Key Clinical Issue

What are the effectiveness, benefits, and adverse effects of options for diagnosis and treatment of urinary incontinence (UI) that are available in primary care?

Evidence-Based Answer

UI can be diagnosed in the primary care setting using the history, physical examination, voiding diaries, and validated diagnostic tools; urodynamic evaluation does not improve outcomes of nonsurgical treatments. Nonpharmacologic therapies are effective and safe. Pelvic floor muscle training (PFMT) promotes continence and reduces the severity of stress UI. Bladder training methods reduce the severity of urgency UI, but combining bladder training and PFMT is effective only for patients with mixed UI. Pharmacologic interventions for urgency (and mixed) UI provide some benefits (20 percent or less difference from placebo). Fifty percent of women discontinue drug treatment within one year, typically because of common adverse effects such as dry eye and constipation. The long-term safety of drugs for urgency UI has not been evaluated in clinical trials, but serious adverse effects have been associated with their use among older persons and in combination with other commonly prescribed drugs. Well-validated instruments are available for evaluating treatment effectiveness in terms of symptom relief and change in quality of life. (Strength of Recommendation: A, based on consistent, good-quality patient-oriented evidence.)

Practice Pointers

UI is common in women, affecting about 25 percent of younger women and increasing in frequency with age to around 75 percent of older women.¹ UI can present as stress UI, urgency UI, or mixed UI, which has primary features of both types. Stress UI is caused by sphincter incompetence during periods of increased intra-abdominal pressure (e.g., with coughing, sneezing, or exercising). Urgency UI is caused by detrusor muscle contractions resulting in an overwhelming urge to void and leakage. Overactive bladder refers to a strong urinary urge without UI.² This AHRQ review evaluated the effectiveness of nonsurgical interventions for improving UI.¹

Patient symptom reports alone had poor diagnostic reliability compared with urodynamic testing, which was used as the diagnostic standard in most studies. Diagnostic algorithms compared favorably with urodynamics and are readily available in the primary care setting, are simple to use, and can be quickly implemented.^1,3 Urodynamic testing did not improve nonsurgical therapeutic outcomes and should be reserved for patients considering surgical intervention.¹

Nonpharmacologic interventions are effective with low risk of adverse outcomes. PFMT for stress UI led to continence in nearly 30 percent of women and to improvement in more than 40 percent of women. Combining PFMT with biofeedback, bladder training, intravaginal electrical stimulation, intravaginal devices, or supervised training did not significantly improve results.¹ A Cochrane review found insufficient evidence to determine the best approach to PFMT.⁴

Intravaginal and intraurethral devices were associated with more adverse events than PFMT. Bladder training resulted in improvement in 43 percent of persons with urgency UI. Adding PFMT to bladder training did not improve outcomes. Electrical and magnetic stimulation alone were effective for stress UI. In women who are obese, exercise and weight loss were associated with improved mixed UI.¹ Some invasive nonsurgical treatments for UI, such as botulinum toxin injections,⁵ midurethral slings, and radiofrequency ablation, lack sufficient evidence of benefit or are beyond the scope of this review.¹

All drugs studied for urgency UI were superior to placebo, although the overall effectiveness was low, with less than 20 percent of patients achieving continence. Indirect comparison between drugs suggested that all common UI drugs are similarly effective. In direct comparisons, fesoterodine (Toviaz) was more effective than tolterodine (Detrol), but had a higher discontinuation rate because of adverse effects. Oxybutynin (Ditropan) and tolterodine had similar effectiveness, although tolterodine was associated with fewer adverse effects. Patient characteristics such as age, race, comorbidities, and baseline severity of UI could not be shown to modify the outcomes of any medication.¹

Adverse effects were relatively common and generally increased with larger doses. Dry mouth (10 to 35 percent of patients), constipation (1 to 8 percent), blurred vision, and dry eye were the most common. One-half of women discontinued medication for UI during the first year.¹

Some off-label pharmacotherapies for stress UI were evaluated. Vaginal estrogen preparations were found to be effective for stress UI, whereas transdermal estrogen preparations worsened symptoms. Duloxetine (Cymbalta) was effective in only 7.5 percent of patients, and almost 13 percent of patients discontinued the treatment because of adverse effects.¹

Limited evidence suggests that nonpharmacologic interventions have similar effectiveness and significantly lower the risk of adverse events.¹

These findings indicate that nonpharmacologic therapy should be the initial treatment of choice for all forms of UI. Vaginal estrogens are a good option for stress UI. When opting for pharmacologic intervention for urgency UI, the choice of drug should be guided by adverse effect profile because effectiveness is similar between the drugs.