral administration.

Functional electrical stimulation

Functional electrical stimulation (FES) has been found useful in therapy for both GSI and DI. In addition to increasing skeletal muscle tone and strength in the pelvic floor, afferent stimulation of the pudendal nerve can result in reflex inhibition of detrusor contractions. The 2 main modes of FES therapy are long-term stimulation and short-term maximal stimulation. Long-term therapy requires the use of an intravaginal or intra-anal probe for several hours a day. Low intensity, subthreshold stimulation is used.

The optimal length of therapy is uncertain, but, most likely, at least several months of therapy are required for reeducation. Improvement rates have ranged from 73-90% in several small series. Reeducation of the bladder and/or nervous system occurs in some individuals, as demonstrated by a carryover effect of 45% for 6 months after discontinuation of therapy. Long-term FES therapy can be useful in patients who have been refractory to other forms of therapy. This treatment also benefits some patients with detrusor hyperreflexia. Patient acceptance of this therapy can be low due to the discomfort of wearing the probe for several hours each day.

Short-term maximal stimulation therapy was developed because it is more practical, and high intensity stimulation may produce a better inhibitory effect. Maximal inhibition of involuntary bladder contractions takes place at stimulation intensity levels that are 2-3 times sensory threshold levels. The closer the proximity of the stimulating device to the selected nerve, the lower the intensity can be and remain effective. In practical terms, maximal tolerance levels usually are approximately 1.5-2 times the sensory perception threshold. Short-term maximal therapy uses high-intensity stimulation for 15-30 minutes once or twice a day. Treatment generally is continued over several weeks. Improvement rates of 52-77% have been documented. Carryover effects of 31-92% have been shown for as long as 1 year after therapy.

Transcutaneous electrical nerve stimulation (TENS) has been tried in patients with DI using several different methods. One method using a positive electrode applied to the area of the anal sphincter and a negative electrode to the posterior tibial nerve has yielded mixed results in 2 studies. TENS of the S2-S3 dermatomes has been tried with some success. An interesting method of alternating stimulation of the hamstring and quadriceps muscle groups recently was reported. In this study, those patients with DI and those patients with detrusor hyperreflexia were included. The subjects underwent 20 minutes of stimulation per day for 14 consecutive days. Clinical improvement was observed in 68%. The mechanism of detrusor inhibition by this method of TENS is unclear but may involve increases in segmental inhibitory tone due to manipulation of peripheral neural input.

Interferential therapy is a type of TENS in which external electrodes are positioned over the pelvis, and the interference produced by the competing electrical fields produces low-level nerve stimulation in the area of interference. A small study showed a 90% improvement rate in 20 patients with DI that was unresponsive to pharmacotherapy. In 18 months of observation, no complications were reported, and no recurrences were observed. Extracorporal magnetic resonance therapy, in addition to being beneficial for stress incontinence, reportedly also has been successful in the treatment of DI. FES via implantable electrode requires a surgical procedure and, thus, is discussed in Surgical therapy.

Acupuncture

Acupuncture is a promising alternative therapy. A 77% improvement rate with 63% of patients dry was reported in a study of weekly therapy sessions for 10-12 weeks.

Behavior modification

Behavioral interventions are based on the assumption that cortical control over a hyperactive micturition reflex can be established or reestablished. Clearly, behavioral therapies can be successful in the highly motivated patient in the short-term. Long-term efficacy is much less certain, and relapse rates, when reported, have been high. In addition to a highly motivated patient, this type of therapy requires a dedicated team to provide support and reinforcement to the patient.

Biofeedback is a type of learning where the patient receives information in the form of visual or auditory stimuli concerning a physiologic function. The patient learns to control this normally unconscious function. In the case of DI, the patient learns awareness of the external striated urethral sphincter that may, in turn, improve the urethrovesical inhibitory reflex. This therapy often is combined with Kegel pelvic floor exercises.

Bladder training can be performed in a number of ways. This author examines the patient's voiding diary and tries to arrive at a reasonable voiding interval that minimizes urge and urge incontinence. The patient uses the chosen interval to void by the clock during waking hours and void as needed at night. The interval is increased by 15 minutes per week until reaching a voiding interval of approximately 3-4 hours. Bladder training can be conducted with or without simultaneous pharmacotherapy. Subjective response rates of 85% and objective response rates of 50% have been achieved with short-term observation.

Intermittent catheterization

This type of management is most appropriate for patients with detrusor hyperreflexia and functional obstruction. Many of these patients have detrusor-sphincter dyssynergia and are at risk for pyelonephritis and upper tract injury. Some patients with urge incontinence and coexisting hypofunctioning detrusors may benefit from self-catheterization. For example, some diabetic patients with bladder neuropathy may have instability requiring bladder-relaxing pharmacotherapy but, at the same time, may have intermittent detrusor hypofunctioning with poor emptying. The addition of bladder-relaxing drugs may worsen the baseline poor detrusor function, resulting in retention and overflow incontinence. In some cases, the solution may be to combine bladder relaxing medical therapy with ISC.

Vaginal prosthetic devices

A disposable vaginal device made of polyurethane, which originally was designed to provide bladder neck support as a treatment for stress incontinence, has been found moderately effective in patients with DI. A combined cure and improvement rate of 56.7% was documented. Few adverse effects were reported, but 21% of the subjects withdrew from the study. Of those who withdrew, 67% cited discomfort or difficulty inserting or retaining the device. The mechanism of action in treating DI is uncertain.

Some theorize that the mechanism is preventing entrance of urine into the proximal urethra and the subsequent triggering of an uninhibited detrusor contraction. The authors offer an alternative hypothesis, ie, local stimulation of the vaginal mucosa leading to detrusor inhibition. More studies of this device need to be conducted. A bladder neck support prosthesis was tested in women with mixed incontinence. A modest reduction in DI was demonstrated, but the discontinuance rate was high. Reasons cited for discontinuation included poor efficacy, the inability to fit the device, or a poor fit.

Medical therapy for nocturnal enuresis

Such basic measures as evening fluid restriction and daytime bladder training can be beneficial. Nasal vasopressin at 10-40 mg decreases nighttime urine production. Imipramine (2 mg/kg/d) has been one of the most common pharmacologic therapies. Oxybutynin and other anticholinergics have been used. Ephedrine recently was found to be moderately beneficial in a recent trial. Some believe that this agent works by increasing urethral alpha-sympathetic tone. Pharmacologic therapies can help, but the underlying disorder often returns after discontinuation. Conditioning therapy with moisture-sensitive alarms are effective. Positive results usually remain even after the device is removed.

Surgical Therapy

Procedures for stress incontinence, although varied in terms of approach, share the common goal of stabilizing the bladder neck and proximal urethra. More than 250 procedures, some of which are modifications of older procedures, have been devised. Many others have been disregarded because of poor long-term efficacy, technical difficulties in performance, and/or unacceptable complication rates. Only a few procedures have been the subject of sufficient scientific scrutiny to demonstrate safety and good long-term efficacy.

One major problem in studying these procedures is that the same basic procedure may yield different results in different hands. This may be due to differences in experience and technical expertise, small variations in technique, choice of suture or supportive material, and differences in preoperative patient selection and postoperative care. A few procedures have been shown to yield good results in various clinical settings by a variety of surgeon researchers. These procedures (eg, Burch retropubic urethropexy, MMK procedure, suburethral sling) should be considered the criterion standard against which new procedures should be judged.

Another variable to be considered when examining the literature concerning procedure efficacy is the method and criteria used to define success. Some studies define a successful outcome as a patient who is dry. Others may use degrees of clinical improvement or the patient's estimate of postoperative satisfaction to determine successful outcomes. Investigators have used such techniques as questionnaires and telephone surveys to gather this information. Others have used more objective criteria, such as stress testing or more extensive urodynamic examinations. Many studies have unacceptably large numbers of patients lost to observation. Comparing studies or performing a meta-analysis of data can be difficult or impossible due to the various criteria used. No widely accepted standardized methods or criteria for defining or measuring surgical outcomes in incontinence surgery have been devised to date.

Mixed incontinence

Surgical therapy for GSI is not contraindicated in the setting of mixed (ie, GSI and DI) incontinence. Many authors have recommended medical treatment of the DI portion of the incontinence disorder before proceeding with surgery. With this strategy, some believe that a certain percentage of patients improve to the point that they no longer are seeking surgical therapy. If the DI is treated successfully with medical therapy, some patients find that the GSI component is minor and are content with conservative management, such as pelvic floor exercises.

In a significant minority of patients, surgery for GSI also results in improvement or cure of the DI component. This especially may be true if the stress incontinence symptoms clearly preceded the urge symptoms chronologically. Unfortunately, many patients find no improvement in their urge incontinence after surgery for GSI, and a few experience worsening symptoms. Although the surgery accomplished what it was intended to do (ie, GSI), the procedure may be viewed as a failure by the patient due to the persisting urge incontinence. In the setting of mixed incontinence counseling, painting a realistic picture of the dual nature of patient incontinence disorder, therapeutic options, and possible outcomes especially is important.

Intrinsic sphincter deficiency

Surgical therapy for ISD is centered on the concept of increasing urethral coaptation. If urethral hypermobility coexists with ISD, most experts believe that the chosen procedure also should address this problem. In the clinical situation of a combined disorder of urethral hypermobility and ISD, suburethral sling procedures are considered the therapy of choice. Retropubic urethropexy may be effective in this situation, but some surgeons' failure rates have been higher if ISD is a significant part of the clinical picture.

Some authors have commented that periurethral bulking injections can be used as a secondary procedure if the results of a retropubic urethropexy are not satisfactory. Periurethral injections alone generally have not been recommended with ISD and hypermobility, but efficacy in this situation may be better than once thought. With a patient who is very frail, bulking injections may be the only reasonable option. ISD in the absence of hypermobility can be addressed with a sling procedure, although failure rates may be higher. With a nonfunctioning, immobile drainpipe urethra, an obstructing sling can be performed with the understanding that many, if not most, of these patients must perform ISC.

Alternatively, urethral mobility can be restored with a urethrolysis procedure followed by performance of a sling procedure under the same anesthesia. Because of the safety, low morbidity, and their relative noninvasive nature, periurethral bulking injections are considered the first-line therapy in this clinical situation. Finally, an artificial sphincter may be indicated, especially if other options have failed.

Detrusor instability/hyperreflexia

Surgical therapy should be considered only in severe and refractory cases. Surgical approaches include bladder augmentation procedures; denervation procedures; and, as a last resort, urinary diversion. An exciting, new innovation has been the implantable sacral nerve-stimulating device. In selected patients, this device offers good efficacy with less morbidity than the other surgical options. As mentioned above, procedures for GSI do treat coexisting DI in some instances, but these procedures are not recommended as a primary treatment for urge incontinence. Few individuals have extensive experience with surgical treatment of DI, and most case series are limited by small numbers.

Preoperative Details

Selection of the surgical procedure

The success or failure of incontinence surgery largely depends on selecting the appropriate procedure for the appropriate patient. A correct diagnosis with knowledge of the pathophysiology in each individual case is the cornerstone of planning good surgical therapy. Some cases may be obvious, for example, a large vesicovaginal fistula. In this case, a procedure would be chosen to excise and close the fistula. A procedure designed to treat stress incontinence would not be expected to work because of the specific pathophysiologic findings.

In other cases, the cause of the incontinence may not be as clear. For example, stress incontinence symptoms can be due to urethral hypermobility, intrinsic urethral sphincter deficiency, stress-induced DI, or any combination of the above. A patient history and a physical examination are helpful but should not be relied upon exclusively when planning stress incontinence surgery. Urodynamic studies can help narrow the diagnostic focus further but are not perfect tests for various reasons. A large percentage of surgical failures may be related to an incorrect or incomplete diagnosis. The following are other factors that may influence the type of surgery chosen and the specific approach:

• The need to perform other indicated procedures: A vaginal approach may be chosen if other vaginal pelvic floor repairs are to be performed. An abdominal approach may be favored in cases requiring abdominal exploration for other reasons.
• Body habitus/obesity
• Previous surgical approaches/failures, previous scars
• Age and medical condition of the patient: A patient who is frail may be less tolerant of a prolonged procedure, and an abdominal approach may potentially result in more postoperative morbidity.
• Evidence-based literature regarding success and complication rates of particular procedures
• Experience and skill of the surgeon
• Personal track record of the surgeon with particular procedures and approaches
• Patient preference
• Medical comorbidities (eg, chronic obstructive pulmonary disease [COPD], chronic cough, chronic constipation)
• Unexpected intraoperative findings or conditions

Preoperative preparation

Appropriate preoperative preparation of the patient may enhance the positive results of a well-performed incontinence procedure. The following are important points to be considered preoperatively:

• Estrogen replacement therapy may enhance urethral, bladder, and vaginal tissue integrity. Estrogen replacement can improve surgical healing; augment the effect of the procedure on urethral competency; and, ultimately, translate into better surgical outcomes. Replacement therapy should be continued postoperatively.
• Smoking cessation reduces chronic cough, improves the condition of connective tissue, and enhances the effect of endogenous or pharmacologic estrogen.
• Treatment of contributing medical problems, such as asthma, COPD, diabetes, connective tissue disorders, UTIs, and DI
• Weight loss in cases of obesity
• Good nutrition helps to maximize tissue integrity and to support good healing. Vitamin C, in particular, may be important in wound healing and connective tissue quality.
• A preoperative commitment on the part of the patient to decrease postoperative activity is vital; the patient cannot engage in any heavy lifting or straining for at least 12 weeks.

A thorough discussion of risks, benefits, anticipated success rates, and potential common complications should be undertaken. A discussion of the possible need for prolonged postoperative bladder drainage or ISC is important. Some surgeons recommend that each patient preoperatively learn and demonstrate proficiency at self-catheterization. A patient with realistic expectations and foreknowledge of possible complications or postoperative difficulties is more likely to express satisfaction with the final result than a patient with unrealistic expectations or one who is surprised by adverse events such as prolonged catheterization.

Intraoperative Details

Open retropubic procedures

The modified Burch retropubic urethropexy

In 1961, as described in his landmark publication, Burch was attempting an MMK procedure but found that the sutures were pulling out of the periosteum. Burch next chose "the origin of the so-called fascia surrounding the vagina" as an alternative fixation point. He believed that this site was both feasible and anatomically correct. Burch's initial experience was favorable with what essentially was a transabdominal approach to a paravaginal fascial defect repair. Although his initial results were good, Burch feared that the suture bites could pull out of the fibromuscular attachment sites. Burch next used Cooper ligament as a fixation point; Cooper ligament is a thick fibrous band running along the superior surface of the pubic bone. Burch used number 2 catgut as his suture of choice but thought that steel wire or fascia might be better.

In the original description, 3 sutures were secured on each side of the bladder neck area. Burch noted that the procedure often corrected mild-to-moderate cystoceles, and, indeed at times, he used the procedure for the sole purpose of cystocele repair. Since its original description, the Burch procedure has undergone several important modifications. Many surgeons now use only 2 sutures on each side, 1 set at the level of the mid-to-proximal urethra and 1 set at the level of the bladder neck. Permanent suture often is favored over absorbables.

In 1972, Tanagho described his modification of the Burch procedure. He emphasized avoiding dissection over or close to the urethra to avoid damage to the delicate muscular, vascular, and neural structures intrinsic to this organ. Tanagho described lateral placement of sutures away from the urethra. He also believed that removal of fat from the retropubic space, lateral to the urethra, was an important step aimed at promoting lateral scarification, perhaps improving the long-term efficacy of the procedure. The Burch procedure often is combined with paravaginal defect repair if a significant displacement cystocele coexists with urethral hypermobility and stress incontinence.

• The essential steps of the modified Burch procedure: The patient is positioned in the low lithotomy position with the buttocks slightly over the edge of the table. A Foley catheter is placed and draped over the right leg. The vagina is prepped, and the vulvovaginal area is surrounded by sterile towels so that later access to the vagina can be achieved with minimal contamination. The abdomen is prepared in routine fashion. The following are the remaining steps:
  • The initial lower abdominal incision can be transverse or midline. Muscle-splitting incisions may be required rarely to increase surgical exposure.
  • After the rectus fascia is incised, the rectus muscles are separated in the midline and attachments of the transversalis fascia to the pubic bone are broken bluntly.
  • The loose areolar tissue found in the space of Retzius is teased away from the pubic bone using blunt and/or sharp dissection. Sharp dissection is preferred if the patient has undergone a previous retropubic procedure. Intentional cystotomy can be performed to facilitate dissection in a heavily scarred retropubic space.
  • The urethra and bladder neck are located with the help of a transurethral guidance Foley catheter. Some operators mark the location of the UVJ with a surgical clip placed in the fat overlying the palpated Foley balloon. Other important surgical landmarks include the midline pubic symphysis, the superior ramus, the iliopectineal (ie, Cooper) ligament, and the obturator neurovascular bundle. The Tanagho modification includes clearing the anterolateral aspect of the paravaginal fascia free of fat. Care is taken not to dissect close to or over the urethra and not to disturb the paravaginal plexus of veins.
  • The surgeon places the first 2 fingers of the nondominant hand into the vagina. With the fingers straddling the Foley bulb at the UVJ, upward pressure to the vagina is applied to facilitate suture placement. The bladder is retracted gently, medially. The paravaginal tissue should have a white glistening appearance, and the presence and location of prominent paravaginal veins should be noted. The needle and suture are placed through nearly the full thickness of the vaginal wall, excluding the epithelium. The direction of the needle should be parallel to the urethra or away from the urethra and bladder. After retrieval of the suture, gentle traction of the suture should confirm placement in strong connective tissue and should demonstrate palpable elevation of the lateral vaginal sulcus (see Image 6).
  • Suture pull-out indicates that the suture was not placed deeply enough or that it was placed through bladder muscle. A single secure bite probably is sufficient, but some surgeons prefer to take a second bite in a figure-eight configuration.
  • An alternative method of suture placement is to grasp the digitally elevated vaginal wall with a long Allis forceps, tract lightly on the Allis, and pass the needle directly underneath the forceps. Two sutures are placed on each side. The first set of sutures is placed approximately 2 cm lateral to the mid-to-proximal urethra. The second set is placed 2 cm lateral to the UVJ, as demarcated by palpation of the Foley bulb. Many surgeons use a permanent suture, although similar outcomes can be obtained with absorbable material, especially if a thorough dissection of fat lateral to the bladder neck and urethra is accomplished. Inadvertent puncture of paravaginal veins is a common source of annoying bleeding at this point in the procedure. A second bite of the suture and/or tying of the suture often can stop this bleeding. Use of surgical clips for hemostasis also has been described. The Cooper ligament is identified next on the anterior superior aspect of the pubic bone if not previouslyperformed.Bothendsofeach previously placed suture should be passed through the ligament. Placing both ends through the ligament may provide for the greatest pull-out strength. Elevation of the vaginal tissue with either pressure from the vaginal finger or traction on the suture can help determine the correct location for suture placement along the length of the iliopectineal ligament. Occasionally, aberrant accessory obturator vessels may be encountered coursing along the ligament. These should be avoided when passing sutures.
  • Many methods have been described to judge the correct suture tension and vaginal wall elevation before tying. No studies exist to recommend one method. Described methods include intraoperative correction of the resting Q-tip angle, transvaginal palpation of slight elevation of the anterior vaginal wall, and placement of a finger between the urethra and the pubic bone. Surgeons generally agree that a suture bridge should be present after tying and that the urethra should not be compressed against the pubic bone. Recent work seems to indicate that major elevation of the anterior vaginal wall and/or large changes in the resting urethral angle probably are not needed to achieve a good outcome. Overelevation of the anterior vaginal wall may, in fact, increase the likelihood of such complications as obstruction, voiding dysfunction, de novo DI, and enterocele formation.
  • Just before or just after tying the sutures, some surgeons place Gelfoam in the retropubic space lateral to the urethra and bladder neck. This may provide for increased lateral scarification and also may decrease venous oozing. No data exist for judging the efficacy of this maneuver. Gelfoam is not thought to increase infection rates, and complications related to Gelfoam usage in retropubic procedures have not been reported.
  • Cystoscopy should be performed to exclude the presence of intravesical suture material and to confirm ureteral functioning and patency. Cystoscopy can be performed transurethrally or via suprapubic telescopy through a small cystotomy incision.

