Catherine E. DuBeau MD
ESSENTIALS OF DIAGNOSIS
The majority of persons remain continent into advanced old age, contradicting the myth that urinary incontinence (UI) is a normal consequence of aging. However, UI increases in prevalence with age. In older women, the prevalence of UI is 15–30% in the community, 50% among the homebound, and ≥ 50% in nursing home residents. In men, the prevalence is about one third that of women until age 85, when the ratio becomes 1:1.
Besides age, risk factors for UI in women are pregnancy and childbirth, pulmonary disease (because of associated cough), hysterectomy, obesity, other lower urinary tract (LUT) symptoms, neurological disorders (eg, delirium, stroke, Parkinson's disease, spinal cord injury), diabetes, and functional and cognitive impairment. In men, additional risk factors include the presence of other LUT symptoms and prostatectomy (risk greater with radical prostatectomy than with transurethral resection). Dementia is associated with UI, but the strongest correlate is impaired mobility, not cognition.
Complicating the recognition and treatment of UI is the failure of many health care providers to ask their patients about leakage and LUT symptoms, coupled with the fact that at least 50% of persons with UI never report it to a health care provider.
Fultz NH, Herzog AR: Epidemiology of urinary symptoms in the geriatric population. Urol Clin North Am 1996;23:1. [PMID: 8677528]
Thom D: Variation in estimates of urinary incontinence prevalence in the community: effects of differences in definition, population characteristics, and study type. J Am Geriatr Soc 1998;46:473. [PMID: 9560071]
Continence depends not only on LUT function but also on intact mobility, cognition, motivation, and manual dexterity. Even persons with a normally functioning LUT may experience UI if factors such as impaired mobility or cognitive impairment are present. Therefore, UI in most older persons is multifactorial and requires evaluation and treatment focused beyond the LUT alone.
The chief precipitants of UI in older persons are age-associated changes in the LUT, comorbid disease, and medications.
Incontinence can be divided into 4 major types:
urge incontinence occurs in frail elderly persons who have detrusor hyperactivity with impaired contractility (DHIC), urge leakage with a high postvoid residual (without urethral obstruction).
Table 24-1. Medications associated with incontinence.
Table 24-2. Mnemonic for causesof transient incontinence.
In older women, symptoms of both urge and stress incontinence (mixed incontinence) are common. The term overactive bladder (OAB) denotes a syndrome that includes urgency, frequency, and nocturia with or without urge incontinence. In older persons, however, urgency and frequency may be caused by many factors besides the bladder. Thus, care should be taken not to attribute OAB symptoms initially to the LUT.
Ouslander JG et al: Overactive bladder: special considerations in the geriatric population. Am J Manag Care 2000;6(Suppl): S599. [PMID: 11183903]
There are no evidenced-based approaches to UI prevention. Avoidance and treatment of risk factors and transient
UI-associated factors may be helpful. Some experts recommend general strategies such as not forestalling voiding for long periods; avoiding diuretic beverages, artificial sweeteners, and high fluid intake; and controlling constipation.
Fonda D et al: Prevention of urinary incontinence in older people. Br J Urol 1998;82(Suppl 1):5. [PMID: not available]
Table 24-3 lists common UI and LUT symptoms with their utility in differential diagnosis. Note that UI is not associated with pelvic pain. Other causes (especially neoplastic) should be sought.
Patients need a full physical examination because of the multifactorial nature of UI in older persons. The examination should include cardiovascular, abdominal, and neurological systems as well as assessment of mobility and cognition.
