Wednesday, 22 August 2007


Today, I would like to discuss about antibiotic choice.

I come from a country where antibiotics are restricted for certain conditions and hence, it is not possible to use certain antibiotics unless there is an agreement from the pharmacy department but especially from the hospital Microbiologist.

The Microbiologist in the UK is a Doctor who trains in medical microbiology and it is they who decide on specialist use of antibiotics not the junior doctors.

For example, if a seriously ill patient is admitted and a resistant organism has been identified, then the doctors will either consult to the Microbiologist directly or they will be contacted by the Microbiologist.

The Microbiologist will then decide on what antibiotics are most appropriate for the condition taking into account local resistance, cost and benefit plus side effect profile etc...

For example, if a junior doctor was to commence a carbapenem antibiotic for a condition that could be treated by an antibiotic of a different class, then it would be stopped by the Microbiologist thereby overriding the junior doctor.

The purpose of this practise of restricting antibiotic use is to prevent resistance. Resistance is an ever increasing problem and those antibiotics which have the least resistance should be used last not first !! If you use these antibiotics then resistance will soon occur to these, leading to a situation of having nothing to fight the bacterial enemy, a situation similar to the pre-antibiotic era.

As doctors, we have a responsibility to look after the FUTURE and not to just think about the now.

A resident I met at another hospital said 'I have to think about the patient in front of me now'. I make no criticism of that comment at all but to me it shows a flaw in the intrinsic understanding of antibiotic knowledge. I always hear, that we should start with a broad spectrum antibiotic and then narrow to a specific one. This is indeed a reasonable way to look at things, but one has to consider the cause of the infection in the first place, to consider the likely organisms involved and from there choose antibiotics that will provide appropriate cover. This is my understanding of using broad spectrum cover. To just inject the 'naypalm' type antibiotics of the carbapenem group with no idea of the cause is irresponsible with the excuse that it is 'broad spectrum'.

As an example, I heard of a patient with a stroke at another hospital who developed a fever and upper respiratory infection. The doctors had no idea of the cause of the infection and hence, they started a carbepenem antibiotic. This was clearly a wrong decision. When one considers the likely cause of the infection, one has to think of aspiration pneumonia e.g. anaerobes, oral streptococci in addition to the usual causes of infection such as penumococcus, haemophilus influenza etc... Of course, nosocomial infections should be considered but this depends on the timing of when the patient develops the infection in hospital. Hence, pseudomonas and MRSA should also be of consideration.

I would have commenced such a patient on a regimen of pencillin with a penicillinase antagonist or a cephalosporin plus anaerobic coverage with a licosamide e.g. ABPC/Sulbactam. In the UK I would have probably used amoxicillin/clavulanate or a 2nd generation cephalosporin plus metronidazole which would provide cover for most of the infections considered above.

Yes, it is always possible to consider the rarest of bacteria with the potential resistance and that every patient has MRSA etc, but common things are common. Pneumococcus causes most pneumonias, stroke patients suffer with aspiration....MRSA pneumonia is in fact, quite uncommon. MRSA tends to cause line infections in hospital and of course, soft tissue infections e.g. cellulitis / infetced ulcers where of course a different antibiotic would be considered in that case.

Hence, the use of the 'broad spectrum' carbapenem shows that the actual cause of the infection was not appreciated, because if it was, then antibiotics within a selective area would have been considered instead.

I also hear that the worse the condition of the patient, the stronger the use of the antibiotic. This again is not a logical statement. When one thinks of infection, one has to consider the bacteria but also the Immune reaction of the Patient. It is a two way process. The same bacteria in one patient may cause just a minor infection whereas in another it may produce septic shock. The bacteria is the same and will respond the same to the identical antibiotic. So, why should a stronger antibiotic be used????

One has to remember that bacteria produce Exo- and Endo-toxins that activate the immune system in various ways and result in cytokine release and it is the cytokines that produce manifestations such as septic shock. It is our own immune system reactivity that results in the severity albeit driven by the presence of the bacterium.

Hence, one needs to consider the sensitivity of the bacteria to killing by an antibiotic, the volume of distribution of the drug, its solubility, renal/hepatic clearance from the circulation etc, and not just how severe the patient is. Remember, the severity is due to the immune system and cytokine release!

For example, I would be more than confident to use benzylpenicillin in a patient with a severe pneumococcal pneumonia with a septic syndrome rather than using a third generation cephalosporin. Even if there is resistance (PRSP) Benzylpenicllin can still be used effectively, albeit at high dose. In the UK, we commonly start Benzylpenicillin with a daily total dose of over 14 mega units per day.

Only if treatment fails to work by a failure of clinical improvement and / or if culture results reveal resistance that categorically show that the current antibiotic regimen will not work, does one then switch to a stronger regimen of antibiotic cover.

The severity of a patient needs to be always be looked at separately as such patients may require a CV Line for fluid management, large quantities of fluid and potentially, catecholamine support until the septic syndrome abates.

Hence, I hope that you can in some way appreciate that my training has come from within a system of where there are limited use of 'strong' antibiotics and patients are generally treated with penicillin e.g. amoxicillin, anti-staph penicillin (flucloxacillin), first and second generation cephalosporins, macrolides, aminoglycosides and fluoroquinolones, and patients still get better from their infections !!! :)

The UK Microbiologists routinely reserve the use of Vancomycin, Piperacillin, Third Generation cephalosporins and especially the Carbapenems. They must be consulted first . If such a system of microbiologist / ID specialist consult existed in all major hospitals, and that there was consistent guidance to all Japanese doctors in hospitals and in the community, I feel sure that resistance to current antibiotics would slow down although it will never be possible to extinguish such resistance.

Please do not abuse the use of antibiotics and consider carefully the cause of the underlying problem and choose your antibiotics according to this. There is no harm in using more than one antibiotic to cover different organisms in different groups so that a broad spectrum cover is then possible.

Do not use the last antibiotics that are left without considering the vast armory of other antibiotics that will still do the same job.

Please look at the following Adobe PDF files below that are NHS Guidance from a Scottish Hospital in Fife produced in 2006 showing that for common infections that require hospitalisation, simple antibiotics are still very much used. Please also check out the use of antibiotics in the community setting. Note that no where will you see carbapenems being used as first line treatment except in ICU patient who have failed on other treatments.

For a more detailed analysis and perspective on antibiotics please read the books by Dr Makoto Aoki.

No comments: