Review: urinary tract infection in elderly people - time to review management? - health management

Age and Ageing, July, 1995 by R.P. Gray, J. Malone-Lee

Recent studies have challenged the traditional approach to the diagnosis and management of urinary tract infection. Is it time, therefore, to review our management of elderly patients?

Urinary tract infection and asymptomatic bacteriuria are extremely common in the elderly population. Cross-sectional and longitudinal studies indicate that bacteriuria is more common in women, increases with age, is more common in institutionalized or hospitalized patients than in those living at home and may be intermittent [1-5]. The prevalence of bacteriuria in women is about 20% between ages 65 and 75, increasing to between 20 and 50% over the age of 80. The prevalence in men is 3% at age 65-70 and about 20% at ages over 80 years.

The characteristic symptoms of lower urinary tract infection are dysuria, frequency, and suprapubic discomfort. Traditionally, the diagnosis of urinary tract infection is dependent on culturing a microorganism from the urine. However, recent studies have redefined the original criteria and it is no longer considered essential to culture the urine in every case of suspected urinary tract infection. Pyuria and bacteriuria, the two key indicators of urinary tract infection, can be measured by direct microscopic examination of the urine or by newer, more rapid assays using the dipstick method. The correlation of these methods with culture-proven urinary tract infection is discussed below.

Significant bacteriuria

On the basis of the work of Kass, [10.sup.5] colony-forming units (C.F.U.) of a single species per ml in a midstream specimen (MSU) of urine became established as `significant' bacteriuria [6, 7]. While this threshold still holds in asymptomatic patients, in many circumstances a lower threshold should be considered 'significant'. Currently accepted thresholds are [10.sup.2] or more C.F.U. of a single species per ml of a known urinary pathogen in acutely symptomatic women [8, 9], [10.sup.3] or more C. F. U. of a single species per ml of any bacterium in men [10], and [10.sup.2] or more C.F.U. per ml of a single species in a specimen obtained by suprapubic aspiration in symptomatic patients [9]. These criteria have specificities of about 85% and if meticulous techniques are used sensitivities of about 95%.

Asymptomatic bacteriuria implies that significant bacteriuria ([10.sup.5] C.F.U.s per ml) has been detected in two consecutive cultures from a patient without symptoms attributable to the urinary tract. The prevalence of asymptomatic bacteriuria increases with age in both sexes but its significance remains unclear. Other than in pregnancy or in association with instrumentation of the urinary tract, there is no convincing evidence linking asymptomatic bacteriuria to the subsequent development of acute infection. In patients with indwelling catheters, bacteriuria is almost universal, but treating asymptomatic bacteriuria has no effect on the incidence of febrile illness, new episodes of bacteriuria, or the number of bacterial strains identified in urine specimens [11]. An association between hypertension and bacteriuria has been found in cross-sectional studies [12, 13] but no prospective study has confirmed this and there is no evidence that treating bacteriuria will reverse hypertension. Bacteriuria has been associated with increased mortality in some studies [2, 14, 15] but the significance of this association remains unclear. With some exceptions there is little evidence that treating asymptomatic bacteriuria protects against subsequent renal damage. Antibiotic treatment of asymptomatic bacteriuria in pregnancy lowers-the incidence of pyelonephritis from 20-40% to approximately 3% [16]. Since asymptomatic bacteriuria in young children may be a marker for underlying versico-ureteric reflux [17], and may also potentiate existing,reflux or renal damage [18, 19], we and others [20] feel that children under the age of 5 with asymptomatic bacteriuria require treatment and urological investigation. However, controversy exists as there is little direct evidence that treatment protects against infection and renal damage. In most other circumstances asymptomatic bacteriuria is not an indication for treatment and it is highly unlikely that it is related to the non-specific symptoms described by many elderly patients. Bacteriuria is too often used as a convenient scapegoat for the explanation of undiagnosed, nonspecific symptoms of elderly people.

Acute symptomatic urinary infection can cause urinary incontinence but the relationship between lasymptomatic bacteriuria' and urinary in is unclear. A patient may have symptoms referrable to the urinary tract, such as the frequency and urgency of an unstable bladder, and demonstrate significant bacteriuria. However, treatment of the bacteriuria may fail to resolve the symptoms as the relationship is not causal. It is important to be aware of this and avoid repeated unnecessary courses of antibiotics.

Diagnosis

The diagnostic tests for urinary tract infection in elderly people are the same as those used in younger patients.