In all patients, perineal innervation should be evaluated by checking resting and volitional anal sphincter tone and perineal sensation. The integrity of sacral roots S2-S4 (the level of the sacral micturition center) are evaluated by the anal wink and bulbocavernosus reflexes. The anal wink is done by lightly scratching (eg, with a tongue depressor) about 1 in. away from the anus, which should contract (“wink”); the reflex should be tested on the right and left sides of the anus. The same “wink” should be seen with the bulbocavernosus reflex, using the stimulus of a light squeeze of either the clitoris or glans penis. False-negative responses can occur if the patient is not completely relaxed. If unsure whether the anus visibly contracts, check by inserting a finger in the rectum and palpating for the reflexes. Pelvic floor prolapse in women is neither sensitive nor specific for UI but may impact treatment. Prolapse is best evaluated using the lower blade of a speculum to first support the posterior wall and then the anterior vaginal wall while the patient strains. Signs of pelvic muscle laxity are forward movement of the urethra (urethral hypermobility), prolapse of the anterior vaginal wall into the vagina, introitus, or beyond (cystocele), and prolapse of posterior vaginal wall (rectocele).
Bimanual exam should be done to check for pelvic masses. Rectal exam should check for masses and, in men, prostate nodules. Estimation of prostate size by digital exam is inexact, even among experienced clinicians.
Table 24-3. Symptoms & signs of UI.
No specific laboratory findings are associated with UI. In general, renal function and urinalysis (for hematuria and, in diabetics, glycosuria) should be checked. If no recent data are available, one should also check for diabetes mellitus and vitamin B12 deficiency. UI should not be attributed to pyuria and bacteriuria unless UI is very recent in onset or associated with fever, dysuria, elevated white blood cell count, or otherwise unexplained inanition. In most cases, the patient will have asymptomatic bacteriuria that does not require treatment.
Women with stress symptoms should have a stress test. To increase sensitivity, this is best done when the bladder is full with the patient in the standing position. After checking that her perineum is relaxed, the patient should be asked to give a single, vigorous cough. Immediate leakage indicates stress incontinence; delayed leakage (often of large volume and difficult to stop) or leakage that occurs after several coughs may be stress- induced urge UI (a triggered detrusor contraction).
A postvoiding residual (PVR) urine test is performed using either catheterization or ultrasonography. However, obtaining a PVR is often not possible or is impractical in primary care settings. A PVR is strongly suggested for the frail elderly (because of possible DHIC); women with a large cystocele, which can obstruct the urethra; patients on medications that suppress detrusor contractility; patients with neurological disease (other than dementia); those with previous pelvic surgery or radiation; patients who have failed empiric therapy; and men with urge UI who will be treated with antimuscarinic agents.
Urodynamic studies are not necessary in the initial evaluation of most older persons. Urodynamics should be considered for women with stress UI who desire surgery; men with elevated PVR who desire prostate surgery; patients with complex neurological disease (especially Parkinson's disease and spinal cord injury); and persons who have failed empiric therapy. This testing includes evaluation of urine flow rate, cystometric pressure with filling and voiding, assessment of urethral function, and a pressure-flow study to evaluate obstruction and detrusor contractility. The best quality testing is done with medium-rate fluid filling of the bladder and simultaneous measure of abdominal pressure; carbon dioxide cystometry is neither sensitive nor specific.
Few patients require special imaging. A renal ultrasonogram is often performed for men with a large PVR, especially if they have impaired renal function. There is no consensus on the definition of large PVR. In the setting of renal insufficiency, even lower volumes (eg, 100-200 mL) should prompt further evaluation.
Evaluation of UI in frail, institutionalized elders should be tailored to their overall functional and cognitive status and goals of care. At the same time, one should recognize the potential for treatment benefit in persons with reversible precipitants (see Table 24-1) and in those with intact mobility despite impaired cognition. Although the Minimum Data Set includes a resident assessment protocol for UI that is completed by nursing staff, full evaluation of UI is the responsibility of the patient's clinician. Stress testing is less feasible and sensitive in this population; it should still be considered for women who are suitable candidates for surgical repair or who have pulmonary conditions or take medications that could precipitate stress leakage. Guidelines for evaluation of UI in these settings are available from the American Medical Directors Association and the National Association for Continence.
Fantl JA et al: Urinary incontinence in adults: acute and chronic management I (clinical practice guideline no. 2, AHCPR Publication No. 96-0682). Public Health Service, Agency for Health Care Policy and Research, 1996.