Other operative procedures, such as hysterectomy and additional pelvic repairs, can be performed during the same anesthesia by either the abdominal or the vaginal route. If the cul-de-sac is deep, a procedure to prevent enterocele should be considered. These procedures include the Halban and Moschcowitz culdoplasties and the McCall colposuspension. If the peritoneal cavity is entered, the parietal peritoneum should be closed to prevent bowel and omentum from adhering in the opened retropubic space. Closure of the peritoneum also may decrease the chance of subsequent hernia formation. Hysterectomy does not appear to increase the efficacy of incontinence procedures, including the Burch procedure. Hysterectomy should be considered if indicated for gynecologic reasons. If hysterectomy is performed, secure suspension of the vaginal cuff to the uterosacral ligaments or other appropriate structures should be accomplished.

Temporary bladder drainage can be obtained with a suprapubic or transurethral catheter. Neither method has been proved to be clinically superior, but the incidence of bacterial colonization of the urine is less with the suprapubic method if drainage is required for more than 3-4 days. Voiding trials can begin on the second or third postoperative day. Suppressive antibiotic administration during the period of catheterization is not warranted and, theoretically, may lead to the selection of virulent organisms. Parameters for removal of the catheter vary among surgeons, but each depends on demonstrating an adequate voided volume with a low residual volume. Attention to preoperative uroflowmetry studies can help predict which patients may have postoperative voiding difficulties. Closed suction drainage of the retropubic space is not required routinely but, occasionally, may be needed if persistent oozing is encountered. Any drains should exit for separate stab wounds.

Modified Marshall-Marchetti-Krantz procedure

In 1949, Marshall, Marchetti, and Krantz published a landmark article on a new retropubic procedure for stress incontinence. The original procedure consisted of suturing the urethra, bladder neck, and a portion of the bladder dome to the pubic bone and rectus muscle and fascia. Number 1 chromic catgut suture was used to take large double bites of the anterior wall of the vagina and the surrounding endopelvic connective tissue, along with the lateral aspect of the urethra, excluding the mucosa.

The sutures then were passed through the periosteum and midline fibrocartilage of the pubic bone. Bites also were taken of the bladder wall and rectus muscle and fascia to bring these structures into apposition. The procedure left the UVJ in a very high retropubic position. Of 50 original patients, 82% showed excellent results, and another 7% demonstrated improvement. Since that time, modifications of the procedure may have decreased complications and morbidity rates. The essential steps of the modified Marshall-Marchetti-Krantz (MMK) procedure are as follows:

• Positioning and draping of the patient and Foley catheter placement are the same as in the Burch procedure. A 3-way Foley catheter is of great benefit if suprapubic telescopy is planned. The abdominal incision and retropubic dissection also are the same as in the Burch procedure. Sharp dissection of the space of Retzius and purposeful cystotomy may be advantageous in repeat procedures.
• Lateral dissection of fat from the retropubic space is performed only as needed to visualize the white, glistening, periurethral endopelvic connective tissue. Direct visualization of the connective tissue aids in correct suture placement and makes inclusion of the urethral and bladder wall less likely.
• A double throw of suture (permanent is preferred by this author) is taken through the endopelvic connective tissue and vaginal wall at the level of the UVJ. The vaginal epithelium is excluded from these bites. Suture placement is facilitated, as in the Burch procedure, by elevating the tissue vaginally with the surgeon's fingers. These sutures are placed on either side of the bladder neck.
• Two to 3 subsequent sets of sutures are placed just lateral to the urethra, approximately 1 cm apart. These bites come close to, but do not include, the wall of the urethra. The sutures then are placed throughout the fibrocartilage of the pubic bone in the midline. Elevation of the urethra with the vaginal fingers with the intention of opposing the urethra against the posterior surface of the symphysis reveals the correct attachment point of each corresponding suture pair. The sutures are tied, bringing the urethra into direct contact with the back of the pubic symphysis. The remainder of the procedure is similar to the previously described Burch procedure.

The MMK has a well-established record of accomplishment as an efficacious and durable procedure. Some surgeons are critical of the procedure because it tends to overcorrect the hypermobile urethra. Urethral kinking, obstruction, and chronic irritative voiding symptoms may result from over-elevation of the UVJ and from suture placement close to the urethra. Unlike the Burch procedure, the MMK does not correct mild-to-moderate cystocele. Because of the potential for increased morbidity, this procedure is used rarely by this author and only in patients who have failed other procedures.

Ball-Burch procedure

The Ball-Burch procedure is a hybrid retropubic procedure combining conventional Burch sutures with imbricating sutures, gathering bites of endopelvic connective tissue on either side of the urethra. These latter sutures then are tied directly on top of the urethra in the hope that these sutures tighten the urethra, decrease funneling, and add functional length. In the original paper, monofilament delayed absorbable suture was used for the Burch sutures and 2 or 3 chromic sutures (3-0) were used over the urethra. The procedure is designed specifically to treat patients with GSI and low-pressure urethras.

Paravaginal repair

In 1909, White published a paper describing his technique of cystocele repair with a vaginal approach to paravaginal defects. White's work essentially was ignored until the 1970s when Richardson revived interest in paravaginal defect repair, this time via the suprapubic approach. Since that time, the procedure has gained acceptance as an anatomically correct surgical correction of displacement cystocele. As noted by Burch, reattachment of the connective tissue lateral to the UVJ to the arcus tendineus corrects urethral hypermobility and, oftentimes, GSI. Paravaginal repair soon was touted as an anatomical surgical solution to GSI and cystocele. Advocates pointed out that the procedure did not overcorrect urethral hypermobility; thus, postoperative voiding dysfunction and the need for prolonged catheterization occurred less often.

The paravaginal repair has been demonstrated as a less durable and reliable cure for stress than the Burch retropubic urethropexy. Although the paravaginal repair correctly restores bladder neck anatomy, overcorrection may be required in many cases to compensate for neuromuscular damage to the pelvic floor, which is not repairable surgically. Recently, the paravaginal repair has been combined with Burch or MMK sutures at the bladder neck to treat stress incontinence and coexisting cystocele. This approach is called the paravaginal-plus procedure.

The paravaginal repair can be performed with an open retropubic, laparoscopic, or vaginal approach. The procedure involves the reattachment of the endopelvic connective tissue lateral to the urethra and bladder to the arcus tendinous fascia pelvis from the back of the pubic bone to just anterior to the ischial spine. Since this procedure is used rarely by itself to treat GSI, the technical steps are not discussed in detail.

Laparoscopic retropubic urethropexies

Laparoscopic procedures for urinary incontinence have been thrust into the surgical limelight in the past decade. Evidence shows that these procedures can be performed successfully. Evidence of short-term efficacy for many procedures recently has emerged. Evidence of long-term efficacy is anticipated eagerly but is not yet present in most cases. Possible advantages of the laparoscopic approach include an improved magnified view of the retropubic space, less blood loss, less postoperative pain, and quicker recovery.

Disadvantages include a steep learning curve with significantly more complications with early cases, the need for laparoscopic suturing skills, and the need for specialized and costly equipment. Uncertain long-term results are another disadvantage. In addition, some authorities have argued that more rapid recovery and a return to normal activities actually may be a disadvantage in disguise by encouraging heavy lifting and activities in the healing phase that may place an otherwise well-performed repair in danger.

In theory, a procedure performed laparoscopically in a fashion that is identical to the open procedure, other than in approach, should yield similar results. Many researchers have taken this theory to heart and are performing Burch procedures through the laparoscope in a way that is true, even in fine detail, to the accepted open modifications of this procedure. The completion and reporting of studies comparing laparoscopic and open approaches to the modified Burch are awaited eagerly.

Other laparoscopic retropubic urethropexies are completely new procedures. Examples are procedures using mesh, staples, and bone screws. Many of these laparoscopic materials and application devices are marketed to private physicians, even though studies validating their efficacy and safety are lacking. These tools are touted for their ease of use, making laparoscopic urethropexy a possibility for even the occasional laparoscopic or incontinence surgeon. The place for these tools is in the hands of researchers and not in widespread use in day-to-day practice.

