At the moment most antibiotics are initiated without waiting for the microbiological result, if they are thought to be clinically necessary.

Quite right too.

This is so called “empirical therapy”.

There is a good reason for this. Traditionally, microbiology tests have neither been good enough nor fast enough to make the antibiotic prescription dependent on the result.

Take for example the patient who sees his GP with a productive cough and fevers. The GP is not going to say to the patient “Let’s just wait a few days until we get the sputum culture result back. By the way there is a good chance it will be negative even if you do happen to have a pneumonia. And on the flipside, if it does grow something it might just represent the bacteria in your throat.”

No chance…, the GP will simply prescribe an antibiotic based on the most likely pathogens, and also the local antibiotic susceptibility patterns.

Along the same lines, the GP is not going to say to the patient a couple of days later. “Your sputum sample has come back negative, so let’s stop your antibiotic.” Sputum cultures are nowhere near sensitive enough to allow this approach.

On the very odd occasion, the treatment will actually change as a consequence of the microbiology result, if there happens to be an unusual or resistant organism.

And sputa are only one sample type. Ear swabs, peri-anal swabs, & ulcer swabs probably have even less impact on patient management…

In fact for the vast majority of  samples (probably > 95%) that get processed by the microbiology laboratory, the impact on patient management is rather small indeed.

However change is coming on to the horizon. The newer microbiological tests, and in particular the polymerase chain reaction (PCR) assays, are both fast enough and sensitive enough to start genuinely impacting on patient management right at the time of prescribing.

Take the following potential examples:

• Macrolide antibiotics could be witheld in a patient with suspected whooping cough until the Bordetella pertussis PCR is back.
• Patients with urethral discharge and suspected gonorrhoea would only be treated if the Neisseria gonorrhoeae PCR result comes back positive.
• Patients with meningo-encephalitis could have acyclovir to cover HSV, dependent on the CSF viral panel result.
• Legionella cover in a patient with moderate to severe community acquired pneumonia could be dependent on the result of the Legionella PCR in a sputum sample.

These are all tests which, if performed quickly enough, can significantly reduce the amount of antibiotics given to the tested cohort, so they all have potential to play a big part in any antimicrobial stewardship program.

So we need to get such assays into our microbiology laboratories, whatever it takes.

Microbiology matters, but we need to ensure that we utilise new tests and technology to make it matter even more…

Michael