Practice recommendations
- Behavioral therapy reduces urinary accidents in elderly patients with urge, stress, and mixed incontinence.
- Bladder training is helpful for urge incontinence; pelvic floor exercises are helpful for stress incontinence; both are helpful for those with mixed incontinence.
- The effect of drug therapy in the elderly is unclear, as there are only a few studies of sufficient methodological quality. However, drug therapy is less effective than behavioral therapy.
ABSTRACT
Objective: To evaluate the effectiveness of conservative treatment in the community-based elderly (aged ≥55 years) with stress, urge, and mixed urinary incontinence.
Design: Systematic review of before-after studies or randomized controlled trials on the effect of exercise and drug therapy in urinary incontinence.
Main outcomes measured: Reduction of urinary accidents, patient’s perception, cystometric measurement, perineometry, and side effects.
Search strategy: MEDLINE (1966–2001), EMBASE (1986–2001), Science Citation Index (1988–2001), The Cochrane Library, and PiCarta were searched.
Results: Four before-after studies and 4 randomized controlled trials were identified. Drug therapy alone: no study of sufficient quality. Drug therapy compared with behavioral therapy, 3 studies: bladder sphincter biofeedback reduced urinary accidents in cases of urge or mixed incontinence by 80.7%, significantly better than oxybutynin (68.5%) or placebo (39.4%). Adding drug to behavioral treatment or behavioral to drug treatment also resulted in significant reduction in urodynamic urge incontinence (57.5% – 88.5% vs 72.7 – 84.3%). Pelvic floor exercises alone reduced urinary accidents by 48% (compared with 53% for phenylpropanolamine) in patients with mixed or stress incontinence. Behavioral therapy, 5 studies: bladder-sphincter biofeedback in case of urge or mixed incontinence, bladder training in case of urge incontinence and pelvic floor exercises in case of stress incontinence reduced the urinary accidents with 68% to 94%.
Conclusion: There are only a few studies of sufficient methodological quality on the effect of conservative treatment of urinary incontinence in the elderly. Behavioral therapy reduced urinary accidents; the effect of drug therapy is unclear. We recommend behavioral therapy as first choice.
The physiologic goals of treatment are strengthening urethral resistance or reducing detrusor muscle contractions. Behavioral technique—pelvic floor exercises and bladder training with biofeedback—and pharmacotherapy are the treatments of choice for the elderly, provided it is possible to assess the likely health gains. Surgery, the most invasive and riskiest treatment, is usually a last resort.
Methods
The authors performed computerized searches of MEDLINE (1966–2001), EMBASE (1986–2001), the Science Citation Index (1988–2001), the Cochrane Library, and PiCarta. The search was limited to publications in English and Dutch. Search terms were elderly and aged combined with urinary incontinence and conservative management, conservative therapy, conservative treatment, bladder training, drug treatment, pelvic floor muscle training, behavior management, behavior therapy, and biofeedback. We supplemented this search strategy by checking articles referenced in other publications.
The titles and abstracts were then screened for the following inclusion criteria: longitudinal cohort, before-after studies or randomized controlled trials, age ≥55 years, community-dwelling population, and conservative therapy.
Types of incontinence
Stress incontinence is involuntary leakage on effort or exertion, or on sneezing or coughing. Stress incontinence may result from diminished bulk and tone of perineal tissue or weakness of the pelvic floor muscle.
Urge incontinence is involuntary leakage accompanied by or immediately preceded by urgency. Causes are “deconditioned” voiding reflexes due to chronic low-volume voiding, infection, or bladder stones.
Mixed incontinence is involuntary leakage associated with urgency and with exertion, effort, sneezing, or coughing.
The methodological quality of the selected studies was evaluated by a modified Delphi-2 scale. (This scale is available online at www.jfponline.com, as Table W1).10 Two researchers (TT, AJ) scored the studies independently; they were blinded for information on authors and journals. In cases of disagreement, the researchers met to reach consensus.
After meeting inclusion criteria, randomized controlled trials were scored from 0 to 9; before-after studies from 0 to 3. A randomized controlled trial needed a score of at least 7 to be included; a before-after studied needed a 2.5; in trials where blinding was not possible, a 4 was needed.
Results
The search yielded 157 publications; 135 studies did not meet inclusion criteria. Of the 22 remaining studies, 6 were excluded because they did not use a general population. Consequently, 16 studies were included: 6 with a before-after design and 11 randomized controlled trials.
Methodological quality
The quality scores for the 6 before-after studies ranged from 0 to 3. Two studies scored less than 2.5 and were excluded. (Information on excluded studies is available online at www.jfponline.com as Table W2.)
Quality scores for the 11 randomized controlled trials ranged from 0 to 9. Four of the 5 studies with the possibility to blind scored <7, and 3 of the 6 studies with no possibility to blind scored <4; they were excluded.11,18