The important question in the long-term care facility is, How does one assess the cause of an episode of acute delirium, especially if there is fever? Such cases are fairly common, yet Lindsay Nicolle suggests that acute changes in clinical status such as fever and delirium should not be attributed to urinary tract infection and that urine samples should not even be obtained. How then are we to tell if a urinary tract infection is indeed causing the delirium? How are we to manage residents with dementia who exhibit delirium with no other obvious cause, particularly if the resident is febrile?

I have seen many residents with dementia who suffer an acute worsening of confusional state, often with behavioural changes and agitation, who show a urine sample positive for white blood cells and are treated with antibiotics and promptly get better. We are now being told that this is strictly coincidence: the antibiotics were not of any help, the urinary tract infection was asymptomatic and there was some other cause for this acute delirium. I do not believe the point has been proven.

I do agree with Walker and colleagues that many elderly residents who have minimal symptoms are inappropriately treated for urinary tract infection. It is clear that a negative dipstick result for white blood cells reasonably excludes the diagnosis of a urinary tract infection but that a positive urine dipstick result or culture is not to be relied upon to diagnose significant infection. Our policy here is to discourage urine cultures and not to look for or treat truly asymptomatic urinary tract infections.