Our objectives were to assess methods to diagnose urinary incontinence (UI) and monitor treatment effectiveness in community-dwelling adult women, and to assess clinical efficacy and comparative effectiveness of pharmacological and nonsurgical treatments for UI.
We searched major electronic bibliographic databases, the FDA (Food and Drug Administration) reviews, trial registries, and research grant databases up to December 30, 2011.
A systematic review of diagnostic studies and therapeutic randomized and nonrandomized studies published in English was performed to synthesize diagnostic accuracy; minimally clinically important differences in validated tools for diagnosing UI; and rates of continence, improvements in UI, and harms of examined treatments. We calculated pooled absolute risk differences to estimate the number needed to treat (NNT) to achieve continence or avoid harms with random effects models.
From a total of 905 eligible references, 99 studies showed minimal diagnostic value of tests to distinguish urodynamic stress or urgency UI; 57 studies suggested specific ranges of improvement in UI frequency (based on voiding diaries) that women considered important, as well as the value of quality-of-life assessment with validated checklists or scales. Pretreatment urodynamic diagnoses were not associated with better predictions of nonsurgical treatment outcomes. Continence was achieved in one woman with urgency UI for every eight women treated with fesoterodine (NNT 8, 95 percent CI [confidence interval], 5 to 17), 12 with tolterodine (NNT=12, 95 percent CI, 8 to 25), nine with oxybutynin (NNT=9, 95 percent CI, 6 to 16), nine with solifenacin (NNT=9, 95 percent CI, 6 to 17), and nine with trospium (NNT=9, 95 percent CI, 7 to 12). Discontinuation of treatment due to adverse effects occurred in one woman for every 33 treated with fesoterodine (NNT=33, 95 percent CI, 18 to 102), 16 with oxybutynin (NNT=16, 95 percent CI, 8 to 86), 56 with trospium (NNT=56, 95 percent CI, 30 to 228), and 78 with solifenacin (NNT=78, 95 percent CI, 39 to 823). Discontinuation due to adverse effects occurred more often with fesoterodine or oxybutynin than with tolterodine. Continence was achieved in one woman for every three treated with pelvic floor muscle training (NNT=3, 95 percent CI, 2 to 5), six with pelvic floor muscle training combined with bladder training (NNT=6, 95 percent CI, 4 to 16), and six with intravaginal electrical stimulations (NNT=6, 95 percent CI, 4 to 16). Weight loss improved UI in obese women. Improvement in UI and quality of life were examined using different definitions, which hampered the synthesis of evidence. Evidence was insufficient from which to conclude prediction of treatment effects by age, race, baseline severity of UI, and comorbidities.
Clinical evaluation with validated tools for diagnosis of UI, its type, frequency, severity, and impact on quality of life informs nonsurgical treatment decisions. Women determine treatment satisfaction and success according to clinically important reductions in UI frequency as recorded in voiding diaries and with clinically important improvements on condition-specific quality-of-life scales. Benefits from pelvic floor muscle training, bladder training, and electrical stimulation are large, and adverse effects are uncommon. Benefits from drugs are small. Drugs for urgency UI have comparable effectiveness. Evidence about long-term adherence to and safety of all available treatments is insufficient.
Shamliyan T, Wyman JF, Ramakrishnan R, et al. Systematic review: benefits and harms of pharmacologic treatment for urinary incontinence in women. Ann Intern Med 2012 Jun 19.