Urinary tract infection in old age: over-diagnosed and over-treated.

A careless and uncritical approach by some clinicians to the diagnosis of urinary tract infection in frail elderly people has resulted in the condition being overdiagnosed and over-treated. Symptoms of lower urinary tract infection are easily recognized and include dysuria, frequency and urgency. But even such typical symptoms require cautious interpretation, as they are common in elderly people without infection [1]. Is laboratory diagnosis by urine culture a more reliable diagnostic tool for urinary tract infection in old age? The isolation of >100 000 colony-forming units of a single species per ml in a mid-stream specimen of urine is regarded as ‘significant’ bacteriuria. This is based on criteria described by Kass [2], although Kass’s studies were performed in healthy young women. There are problems in using these criteria as the standard, since bacteriuria is common in old age, occurring for example in 25–40% of non-catheterized patients in long-stay units [3, 4]. Pyuria is also common in nursing-home residents—both in the presence and absence of bacteriuria—and should not be the sole determinant of the need for antimicrobial treatment [5]. The presence of bacteria in a urine sample does not help to distinguish between true infection and colonization or contamination. The chances of detecting coincidental asymptomatic bacteriuria in a frail elderly patient are high [6, 7]. A familiar scenario begins when a frail elderly patient is non-specifically unwell, and nurses ask if a urinary tract infection might be responsible. No other symptoms or signs of systemic infection are present but a urine specimen is sent for culture. The microbiology laboratory reports the presence of bacteria in the urine, and the ward doctor (perhaps without assessing the patient) prescribes an antibiotic. This leads to overuse of antimicrobials, provides opportunities for the colonizing organism to acquire antimicrobial resistance and exposes the patient to unnecessary antibiotic side-effects [8–11]. Awareness of this situation has led to the suggestion that urine culture in older patients be abandoned, but this remains one of the most commonly requested bacteriological investigations in elderly hospital patients [12]. Another common event is when ‘routine’ urinalysis is carried out for elderly patients in the absence of clinical symptoms suggestive of infection, and a positive stick test results in a urine sample being sent for culture. This misguided practice is perpetuated by the notion that urinary infection can occur in an asymptomatic older patient, and ignores the fact that asymptomatic bacteriuria is not an indication for treatment [13]. While dipstick testing and urine clarity can predict which urine specimens are likely to be culture-negative, the use of these procedures should be highly selective and confined to symptomatic patients [14].