The laparoscopic approach to GSI and to pelvic floor surgery, in general, holds promise. Unlike extirpative surgery, equivalency to open approaches is difficult to demonstrate. Until research is completed, the laparoscopic approach should be considered experimental. The argument is plausible that equivalent outcomes should be achieved if the procedure performed laparoscopically is otherwise identical to the open procedure. Scientific proof is better than a plausible argument, and no such argument can be put forth for new laparoscopic procedures. These new procedures should not be adopted at this time.

Needle urethropexies

The development of these procedures stemmed from a desire to develop a simple and minimally invasive procedure for GSI. In 1959, Pereyra described the first such procedure, which used #30 stainless steel wire to support the bladder neck. Since that time, many variations and modifications of the original needle urethropexy have been described.

In general, these procedures involve isolation of the pubourethral ligaments by way of a vaginal entry. Midline anterior, inverted U-shaped, and bilateral periurethral vaginal incisions have been used. Next, digital or instrument perforation of the endopelvic connective tissue is made along the pubic bone at the level of the bladder neck. Access to the retropubic space is obtained in this way. Permanent or absorbable suture then is placed through the pubourethral ligaments and/or paravaginal connective tissue in a helical fashion at the level of the bladder neck. Some procedures also include part or all of the vaginal wall in the suture. A small abdominal incision is made transversely just above the pubic bone and is carried down to the rectus fascia.

Long suture-carrying needles then are passed through the rectus fascia and into the retropubic space using finger guidance through the vagina. The needles are advanced through the previously created defect in the endopelvic connective tissue to emerge through the vaginal incision. The bilateral sutures are threaded through the needles and then pulled back through the retropubic space and rectus fascia to emerge from the small abdominal incision.

Traction on the sutures produces elevation of the UVJ and closure of the bladder neck and proximal urethra. Using a free needle, a bite is taken with 1 strand from each side, through the rectus fascia. The 2 sutures are tied independently over the rectus fascia. The 2 sutures also can be tied to each other over the midline for extra security. The proper tension on the sutures before tying has been determined by various methods, including correction of the Q-tip angle and urethroscopic visualization of proximal urethral closure. No one method has proved to be superior. The following are a few common variations of the needle urethropexy procedure:

• Modified Pereyra procedure: One suture is placed on each side of the bladder neck with helical passes through pubourethral endopelvic connective tissue.
• Stamey procedure: A needle is passed and the suture is threaded and pulled up through the abdominal incision. The empty needle is passed a second time lateral to the first puncture. The other end of the suture is retrieved. The loop of suture that remains in the vagina is attached to a polyester buttress, which is intended to prevent the suture from pulling through the pubourethral tissue.
• Raz procedure: This is similar to the Pereyra procedure except that an inverted U-shaped vaginal incision is used and the helical suturing of the endopelvic connective tissue includes a partial thickness bite of the vaginal wall lateral to the urethra.
• Muzsnai procedure: Three pairs of sutures are used on each side. The sutures incorporate the full thickness of the vaginal wall, excluding the epithelium. The first of these sutures is placed at the level of the UVJ, and the remaining 2 are placed progressively, 1 centimeter more caudally each. The sutures are, in essence, parallel to the urethra.
• Gittes procedure: This is called the no-incision suspension procedure. A small puncture wound is made in the skin bilaterally, approximately 2 centimeters above the pubic bone. The needle ligature carriers are passed though these defects blindly, all the way into the vagina, lateral to the bladder neck. Sutures are threaded and retrieved suprapubically. Full-thickness helical bites of vaginal tissue then are taken with the other end of the sutures at the level of the bladder neck bilaterally. The suture carriers are passed from above a second time into the vagina, and the remaining ends of the sutures are pulled up suprapubically. Both ends of each suture are pulled up and tied. The sutures are cut, and the knots retract into the subcutaneous fat.

The needle urethropexy is a relatively quick and simple procedure for stress incontinence. The short-term results generally have been good, but long-term efficacy, in most studies, has been less favorable. Failure of these procedures often is due to suture pull-out. Many of these procedures rely on single sutures on either side of the bladder neck. Formation of significant retropubic scarring, which is thought to augment the supportive effect of sutures in open retropubic procedures, probably is minimal with needle urethropexies.

These procedures are used rarely at the institution of this author because of their unacceptably high long-term failure rate. A possible role for these procedures is in the patient who is elderly and debilitated with a short life expectancy or in an inactive patient who is expected to place minimal stress on the repair. Some newer procedures have been developed that use bone anchors rather than rectus fascia for the suprapubic attachment. Because the site of suture pull-out more commonly is from the vaginal side, whether bone anchors improve long-term efficacy is uncertain. The possibility of osteitis pubis and osteomyelitis also is a significant concern.

Suburethral sling procedures

Sling procedures of various types are among the oldest and most successful procedures for incontinence. These procedures have been used for approximately 100 years. In the early 1900s, Goebel and Stockell independently described the use of the pyramidalis muscle to form a suburethral sling. In 1942, Aldridge described his classic rectus fascial sling procedure. He developed strips of rectus fascia that were left attached in the midline. These strips of fascia were tunneled through the retropubic space and delivered to the suburethral area that previously had been exposed via a vaginal incision. The strips were sutured in the midline under the bladder neck. Modern sling procedures are based on this procedure. Numerous modifications and simplifications of sling procedures have been described, including patch slings with suture arms, inferior pubic ramus bone attachments, vaginal wall slings, overlapping suburethral fascial flaps, and tension-free slings.

Various materials have been used, ranging from a host of permanent synthetic materials, cadaveric donor fascia, and endogenous rectus fascia to fascia lata and vaginal wall materials. These modifications are a testament to the ingenuity and creativity of incontinence surgeons. Each modification is an attempt to refine sling procedures by improving efficacy, lowering morbidity and complications, simplifying the procedure, and decreasing invasiveness. However, a recent review article pointed out that no randomized trials comparing sling procedures or methods have been conducted. As a result, no solid data exist to help decide which procedure is best.

Surgeons in the last decade have witnessed a renewed interest in sling procedures. This interest can be linked to the realization that, in most cases, a sling does not need to be obstructive to be successful. The most feared complications of sling procedures, including permanent obstruction, urethral erosion, and urethral transection, are related partly to sling tension. Tying slings with little or no tension likely decreases the incidence of these complications. As a result, procedures that were once reserved for only the most severe, recurrent, and refractory cases are becoming accepted as mainstream and even primary procedures. Most recognize that GSI due to hypermobility often carries with it some degree of ISD. Sling procedures especially are attractive in light of this fact because the procedure ameliorates both problems.

The diagnosis of ISD is difficult, and the best test(s) and diagnostic criteria are controversial. Proponents of primary sling procedures argue that if a sling procedure is chosen, the need to diagnose ISD in the setting of stress incontinence and the hypermobile urethra is obviated. Opponents of primary sling procedures state that these procedures may be technically more difficult than other incontinence procedures, and lower complication rates due to less sling tension, although hoped for, have not yet been confirmed.

Tension-free vaginal tape

A promising and less invasive variation of the combined abdominovaginal approach to the sling procedure has been described recently in several studies. This technique uses a thin strip of Prolene mesh with barbed edges to serve as artificial pubourethral ligaments. Both ends of the tape are attached to sharp curved trocars, and the tape is encased in plastic. After minimal suburethral and periurethral dissection is performed transvaginally, the trocars are tunneled in the periurethral area, under the pubic ramus and through the space of Retzius. The trocars are delivered through 2 small transverse incisions just above the pubic bone.

Alternatively abdominal needles, similar to those used in traditional needle procedures can be passed through the small abdominal incisions, through the retropubic space, emerging lateral to the midurethra. When passing the needles, care should be taken to remain in contact with the retropubic surface of the pubic bone throughout. The initial thrust of the needle should be at or just medial to the pubic tubercle and directed slightly laterally while penetrating the rectus fascia and tendon. This will help minimize the chances of injuring the dome of the bladder.

Following the penetration of the abdominal fascia, the direction of the needle is adjusted so that it is directed slightly medially, in the same plane as a line drawn from the ipsilateral shoulder to the suburethral incision. Adjusting the needle direction in this way is important to avoid injury to lateral vessels such the external iliac and femoral. A coupling device is then placed on the needle tips and the standard trocar is, in turn, attached to the coupling device. The trocars are then pushed up through the vaginal tunnels and the space of Retzius using a small amount of upward traction on the abdominal needles to assist. Some surgeons feel that this approach is safer in terms of avoiding major lateral vascular injury. Surgeons familiar with needle procedures may be more comfortable with this approach as well.

The central body of the mesh should be located at the mid urethra, although some experts believe that the final location usually is closer to the proximal urethra and UVJ. When the sling ends are visible through the abdominal incisions, they are grasped with clamps, and the trocars are released. The plastic encasement then can be released in the middle, and the 2 pieces can be pulled up and out through the abdominal incisions. The plastic is intended to minimize contamination of the mesh with vaginal bacteria and to aid in sliding the tape into place. Cystoscopic inspection of the urethra and bladder should be performed, looking for injury and foreign material.

To facilitate adjustment of sling tension, performing the procedure under local anesthesia and sedation is recommended. The awake patient is asked to cough repetitively, while tension on the sling arms is increased until stress-induced leakage is eliminated. At this point, the sling arms are trimmed flush with the skin and released, and the arms retract below the skin level immediately. The sling is not sutured in place. Some believe that friction of the barbed edges of the sling material against the trocar-perforated edges of the endopelvic connective tissue and rectus fascia provides points of fixation and stability. Eventually, scarification may provide additional support. Excellent short-term success rates, low rates of voiding dysfunction, and low rates of sling rejection have been demonstrated. In some series, high rates of bladder perforation at the start of the learning curve have been evident, but, if recognized and corrected, this complication should rarely increase long-term morbidity.

All-vaginal methods

Two all-vaginal methods of creating a suburethral sling have been described.

Patch sling with suture arms

The first involves placement of a patch sling of synthetic material with suture arms. The suture arms are secured via bone anchors to the vaginally exposed underside of the pubic ramus. In addition to the all-vaginal approach, the biggest advantage of this procedure is that using the inferior side of the pubic ramus as an anchoring site makes it very difficult to tie the sling under too much tension. A serious theoretical concern is the potential for osteomyelitis, especially concerning transvaginal placement of bone anchors. At present, this author is not aware of any reported cases of bone infection with this procedure.

Paraurethral fascial sling urethropexy

The second all-vaginal approach is a procedure used extensively at this author's institution. The paraurethral fascial sling urethropexy (PFSU) procedure is described in detail by the procedure's originator, Marvin H. Terry Grody, MD, in his textbook Benign Postreproductive Gynecologic Surgery. The interested reader is directed to this text to supplement the description presented below.

The PFSU procedure begins with a midline vaginal incision as is made in a standard anterior colporrhaphy. The incision is carried to within 1.5-2 cm of the urethral meatus using the relatively avascular space between the vaginal wall and the pubovesicocervical connective tissue. The endopelvic connective tissue then is dissected off the vaginal wall flaps using sharp and blunt dissection as needed. This dissection is carried far laterally until the natural insertion of this tissue is reached at the ATFP or white line. The white line extends from the inferior pubic ramus to just anterior to the ischial spine. This dissection must be complete.

At this point, the bladder neck and proximal urethra are located with the aid of a transurethral Foley catheter. The endopelvic connective tissue then is grasped just lateral to the midline on either side with a succession of 3 Allis clamps. The clamps are at the level of the proximal urethra, UVJ, and bladder neck, respectively. The clamps are placed on gentle traction in a direction out toward the surgeon, and sharp dissection is initiated on the medial aspect of the clamped tissue. The dissection must be carried out slowly and meticulously to avoid entry into the urethra or bladder. Bilateral flaps of approximately 2-2.5 cm in depth are created.

Transillumination of the tissue flaps with a small light source helps the surgeon determine that no bladder wall or urethral tissue remains attached. This tissue is strong but can be torn, especially in the direction of the connective tissue fibers. To avoid tearing, the clamps should be handled gently and should not be twisted.

Whether this tissue, once dissected, retains its own neurovascular supply is controversial. Pelvic floor denervation injury and partial recovery have been shown to occur with extensive vaginal dissection, but the clinical consequences of this injury are uncertain. In the experience of surgeons performing the PFSU procedure, no clinical evidence of tissue necrosis or urethral sphincter denervation has been apparent.

Once the sling flaps are created, they are sutured together in an overlapping fashion using permanent 0 or 2.0 polytetrafluoroethylene suture. The resulting double-thickness tissue provides an excellent hammock or backboard for the urethra and bladder neck. Some believe that the overlapping of these flaps may decrease urethral funneling and increase the anatomic urethral length.

After suturing the flaps, reestablishment of a firm connection of the lateral aspect of this connective tissue to the fascial white line is necessary. Next, a vaginal-paravaginal repair at the level of the bladder neck is performed. Usually, 2 sutures per side are required, approximately 1-1.5 cm apart at the level of the proximal urethra and bladder neck. These sutures are placed far laterally in the same connective tissue that makes up the previously described flaps. Bites then are taken at the corresponding location on the white line.

The proximal urethral stitch is placed through the white line just under the pubic bone and the second stitch, approximately 1.5 cm below the first stitch. Some surgeons take a third tissue bite with each stitch through the underside of the vaginal mucosa at the anatomically correct level. This bite pulls the vaginal mucosa up into the lateral sulcus once the suture is tied. This 3-point suturing technique is optional and probably does not add to the efficacy of the procedure.

When the suturing is complete and the strands are tied, the result should be a gentle elevation of the UVJ. If the endopelvic fascial bite is taken too medially, overelevation and anatomic distortion results. Such aggressive elevation also may kink the ureters. Therefore, this author performs cystoscopy in all patients to ensure bladder and urethral integrity and to demonstrate ureteral patency.

After the PFSU is complete, the paravaginal repair can be continued, if needed, with interrupted sutures to approximate the pubovesicocervical connective tissue for the full length of the white line. Midline plication of this tissue also can be performed if such a defect or weakness exists. Each case is individualized to repair each defect as encountered.

Vaginal procedures

In addition to the 2 all-vaginal approaches to the suburethral sling procedure described above, simple suburethral plication of the endopelvic connective tissue has been used as an incontinence procedure in the past. Kelly described this approach, which was the first widely used procedure for stress incontinence. This procedure usually is performed in conjunction with an anterior colporrhaphy for cystocele and is referred to as the Kelly or Kelly-Kennedy plication. The procedure is simple to perform, but the long-term durability of the procedure has been poor in several different studies.

Kelly plication may still have a role in patients with only minor degrees of stress incontinence or in patients with urethral hypermobility and severe anterior compartment prolapse in whom incontinence cannot be demonstrated with pessary placement or prolapse reduction and stress testing. Further research is required to define the role of this procedure in the surgical incontinence armamentarium. Similar procedures have been described by Beck and Nichols, which may provide for better long-term efficacy. The Beck procedure involves taking a suture bite of the periosteum of the inferior pubic ramus on either side of the urethra before tying. Nichols plicates the suburethral tissue in a more aggressive overlapping fashion. Validating studies of the long-term effectiveness of these procedures currently are not available.

Microwave therapy

Microwave therapy is a recent, minimally invasive innovation that results in only subtle alterations in anatomy. The procedure uses microwave energy to heat the connective tissue lateral to the bladder neck and proximal urethra, resulting in shrinkage of the tissue. Both vaginal (through small periurethral incisions) and laparoscopic approaches have been described. It is thought that the contracted posttherapy tissue provides better support to the bladder neck and some extrinsic compression to the proximal urethra.

Efficacy in patients with mild stress incontinence has been good in the short term. Additional studies are needed to gauge long-term efficacy. This procedure may have a role in initial surgical treatment in young patients, even if the long-term efficacy proves to be less than ideal. The procedure appears to cause minimal scarring and alteration in the surgical anatomy; therefore, traditional operations, when performed after failed microwave therapy, may not carry the high complication and failure rates currently observed with repeat incontinence surgery.

Additionally, repeat microwave procedures, if shown to have good short- and medium-term efficacy, may prove to be an acceptable way of managing stress incontinence because of its minimally invasive nature. Finally, this procedure may have a role in medically complicated or frail patients in whom a quick procedure with minimal tissue dissection is desirable. The major drawback of this operation is associated with patient selection. Efficacy in patients with more severe stress incontinence has been poor to date.

Periurethral injection procedures

Periurethral injections are the procedures of choice for patients with ISD and immobility of the bladder neck. Emerging evidence points to efficacy in patients with ISD and bladder neck hypermobility. Various bulking agents have been used, including autologous fat, Teflon (Dupont Co, Wilmington, Del), and silicone microspheres. In 1993, collagen was approved for use as a periurethral bulking agent by the FDA. Since that time, collagen quickly became the material of choice in the United States. In addition to the above criteria, good candidates for periurethral injection should have no evidence of UTI, no significant PVR volume, and a normal bladder capacity.

The periurethral method is described first. Although considered technically more difficult, this method has the advantage of requiring little specialized equipment. Generally, prophylactic antibiotics are administered. The essential steps of periurethral collagen injection are as follows:

• Skin testing for collagen sensitivity must be performed preoperatively. A skin wheal is made with a small amount of collagen. The test is read 28 days after placement, and a negative finding is required before proceeding.
• A small wheal is made with 1% lidocaine, 0.5-1 cm lateral to the urethra on either side using a 30-gauge needle. A urethroscope of 0 degrees is inserted, and any residual urine is drained. A special 20- or 22-gauge needle is advanced parallel to the urethra with a small syringe containing lidocaine stained with indigo carmine. The needle is directed slightly medially as the proximal urethra is approached. Gently wiggling the needle helps determine its location along the urethra. When the area of the proximal urethra and UVJ is reached, injection of the lidocaine and dye solution helps to fine tune the needle placement.
• Blue staining and/or bulging of the mucosa approximately 0.5-1 cm distal to the UVJ confirms the correct needle location. When this location is found, the collagen syringe is placed, and collagen is injected slowly, using firm pressure. The mucosa should be observed bulging into the field of view. If this is not occurring, the needle may be at the UVJ, and the collagen is dissipating around the bladder neck. Alternatively, the needle may be too deep and not in the desired superficial submucosal location. The needle should be repositioned if this is the case.
• If bulging of the mucosa is observed, injection should continue until a coaptation of the urethral mucosa occurs or until no further bulging is observed. A second injection usually is necessary from the other side to obtain adequate closure. On occasion, the collagen travels 360 degrees around the urethra from a single injection site. At a single session, 5-15 mL of collagen may be used.
• The needle and urethroscope are removed. If the patient is awake, a stress test can be performed. The PVR can be determined with US or by passing a small pediatric 8F catheter. If a significant residual volume is noted, ISC should be repeated until residuals are acceptable. Indwelling catheter drainage is not recommended due to the possibility of molding of the collagen around the catheter. ISC may be required for the first 6-24 hours. Prolonged self-catheterization rarely is needed.
• Injections can be repeated at monthly intervals for as many as 3 times if improvement is observed, but results are not yet satisfactory. The need for repeat procedures within the next 6 months to 3 years is very common. The use of collagen injections does not preclude other forms of surgical therapy in the future. Likewise, periurethral injections can be performed after surgical procedures for stress incontinence if the stress incontinence persists after urethral hypermobility is treated. Many prefer the transurethral method because of the relative ease of precise needle localization. The special equipment required is a urethroscope of 0-30 degrees, a 21F operating sheath, a 5F injection catheter, and a beveled 20-gauge needle.
• Under direct visualization with the urethroscope, collagen is injected into the proximal urethra just below the mucosa. Periurethral analgesia is provided as previously described. The endpoints for injection are the same as above. One or more injection sites may be required to produce urethral coaptation. Small amounts of collagen may leak from the injection sites. The leakage is of no consequence.
• To avoid deforming the collagen with either type of approach, the urethroscope should not be passed into the bladder once significant mucosal bulging is observed. In general, the collagen is absorbed gradually over a variable period. This partly explains why long-term results are poor and the need for repeat procedures is great. Nevertheless, patients who are poor surgical risks may benefit from these procedures. New injection materials are being developed with the hope of obtaining more predictable and better long-term results.