UI can be the presenting symptom of serious underlying diseases. However, patient care goals will determine the extent of evaluation. Important conditions to consider are abdominal and pelvic malignancies (especially in the setting of hematuria or pelvic pain and sudden onset of UI); bladder stones (with recurrent urinary tract infections [UTIs] and pelvic pain); spinal cord injuries (when the neurological exam is abnormal beyond known comorbidity); and fistulas (with vaginal leakage of urine or stool). Urinary retention with overflow incontinence should be suspected in men with underlying prostate disease and in women with a large cystocele, prior pelvic surgery, or irradiation. Although uncommon beyond the sixth decade, women with nonmalignant pelvic pain, dysuria, and frequent small voids may have interstitial cystitis and should be referred for evaluation.
Morbidity associated with UI includes falls (and attendant fractures), skin infections, and pressure ulcers.
Most significant is its impact on many domains of quality of life, including psychological distress (decreased self-esteem, worry about coping strategies), impaired social interactions (at work, church, leisure time, and with intimate relationships), and limitations of activities.
Naughton MJ, Wyman JF: Quality of life in geriatric patients with lower urinary tract dysfunction. Am J Med Sci 1997; 314: 219. [PMID: 9332259]
The model for UI treatment mirrors that of many chronic diseases: a stepped approach over time, starting with treatment of precipitating or aggravating factors (eg, comorbid disease, immobility, medications) followed by lifestyle changes, behavioral therapies, medications, and finally surgery. Combining treatments, especially behavioral and medications, works better than either alone. Older persons should have treatment individualized to their overall care goals, most bothersome aspect of UI, and desired outcomes. Table 24-4 presents treatments and evidence-based efficacy. Many of these methods work for several types of UI.
Modification of the volume and types of fluid intake can have a large impact on leakage and frequency. Patients should aim for a 24-h output of ~2 L (adjusted for body size), avoid diuretic drinks with caffeine (coffee, tea, colas), and if nocturia is a problem—curtail late afternoon and evening intake.
Bladder retraining, prompted voiding, and habit training reduce urge and stress UI by keeping bladder volume low by regular voiding. Bladder retraining is performed with mobile, cognitively intact persons. In addition to timed voiding, patients are instructed to inhibit urgency by sitting or standing still, relaxing, and pelvic muscle contracting. They should continue to the bathroom when the urge has subsided. For frail patients, including those with impaired mobility prompted voiding (taking the patient to the toilet with praise reinforcement) and habit training (simply taking the patient to the toilet on a schedule) are effective. None have side effects. Prompted voiding and habit training depend on caregivers and, therefore, can have high personal and labor costs.
Pelvic muscle exercises (PME) strengthen the muscles that support the urethra and augment its closure. Often used for stress UI, PME may help with urge leakage as well. Similar to other muscle-strengthening regimens, PME are based on low repetitions of high-intensity contractions held as long as possible. A starting regimen could be 3 sets of 10 contractions (with adequate relaxation between each contraction) repeated 2-3 times per week. As patients progress, they increase the intensity and duration of each contraction. Keys to PME success are correct identification of the target muscles and motivation to continue the program. Biofeedback and weighted vaginal cones (held in the vagina while the patient is upright) can be added to help patients learn PME.
Electrical stimulation is an alternative for patients who have difficulty localizing or contracting muscles for PME. A tubular sensor with electrodes is placed in the vagina or rectum, and a low electrical current causes rhythmic pelvic muscle contractions.
Antimuscarinic agents that decrease parasympathetic activation of the detrusor are used for urge UI. The agents with well-established efficacy are oxybutynin and tolterodine. Both are available in immediate-release (IR) and extended-release (ER) formulations; doses are oxybutynin 2.5-5 mg 2-4 times daily (maximum, 20 mg/day), oxybutynin-ER 5-30 mg once daily, tolterodine 1-2 mg twice daily, and tolterodine-ER 2- 4 mg once daily. ER formulations have similar efficacy but fewer side effects than IR formulations. A topical oxybutynin patch is now available as well (Oxytrol 3.9-mg patch every 3 days). Troublesome anticholinergic side effects in older persons are dry mouth (40% occurrence with IR forms, < 20% with ER forms and a risk factor for caries), constipation, blurred vision, acute narrow-angle glaucoma, worsening of reflux, and confusion. Although the absolute number of patients studied are relatively small, oxybutynin-ER has better efficacy and tolterodine-ER better tolerability. Lack of response to one agent does not preclude response to the other.
