There seems to be some confusion on how to manage seriously ill patients will the right cover of antibiotics.
From my experience in Japan, I often hear that the patient with sepsis and no definite focus of infection is given a carbapenem antibiotic because of the various routes of coverage e.g. gram positive / negative and anaerobic.
My response is usually less than favourable because it is the last antibiotic that should be used and not the first !
I recently wrote another article on antibiotics but I feel it is necessary to reiterate the point.
Broad spectrum cover can be achieved in many ways with different combinations of antibiotics rather than with the carbapenems. For example, a second / third generation cephalosporin plus metronidazole / clindamycin has good coverage of gram +/- and anaerobes. Another good combination is the amoxicillin-clavulanate or ampicillin-sulbactam combinations which again have good broad spectrum cover. Broad spectrum cover can be enhanced by using gentamicin especially in patients with an undefined cause of sepsis and it is especially good against infections including e.g. endocarditis, pyelonephritis....
The reason I continue to push for other combinations of antibiotics for broad spectrum use is because of RESISTANCE.
From the USA, there have already been reports of bacteria producing Carbepenemases which destroy the carbapenems and if this occurs then these bacteria are resistant to all beta lactam antibiotics including penicillins and cephalosporins.
Just imagine a situation where bacteria cannot be killed by third or fourth generation cephalosporins !!! That does not leave a great choice of antimicrobials to choose from.
The UK and American physicians strongly advocate not to use the carbapenem antibiotics unless other therapies have failed. It is a hidden weapon to coin a phrase. However, if we reach a situation where the carbapenems are being used in place of other antibiotics, the latter which in combination provide a similar coverage, then resistance to this antibiotic will soon occur and then Japan will have major infectious disease problems.
Use of carbapenems should not be first line. Yes, they may be simple to give and reduce the work of the nursing staff, but that should not be the reason for their use. Resistance should always be considered and the use of such antibiotics should normally be restricted to ICU patients where all other treatments have failed.
The source /focus of infection should always be considered on initial presentation and the type of organism considered that could cause the problem. Then, a combination of antimicrobials can be chosen to cover the considered organism(s). Empiric therapy, as it is actually called, is an educated guess, but using combination drugs for broad spectrum cover will treat the vast majority of likely causes. Then as results become available, and narrow spectrum antibiotics can be used, the other antibiotics in the combination which are not necessary can simply be stopped :)
In all fields of medicine, there is always uncertainty whether the treatment is going to be the right one and especially when considering antibiotic coverage. However, uncertainty is something that we as doctors all need to accept and live with. Our medical practices should not be driven by defensive protocols but by clinical need considering history, physical, radiological and laboratory data and the likely differentials diagnoses.
If we reach a situation when all treatments are provided because the medical profession is being defensive, then we cease to be free thinking physicians.