Fistula repair

A detailed description of the operative management of urinary tract fistulas is beyond the scope of this article, but several important points are highlighted. A clear understanding of the anatomy and the structures involved in the fistula must be obtained before surgical closure. Involvement of the ureters, bladder neck, trigone, and urethra complicates surgery considerably. On occasion, the uterus and/or cervix can be involved on the genital side of the fistula. Cystoscopy, retrograde urography and IV pyelography are useful in defining the anatomy and in determining which structures are involved. The timing of surgical repair can be controversial. Fistula repair is best accomplished at a time when infection, inflammation, edema, and granulation tissue are absent. A 3- to 6-month waiting period, from the time of injury to the time of repair, has been recommended for simple vesicovaginal fistulas. Serial cystoscopic examinations have been recommended to determine the earliest possible date of closure.

Two common methods are used to repair simple, small (<2-3 cm), vesicovaginal fistulas. The classic method involves excision of the fistula tract with mobilization and closure of the various anatomic layers of the bladder and vaginal wall. The second method, the Latzko partial colpocleisis, involves partial ablation of the fistula by denudation of the vaginal mucosa around the fistula and interposition of layers of the vaginal wall tissue over the fistula.

Successful fistula repair requires adequate dissection and mobilization of tissues, meticulous hemostasis, and reapproximation under no tension. Blood supply to the healing tissues is important. If the tissue is thin, avascular, or irradiated, fresh blood supply must be brought surgically into the field. Omental or bulbocavernosus fat pad flaps are time-honored methods of augmenting vascularity. Surgical approaches to fistula repair must be individualized. Vaginal, abdominal, and combined abdominovaginal approaches each may be appropriate in certain situations as defined by the relevant surgical anatomy. Laparoscopic approaches recently have been described.

Urethral diverticulum repair

Urethral diverticula occasionally can cause urinary incontinence. Classic symptoms include postvoid dribbling of urine, dyspareunia, and recurrent UTIs. The surgical repair of these lesions shares many principles of fistula repair. Adequate mobilization of tissue and meticulous hemostasis are important to avoid complications such as urethrovaginal fistula. Diverticula distal to the urethral sphincter usually can be treated with simple marsupialization (ie, Spence procedure). Diverticula in the mid-to-proximal urethra must be treated with more extensive surgical excision, as outlined below.

• Partial ablation, as described by Tancer, involves incision of the vaginal mucosa over the diverticulum. Next, the periurethral connective tissue over the sac is incised and mobilized as flaps. The diverticular sac then is dissected free and entered. The body of the sac is excised. No attempt is made to excise the neck of the sac or to dissect close to the junction with the urethra. If possible, tissue is closed over the neck in several layers. The periurethral connective tissue flaps are closed in an overlapping fashion. Finally, the mucosa is reapproximated.
• Urethral diverticulectomy is a more extensive procedure in terms of the dissection of the diverticular sac. Some surgeons inflate the diverticulum by way of a double balloon catheter to facilitate identification and dissection and to avoid entry in the sac. The vaginal incision is made. Periurethral connective tissue is incised and mobilized into flaps completely around the diverticulum. Next, the diverticulum is dissected completely, including the neck, but not entered if possible. The sac is excised flush with the urethra. The urethra is closed longitudinally over a catheter. The connective tissue flaps are closed in an overlapping fashion to avoid superimposed suture lines. The vaginal mucosa is sutured to complete the procedure.

With both repair techniques, cure rates of approximately 90% can be expected if meticulous surgical principles are followed. Stress incontinence procedures, such as the suburethral sling, have been combined successfully with surgical diverticulum repair.

Cystoplasty

Augmentation cystoplasty has been used in cases of refractory DI. The basic technique involves bisecting the bladder with the suturing of an opened pedicled section of bowel into the defect. Most commonly, ileum is used, although large bowel and stomach also have been used. Cure rates for DI approach 90% with this technique, but approximately 25% of patients have voiding difficulties, including incomplete emptying. Many of these cases require ISC. Mucus production and stone formation can be problematic for some patients. Malignant transformation of the bowel epithelium has been reported and is a major concern. Recently, detrusor myomectomy has been used to create an intentional bladder diverticulum. Most of the experience with these autoaugmentation procedures has been in neuropathic patients with decreased compliance and/or contracted capacity. Further study is required to determine the role of these procedures in refractory DI.

Denervation procedures

Denervation procedures for refractory DI have been performed for many years. Such procedures include selective sacral neurectomy, the Ingelman-Sundberg procedure, bladder transection, subtrigonal phenol injection, and bladder distention. Selective sacral neurectomy is a procedure performed by neurosurgeons. Electrical stimulation is used to identify which nerve roots increase intravesical pressure. A limited sacral laminectomy is performed, and the selected sacral nerve roots are transected bilaterally. Limited experience with this procedure exists, although reported results have been favorable.

The Ingelman-Sundberg denervation procedure is performed via a vaginal subtrigonal dissection. First, an inferior pelvic nerve block is performed in an attempt to predict a successful outcome. If the injection decreases symptoms, proceeding with surgery is reasonable. A midline vaginal incision or inverted U-shaped incision is made to gain access to the subtrigonal area. Soft tissue beneath the trigone is dissected and excised between 2 clamps. Care must be taken to avoid dissection in the suburethral area to avoid damage to the urethral sphincter. In addition, the lateral attachments of the endopelvic connective tissue at the bladder neck must be preserved to avoid new-onset stress incontinence.

Occasionally, patients may require prolonged catheterization or ISC after this procedure. In addition, injury to the ureters is possible, so confirmation of ureteral functioning before concluding the procedure is important. Some surgeons catheterize the ureters to aid in their recognition intraoperatively. A recent study of a modified version of the Ingelman-Sundberg procedure showed a subjective cure rate of 64% with a mean observation of 14.8 months.

In 1967, Turner Warwick first reported on bladder transection. A full-thickness incision is made through the bladder wall from 1-2 cm lateral to 1 ureteral orifice across the dome and to the opposite ureteral orifice. The bladder is sutured shut immediately using a single layer closure. Procedures involving multiple detrusor myomectomies and circumferential dissection of the bladder (ie, cystolysis) also have been described. These procedures are not in widespread use, and their role in the management of refractory DI is uncertain. Subtrigonal 6% phenol injections have been used to achieve local neurolysis.

Short-term successes have been reported, but long-term efficacy has not been good. Morbidity can be high, and serious complications, such as fistula formation and ureteral injury, have been reported.

Many authorities no longer recommend the procedure. Some researchers believe that this procedure is effective in cases of detrusor hyperreflexia secondary to MS. Bladder distension causes degeneration of unmyelinated nerve fibers. The result can be decreased sensory and motor neurologic function. Success rates can be as high as 60-70% initially but fall off to less than 10% by 6 months.

Implantable sacral neuromodulation devices

Implantable sacral nerve stimulating devices have been used successfully to treat refractory urge incontinence, urgency-frequency syndrome, and voiding dysfunction. In cases of urge incontinence, the rationale behind this therapy is to restore the spinal reflexes responsible for the normal bladder function of urinary storage. The mechanism of action of neuromodulation is understood poorly. The activation of spinal inhibitory pathways is a plausible but probably incomplete explanation.

The commercially produced stimulating device is called the InterStim (Medtronics Inc, Minneapolis, Minn). Appropriate candidates include patients who have failed or cannot tolerate other treatments for urge incontinence and who have demonstrated improvement of symptoms during test stimulation. In addition, patients should have the ability to operate the device and should have no evidence of outlet obstruction. A trial with a test stimulator provided by Medtronics should be conducted. By conducting the test stimulation for 3-7 days, the clinician can determine the integrity of the sacral nerves and the effect of stimulation on the patient's symptoms. In addition, the patient can determine if the resulting sensation is tolerable and acceptable.

The details of surgically implanting the permanent device are not described; however, unilateral implantation is typical. Usually, the third sacral segment is selected. The neurostimulator is implanted subcutaneously in the abdomen. The lead is placed adjacent to the appropriate sacral nerve root in the foramina through an incision in the back overlying the sacrum. The lead is tunneled around the side subcutaneously to connect with the neurostimulator. A handheld wireless patient programmer is used to adjust levels of stimulation as prescribed by the physician. Stimulation parameters can be adjusted noninvasively as needed. The programmer also can be used to turn the device on and off.

