I wonder how often the following question is asked on hospital wards: “[patient] doesn’t look right today. I’ve dipped her/his urine and it’s got everything in it. Do you want to start antibiotics now or wait for the culture result first?”

 

Comprehensive geriatric assessment (GCA) is known to be effective in the evaluation of older people but takes time and skill to do well.

Wouldn’t it be great if there was a rapid bedside test that could be performed by people with only minimal training and could be reliably used as a surrogate for this process?

This would be especially true for the frail who can be challenging to assess due to multiple co-morbidities, including cognitive impairment, and atypical disease presentations.

One possible solution is the urinalysis dipstick test.

 

The diagnosis of urinary tract infection (UTI) in older people is trickier than many imagine.

Most diagnostic schemes rely on the presence of classical urinary tract symptoms (e.g. dysuria, frequency, urgency, haematuria and costovertebral tenderness), but these are present in just half of people aged over 75 with bacteraemic UTI and less than 10 per cent of care home residents with advanced dementia and suspected UTI.(1,2)

So, how do all the other people with UTI present? The answer is the geriatric giants (i.e. falls, confusion, incontinence and/or loss of mobility). So how can we distinguish these presentations caused by a UTI from all the other possible causes (i.e. every other illness)?

Indeed, UTI is commonly blamed for non-specific illnesses in older people even though this is not supported by sound evidence and is often overtly wrong.(3,4)

If we knew there were bacteria in the urine then this might help - a role for urine dipsticks?

 

Although dipsticks are available in a variety of forms to detect a wide range of substances in the urine, when used for old people in hospitals it is mostly for the diagnostic evaluation of UTI.

Leukocyte esterase (LE) is a surrogate marker for the presence of white blood cells and nitrites are present when nitrates in the urine are metabolised by some bacteria (e.g. E. coli but not pseudomonas).

Typically urinalysis results are considered suggestive of bacteriuria if either nitrites or LE are positive and not suggestive if both nitrites and LE are negative.

​

Unfortunately in older people it’s seldom so simple. Asymptomatic bacteriuria (ASB) is the presence of bacteria in the bladder urine with no attributable symptoms.

It is common in older people (10-20 per cent in the community), especially the most frail (up to 40-50 per cent of care home residents) and universal in those with long­term catheters.(3,5,6)

ASB is frequently misinterpreted as symptomatic in the presence of non-specific symptoms, e.g. subtle changes in functional or cognitive status (‘not their usual self’ or ‘foul-smelling urine’).(7)

 

Urinalysis cannot be used to distinguish ASB from UTI. Additional problems with relying on urine testing are difficulty obtaining a sample (e.g. in the presence of cognitive impairment or incontinence) or unreliable results (e.g. contamination, catheter samples or the prior receipt of antibiotics).

 

Putting these concerns aside, just how reliable is urinalysis to detect bacteriuria?

 

False positive results

 

Due to both the lack of specificity of the test and the high rate of ASB, a large proportion of older people will have a positive urinalysis result irrespective of whether or notthey have a UTI.

In studies of older people in hospitals and nursing homes, 34-97 per cent of people tested had a positive urinalysis result (LE or nitrites) but only 17-57 per cent actually had a positive urine culture.(3,8,9)

So, in each study approximately 50 per cent of people with a positive urinalysis had a positive urine culture.

 

Or to put it another way, a positive urinalysis result is about as accurate as flipping a coin to detect the presence of bacteriuria in frail older people.

 

False negative results

 

Some have suggested that urinalysis is useful to exclude UTI if both LE and nitrites are negative.(8,10)0

However, given that many pathogenic organisms can’t produce nitrites it is an inherently risky approach that depends on LE.

The majority of studies have shown false negative rates of 6 to 30%.(3,10-15)

As urinalysis is inaccurate for bacteriuria and can’t distinguish between UTI and ASB, it can’t be used as a replacement for clinical assessment of the patient.

So after your assessment, if you aren’t thinking the diagnosis is UTI but the urinalysis is positive then you should ignore the result, and if you are thinking the diagnosis is UTI but the urinalysis is negative then you should ignore the result.

Or you could say that the urinalysis result is irrelevant to your clinical assessment and is therefore pointless.