Tag Archives: antibiotic reporting

“Reporting susceptibilities on UTIs, not urinary isolates…”

Antimicrobial Resistance, Antimicrobial Stewardship, , , ,

Urines arrive at diagnostic microbiology laboratories in considerable numbers. My own lab in New Zealand processes a couple of thousand urines a week. A significant proportion of these will have positive cultures. Therefore, the potential for the laboratory to promote good antimicrobial stewardship with respect to urinary tract infection is considerable.

My mantra on this is as follows: “The microbiology lab should never release antibiotic susceptibilities on a positive culture from a urine sample unless there is reasonable evidence accompanying the request that the patient has a UTI.”

The fact that the urine sample has turned up at the microbiology lab is insufficient evidence per se that the patient has a UTI. Urines get sent to microbiology laboratories for all sorts of spurious reasons, see below for a few examples:

  • Urines often get sent “automatically” from acute receiving wards as part of a blanket laboratory screen, where the patient may have a diverse spectrum of symptoms such as chest pain, shortness of breath, collapse, etc.
  • Urines can get sent from Long Term Care Facilities when someone decides to dipstick all their patients and send the urine samples with positive dipsticks to the lab for culture. Yes, it happens, and a lot more often than you might think!
  • Urine from indwelling catheters can get sent when the patient has a blocked catheter, or the catheter bag is cloudy.
  • Urines from patients attending outpatient clinics should also raise a flag. With the exception of urology clinics, patients who attend a pre-planned elective clinic appointment generally do not have an acute UTI. The same principle can apply for patients who are in hospital wards for other reasons.
  • Urines where the clinicians are looking for other tests, i.e. albumin/creatinine ratio, and due to laboratory processes the urine ends up getting cultured as well…

So, my argument is that if a urine sample turns up at the laboratory without any clinical details or with inappropriate clinical details, the lab is under no obligation whatsoever to release antibiotic susceptibilities on any organisms grown. 

The best approach of course is not to process the sample at all unless relevant clinical details are received. I would regard all of the following clinical details as being unacceptable to justify proceeding to urine culture:

  • No clinical details
  • Cloudy urine
  • Concentrated urine
  • Dark urine
  • Smelly urine
  • Urine dipstick urinalysis results only
  • Routine/monitoring/screening urine
  • Fatigue
  • Increased CRP
  • Lots of other non-specific symptoms!

The easy option for the lab of course is just to accept the sample, report the organisms, and the accompanying susceptibilities. However, this is almost certainly not the best way…

Michael

“Keeping it simple or keeping it accurate…”

Antimicrobial Resistance, The Science of Microbiology, , ,

Let’s say you are a clinician and you are looking at two different microbiology results on two different patients (A&B). Both have an E. coli UTI. Both results state that the E. coli is susceptible to trimethoprim. However what you don’t know is that the E. coli isolate on Patient A had a trimethoprim disc diffusion zone of 18mm (right on the EUCAST breakpoint), whilst for patient B the corresponding zone was a much more comfortable 26 mm.

And who knows, if you repeated the same testing on patient A a dozen times, the chances are you would have a few “non-susceptible”results, due to the natural margin of error of the test.

If I was a clinician, and had this extra (zone diameter) information, I would be a lot happier prescribing trimethoprim to patient B, even though they both have in-vitro “susceptibility” reported on the result. (The same principle of course applies if we were talking about Minimum Inhibitory Concentration (MIC) values instead of zone diameters.)

But do clinicians really want this extra information?

They are usually very busy, …and not particularly interested in microbiology.

In my experience all clinicians generally want to know is if an isolate is susceptible or resistant. They are not particularly interested in the details, with the exception of blood culture and sterile site isolates, when there is at least a modicum of interest in the degree of susceptibility or resistance.

So which is better.. a susceptibility result full of information, but potentially difficult to understand and interpret, or a result reduced to its simplest form.

There is no right answer of course…

I am not even convinced antimicrobial susceptibility breakpoints have a long term future.

More and more, year by year, the anti-microbial susceptibility committees (EUCAST, CLSI) are trying to take into account antibiotic dose,  renal function, degree of infection, etc. when setting antimicrobial breakpoints.

But they are really only scratching the surface…

Time for some major disruption!

Michael

“The Whole Picture”

The Art of Microbiology,

For the clinical microbiologists and ID physicians….

It might be the most susceptible antibiotic on the report but the child will only take syrup and not tablets.

It might be the most susceptible antibiotic on the report but there is little chance that this patient will take antibiotics four times a day.

It might be the most susceptible antibiotic on the report but it tastes horrible. Is this patient likely to stomach it?

It might be the most susceptible antibiotic on the report but it causes diarrhoea in a good proportion of patients. Will this patient “run” with it?

It might be the most susceptible antibiotic on the report but this patient swears by another antibiotic.

These are only a few of the things that should be considered when individualising treatment. Advising on the best antibiotic is not just about looking at the report and looking for the most susceptible antibiotic available. It is about “What antibiotic is most likely to work best in this particular patient?”. Getting good at this sort of decision making takes time and experience and a good deal of background knowledge, not all of which can be learned from textbooks. We certainly don’t always get these decisions right.

Michael