Other agents used for treatment of urge incontinence are imipramine, hyocyamine, probanthine, and flavoxate; these have no established efficacy in older patients and are best avoided. The antidiuretic desmopressin is sometimes used to treat nocturia but must be used with caution in older persons because of the high risk of fluid retention, congestive failure, and hyponatremia.
Estrogen (oral and topical) previously was widely used in postmenopausal women with stress and urge UI. Oral estrogen with progesterone has been found to worsen UI; the data on topical estrogen are conflicting and scant (especially in older women). Estrogen is effective in reducing recurrent UTIs in women. Topical estrogen is available as a cream (Premarin), ring insert (Estring), and dissolving tablet (Vagifem).
Currently, there are no drugs available to treat stress incontinence. Duloxetine, a serotonin and norepinephrine
uptake inhibitor that may be released in the near future, has good to moderate efficacy in middle-aged women, although initial nausea may decrease compliance.
Table 24-4. Evidence based efficacy of urinary incontinence (UI) treatment.
Numerous operations are used to treat stress UI. In general, these operations use a variety of approaches to resuspend the muscles, ligaments, and fascia that support the urethra and anterior vaginal wall. Data on long-term outcomes (beyond 5 years) and surgery in older-old women are scant. As with other surgical procedures, outcomes from more recent techniques (eg, tension-free vaginal tape) are based on case series or limited randomized trials of selected patients in tertiary centers and should be interpreted with caution.
Pelvic organ prolapse, causing urethral obstruction with overflow UI or exacerbating stress UI, may respond to a pessary. Numerous models are available. Ring and Gelhorn pessaries often are easiest to fit in older women and need to be changed only monthly. Cube pessaries require daily changing. Many older women with prolapse are difficult to fit because of a shortened vagina.
Drainage catheters cause morbidity, resulting in universal bacteriuria by 30 days and increasing the rates of (1) infections with resistant organisms, (2) chronic pyelonephritis by 3 mo, and (3) urethral meatal damage. They should be reserved for patients with sacral or lower extremity wounds, those with chronic retention not amenable to other treatment, and those for whom palliative measures are too uncomfortable or disruptive (eg, end of life).
Absorbant garments and pads should be used only when all other methods have failed or when incontinence persists despite adequate, appropriate treatment. They are costly for patients, especially over time. Products range from penile shields for men to numerous varieties of pads and undergarments for both men and
women. Patient advocacy organizations have information to help patients pick the product most appropriate for their type of leakage and lifestyle.
Abrams P et al (eds): Incontinence, 2nd edition. Second International Consultation on Incontinence. Health Publications Ltd, 2002.
The Cochrane Library: Update Software, 2002.
The majority of persons with UI will improve with treatment. Virtually no long-term outcomes data (> 5 years) exist. Whether UI severity increases over time is not clear. The finding of involuntary detrusor contractions in healthy continent older persons raises the possibility that detrusor overactivity may progress over time. However, other reasons for UI to worsen are failure of compensatory mechanisms (eg, urethral sphincter function in women), increasing comorbidity, and multiple medications.
RELEVANT WORLD WIDE WEB SITES
International Continence Society: http://www.continet.org (Includes links to other continence organizations and resources.)
American Urological Association: http://www.auanet.org
American Foundation for Urologic Disease: http://www.afud.org
American Urogynecologic Association: http://www.augs.org
American Medical Directors Association: http://www.amda.com) (UI treatment in long-term care.)
National Association for Continence: http://www.nafc.org
Simon Foundation for Continence: http://www.simonfoundation.org