In a study of patients with refractory urge incontinence, 47% were dry 6 months after implantation and another 29% experienced a greater than 50% reduction in incontinence episodes. Complications included generator site pain, implant site pain, and lead migration. Approximately one third of the patients required surgical revision to treat or resolve a complication. Approximately 4% required permanent removal of the device because of complications. The infection rate was 2.5%. No permanent nerve injuries were reported. In another small study, 50% of the patients were dry with another 25% greatly improved. Longer-term studies of the efficacy, safety, and patient acceptability of this device are needed.

Artificial urethral sphincter

In 1972, the artificial urethral sphincter (AUS) was introduced for the treatment of severe ISD. Since that time, the construction and design of the device has improved. The current model (American Medical Systems 800) is made of silicone and consists of a pressure-balloon reservoir, a control pump with deactivation button, an inflatable cuff, and connective tubing. The balloon reservoir is designed to accommodate a volume of fluid that can provide for a range of preset pressures.

Required pressures can vary depending on urethral size, location of the cuff, and specific clinical circumstances. Activation of the device is achieved with sustained compression of the pump. In the activated mode, the cuff is depressurized by squeezing the pump and by forcing fluid through a unidirectional valve and into the reservoir balloon. The cuff stays deflated for 3-5 minutes to allow voiding and then automatically reinflates. The deactivation button can be pushed to prevent reinflation. This places the device in deactivated mode.

The AUS can be implanted surgically via the abdominal or vaginal routes. The abdominal route is the only available approach in the male patient. This approach has the advantage of less bacterial contamination, but the dissection around the urethra is difficult, and injury to the urethra and bladder neck may be more common. The site of placement in the male patient is the bulbous urethra, and, in the female patient, it is the bladder neck and proximal urethra. Strict adherence to sterile technique is of utmost importance. The use of surgical hoods and the limiting of traffic in the operating room sometimes are recommended.

The abdominal approach is performed in the dorsal lithotomy position. In the female patient, the vagina is prepped, and some surgeons pack it with iodine-soaked gauze. A 16F transurethral Foley catheter with a 30 mL balloon is placed. The retropubic space is approached most commonly through a transverse skin incision. Muscle-splitting incisions may increase surgical exposure. After the retropubic space is entered, the urethra and bladder neck are located.

Some surgeons place a Babcock clamp around the urethra to provide elevation to facilitate suburethral dissection. The endopelvic connective tissue on either side of the urethra and/or bladder neck is incised. A right angle clamp is used to create a tunnel under the endopelvic connective tissue and urethra. Gentle traction on the Babcock clamp can aid in this dissection. In the female patient, some surgeons use a finger inside the vagina to guide the dissection. Others prefer the vagina to be packed (as described). Caution must be used to avoid dissection in the area of the trigone and ureteral orifices.

Once a suburethral tunnel of approximately 2 cm in width is created, a cuff sizer is passed around the urethra or bladder neck to determine the size of the cuff required. The cuff is tunneled through and snapped into place. The tubing should exit the cuff laterally. The tubing is placed through the layers of the anterior abdominal wall, including the belly of the rectus muscle on one side. The reservoir balloon is placed in the prevesical space on the same side.

The tubing from the balloon to the pump pierces the layers of the abdominal wall (as described) until it reaches a subcutaneous location. The tubing and pump are tunneled subcutaneously into the labia majus or scrotum. Hegar dilators can be used to help create space for the pump. The tubing from the pump to the cuff runs subcutaneously and also pierces the abdominal wall and enters the retropubic space where it connects to the cuff. Once placed, the device is cycled and then deactivated. The device remains deactivated for 6-8 weeks, while healing and encapsulation occur.

The vaginal approach to artificial sphincter placement begins with an inverted horseshoe-shaped flap incision with the apex midway between the urethral meatus and the bladder neck. The vaginal mucosa is dissected completely away from the underlying connective tissue. The retropubic space is entered via sharp or blunt dissection, with the dissecting instrument pointed towards the ipsilateral shoulder. With blunt dissection, the proximal urethra and bladder neck are mobilized off the pubic bone. The previously placed guidance Foley catheter is removed so that the urethra can be sized accurately. The measuring tape is passed around the urethra. The appropriately sized cuff is passed around the urethra and closed shut. The cuff should be oriented so that the tubing is exiting laterally.

Cystoscopy is now performed to check for bladder integrity and the location of the ureteral orifices in relation to the cuff. A transverse suprapubic abdominal incision is made down to the rectus fascia. A small midline incision is made in the fascia. The water balloon pressure reservoir is passed through the incision and into the prevesical space. The attached tubing is brought out through a small puncture wound in the rectus muscle and fascia.

The cuff tubing then is transferred into the retropubic space on the same side through the vaginal incision. The cuff tubing then is brought out through the rectus muscle and fascia near to the area that the balloon tubing pierces these structures. As with the abdominal approach, Hegar dilators are used to create a pocket in the ipsilateral labia majus. This pocket houses the pump. The tube, pump, and reservoir connections are made. The device is tested and deactivated, and the incisions are closed. Postoperative care and device activation are the same as with the abdominal approach.

Urinary diversion

Although they are valuable reconstructive procedures in cases of neoplastic and severe inflammatory diseases of the bladder, urinary diversion procedures are the last resort in incontinence disorders. Instances where these procedures might be used include failed closure of bladder exstrophy and severe, debilitating, refractory urge incontinence. Continent diversions, such as catheterizable pouches and orthotopic bladder substitutions, almost always are possible. The details of these complex procedures are beyond the scope of this article.

Complex reconstructive procedures

Bladder exstrophy is a rare congenital anomaly that requires extensive and complex reconstructive procedures. The reported incidence of this disorder is 1 case per 50,000 population to 3.3 cases per 100,000 population. The male-to-female ratio is 2-3:1. In classic exstrophy, the bladder is everted with a firm, hyperemic, and polypoid mucosa. Chronic inflammation may be followed by metaplasia and malignant change unless early repair occurs. Fibrosis often occurs within the detrusor muscle, resulting in bladder dysfunction and possible upper tract damage after a successful closure. Often, the urethra is short, and the pelvic floor is deficient due to attenuated attachments and pubic diastasis.

Surgical treatment consists of staged reconstruction. The goals of surgical management include urinary continence, preservation of renal function, sexual functionality, and preservation of fertility. Epispadias sometimes accompanies exstrophy or may be found only rarely as an isolated disorder. For defects distal to the urethral sphincter, repair is easy and usually results in continence. Defects that are more proximal and associated with exstrophy are more complicated to manage. In some cases, the urethra may be difficult to locate. Staged repair is described briefly as follows:

• Stage I: Bladder closure with or without pelvic osteotomy is best performed in the neonatal period, ideally in the first 72 hours to prevent mucosal damage.
• Stage II: Bladder neck reconstruction and bladder augmentation and possible diversion generally is performed in children aged approximately 3-5 years.
• Stage III: Genital reconstruction and cosmesis and abdominal wall reconstruction usually is performed at puberty and beyond.

Postoperative Details

Postoperative care is similar for patients undergoing each of the incontinence procedures discussed above, with the exception of periurethral injections, sacral neuromodulator implants, and complex reconstructive procedures. Bladder drainage is an essential aspect of postoperative care. The details have been discussed. Drainage allows the bladder to rest while postoperative edema and inflammation resolve. Most patients are able to void spontaneously in 3-7 days. A few patients require drainage for several weeks. Catheterization rarely is needed longer than 3-4 weeks.

Consider clean ISC at this point. Urethrolysis may be indicated if the patient is unable to void spontaneously after a long period and is unable or unwilling to perform self-catheterization. Straining and large sudden increases in intra-abdominal pressure should be avoided. Take measures to control chronic coughs and to avoid or treat constipation. The patient should avoid lifting anything heavier than 10 pounds for 12 weeks. The long-term success of the repair may depend partly on the patient's lifestyle and activities. Smoking and activities that repetitively stress the pelvic floor may result in long-term failure of the procedure.

Nutrition in the postoperative healing phase also may affect long-term outcomes. Vitamin C especially may be important in collagen formation, and supplementation should be provided liberally. Estrogen may have a positive effect on the healing tissues and should be replaced as indicated. De novo DI can complicate a certain number of procedures, even when performed by the best physicians, and can result in a high degree of patient dissatisfaction if not treated. Most cases respond to conventional treatment, and many cases may be self-limited.

Follow-up

Follow-up care of patients after incontinence surgery consists of surveillance for persistent or recurrent incontinence, voiding dysfunction, and signs or symptoms of pelvic organ prolapse. If postoperative incontinence or voiding dysfunction is identified, complex urodynamic testing is indicated. Patients with foreign bodies, such as synthetic slings and artificial sphincters, should be followed closely for infections, erosion, and rejection. Patients with artificial sphincters should receive follow-up care because of the high incidence of device malfunction. Surgeons should consider referral to a urogynecologic specialist if they do not have experience with complex incontinence issues.

In instances of loss of compliance and increased bladder pressures, monitoring for upper tract damage needs to be considered. Correction of this situation with bladder augmentation or other procedures usually constitutes the best management. In cases of urinary diversion, late complications are not uncommon. Metabolic disturbances, stone formation, mucous obstruction, secondary malignancy, ureteral reflux, ureteral stenosis, and loss of reservoir compliance are some of the more worrisome long-term problems encountered.

Patient Education:

For excellent patient education resources, visit eMedicine's Kidneys and Urinary System Center and Procedures Center. Also, see eMedicine's patient education articles Bladder Control Problems, Understanding Bladder Control Medications, Cystoscopy, Suture Care, and Prolapsed Bladder.

Complications

Surgical and postsurgical complications are an unfortunate reality in incontinence surgery. Even for patients treated by the most experienced and skilled physicians, a certain small number of complications occurs. Recovery can be difficult, and long-term or permanent disability can occur. These potential problems must be taken into account when counseling patients and weighed against the disability caused by the incontinence itself. The patients should be prepared for the possibility of complications, procedure failure, and prolonged catheterization or ISC.

Avoiding complications

The following measures can be taken by the surgeon to minimize the number and severity of these complications:

• Appropriate patient selection
• Management of medical comorbidities, appropriate consultation with specialists, and consultation with anesthesiologists
• Thorough knowledge of the relevant surgical anatomy
• Meticulous hemostasis
• Attention to surgical technique
• Prophylactic antibiotics
• Routine use of intraoperative cystoscopy to ensure bladder and ureteral integrity
• Attention to patient positioning on the operating room table
• Resisting the urge to overcorrect defects
• Appropriate use of bladder drainage
• Knowledge of personal surgical limitations and willingness to seek help if needed

Complications such as intraoperative hemorrhage or visceral injury can be immediate. Other complications, including erosion of sling material or wound infection, can be delayed. Minor and transient injury or severe, permanent, and debilitating problems may exist. Patient prognosis for hemorrhage, urinary tract, and visceral injuries is better if diagnosed and repaired intraoperatively rather than in the postoperative period.

Routine intraoperative cystoscopy detects most of these injuries. Ureteral patency can be demonstrated by observation of the free flow of blue-stained urine from each ureteral orifice following IV administration of indigo carmine dye (see Image 1). One recent review found that 90% of unsuspected bladder injuries and 85% of unsuspected ureteral injuries were detected with routine intraoperative cystoscopy and were managed successfully under the same anesthesia. Sling procedures traditionally have resulted in the highest rates of long-term voiding problems.

Past studies have demonstrated that approximately 8% of patients require long-term or permanent ISC. The incidence of voiding dysfunction after various procedures varies widely and depends partly on the type of procedure but also on technique—most importantly, how tightly the suspension sutures or slings are tied or placed. Increasingly, many recognize that both slings and colposuspension sutures do not need to be tight to be effective. Voiding complications may be on the decline due to this realization. In addition to ISC, postsurgical voiding problems have been managed with varying results with cholinergic agents, alpha-blockers, and intravesical prostaglandin therapy.

Tension-free midurethral sling procedures appear to have lower rates of postoperative urinary retention than traditional bladder neck slings. When retention does occur, several small series have demonstrated the efficacy of midline sectioning of the sling in restoring normal voiding function. Encouragingly, about 80-90% of patients treated in this way still have improvement or cure of stress incontinence symptoms even after division of the sling. Vaginal erosion of synthetic sling material is one of the more common complications of this procedure (1-2%) and is usually treatable by transvaginal partial excision. A more rare and difficult to treat complication is urethral erosion or transection. Intraoperative cystoscopy can eliminate most but not all cases of intravesical sling material. Small breaches of the bladder can be missed and delayed erosion of sling material into the bladder is thought to occur on occasion.

The transobturator approach to tension-free slings may have the advantage of a lower incidence of bladder or urethral injuries. However, procedure-specific complications are being reported, including injury to the obturator neurovascular bundle, leg pain, and thigh abscesses.

Hemorrhage

• Bleeding is usually from the perivesical venous plexus. The obturator vascular bundle or an accessory vessel crossing the Cooper ligament rarely may be injured.
• Endoscopic needle suspensions may have higher rates of hemorrhage requiring surgical exploration than retropubic procedures (5-7% vs 2%).
• Laparoscopic procedures have the possible advantage of decreased blood loss. Hemostasis can be obtained with suture ligation, surgical clips, direct pressure, and absorbable hemostatic agents.
• Closed suction drainage of the retropubic space may be required on occasion to prevent hematoma formation if venous oozing persists.

Urinary tract and visceral injury

• Bladder injuries are the most common. The laparoscopic approach may have the highest rate of injury (10%), especially on the steep beginning of the learning curve.
• The rate of urinary tract injuries with endoscopic and sling procedures is 1-7%. The rate for the Burch procedure is as much as 6%, with most of these being bladder injuries. Urethral injuries may be more common with the MMK procedure, the Kelly-Kennedy suburethral plication, and the artificial sphincter.
• Injury to the ureter usually is a kinking or angulation rather than a transection or ligation. The ureter can be injured in any of the retropubic procedures.
• Blind placement of suprapubic catheters and suture carrying long needles during urethropexy or sling procedures can injure the bowel in rare incidences.
• Adequate distension of the bladder during suprapubic catheter insertion can minimize the risk to bowel.

Urinary tract infection

• UTIs are especially associated with postoperative voiding difficulties with prolonged catheterization. Bacteruria is related to the number of days of catheterization and the type of catheter.
• Bacterial colonization with suprapubic catheters occurs in 17-21% of cases. With transurethral catheters, the rate can be as high as 46-63%. With ISC, the rate is 33%.
• Most cases of colonization do not result in clinical infection, but upper tract infection and urosepsis can occur, especially in elderly and debilitated patients.
• Antibiotic treatment of asymptomatic catheter-associated colonization is not warranted and may result in the selection of more virulent organisms.

Wound infection

• Rates of infection with retropubic procedures, such as the Burch, range from 1-10%. Needle procedures carry approximately a 7% risk.
• Sling procedures have a wound infection rate of 4-6%. Sling rejection and erosion rates have been reported to be as high as 21%. Erosion and other delayed complications of periurethral bolsters that are used in the Stamey procedure can occur even up to 10-12 years after the procedure. Symptoms are usually relieved by transvaginal excision.
• Infection occurs in approximately 9.5% of artificial sphincter procedures. Infections and abscesses have been reported in association with periurethral injections.
• Risk factors for wound infection in incontinence surgery include the use of artificial materials, length of procedure, medical problems (eg, diabetes, steroid-dependent illnesses), and concomitant vaginal hysterectomy.
• Prophylactic antibiotics in the form of a single dose of a broad-spectrum agent are efficacious in vaginal and retropubic procedures.

Osteitis pubis

• Inflammation of the pubic bone related to foreign body placement (eg, suture, bone anchors)
• Usually self-limited, but can complicate MMK (2-4%) and needle procedures
• Rarely complicates Burch procedures and may be related to misplacement of sutures
• The incidence in procedures with bone anchors and screws is unknown.

Osteomyelitis

• Infection of the pubic bone
• More severe and more rare than osteitis pubis

Urogenital fistula

• Rare (<1%)

Nerve injuries

• Common peroneal, sciatic, obturator, femoral, saphenous, and ilioinguinal injuries have been reported in association with incontinence operations.
• Many are related to hyperflexion and external rotation of the hips in the lithotomy position. Compression injury can occur, especially with the use of self-retaining retractors.
• Direct surgical injury can occur rarely, such as an obturator nerve injury during a retropubic urethropexy.
• Ilioinguinal nerve entrapment is a possible complication of needle urethropexy if the needle is passed lateral to the pubic tubercle.

Voiding dysfunction

• This may be manifest as a slow or poor urinary stream.
• The more severe form may result in the inability to void due to relative obstruction and the need for prolonged catheterization or ISC.
• Abnormal uroflowmetry findings preoperatively indicating Valsalva voiding and/or poor detrusor contractility may suggest higher risk.

Detrusor instability

• With cases of mixed incontinence (ie, GSI and DI) managed surgically, the risk of persistence of DI is approximately 40-45%.
• New-onset DI or de novo DI may develop postoperatively and may be due to partial obstruction, dissection-related nerve damage, or a missed preoperative diagnosis.
• De novo DI reportedly complicates 5-27% of incontinence surgery cases.

Genital prolapse

• After the Burch urethropexy, the rate of enterocele formation may be as high as 8%. Some experts have recommended prophylactic culdoplasty to reduce the incidence of postsurgical enterocele.
• Rates of rectocele formation also may be increased.
• These complications are thought to be due to elevation of the anterior vaginal wall and subsequent exposure of the apex and posterior vaginal wall to the more direct action of increases in intra-abdominal pressure.
• With the recognition that colposuspension sutures do not have to be pulled tightly to be effective, the incidence of this complication may be decreasing.

Other complications

• Dyspareunia
• Chronic suprapubic pain
• Sinus tract formation

• Search for CME/CE on This Topic »

• Urinary Incontinence (Emergency Medicine)
• Urinary Incontinence, Nonsurgical Therapies (Urology)
• Urinary Incontinence, Surgical Therapies (Urology)
• Overactive Bladder - Etiology, Diagnosis, and Impact (Urology)

• Bladder Control Problems Causes
• Kidneys and Urinary System Center
• Bladder Control Problems Overview
• Bladder Control Problems Symptoms
• Bladder Control Problems Treatment
• Prolapsed Bladder Overview
• Procedures Center
• Bladder Control Medications Introduction
• Cystoscopy Introduction
• Suture Care Introduction

• Transobturator Midurethral Sling Is an Effective Treatment for Urinary Incontinence
• First RCT Shows Similar Efficacy Among Nonsurgical Treatments for Stress Urinary Incontinence
• Vaginal Tape Procedure Effective for Female Stress Urinary Incontinence

• Stress Urinary Incontinence: Prevention, Management, and Provider Education
• Urinary Incontinence in the Childbearing Woman
• Rating Improvements in Urinary Incontinence: Do Patients and Their Physicians Agree?

• Urinary Incontinence & OAB Resource Center