What’s stopping antibiotic de-escalation therapy?

Peter West

Wyeth Europa

Everyone seems to agree that antibiotic de-escalation therapy seems to make sense. The big question is therefore why aren’t we doing it?

Antibiotic de-escalation therapy is the practice of using more powerful antibiotics, earlier in treatment, for a short period of time – and then switching to a less powerful antibiotic once the infection is accurately diagnosed and under control. One could argue that the concept of “hit hard – hit fast” is not new. Paul Ehrlich described something similar to this with his “magic bullet” theory at the turn of the last century.

Over recent years the momentum appears to have grown and studies have demonstrated that early intervention with appropriate antibiotic therapy saves lives and reduces morbidity.

At a recent international congress I was amongst an audience of clinicians where the chairman was seeking individual views on de-escalation. The responses were consistent to the point of being unanimous – de-escalation makes good sense and should be practiced.

But frequently, it isn’t. So why is this the case?

I think that part of the conundrum is the disease process itself, which frequently is one of escalation. A patient presenting on a ward in hospital may not initially have a “serious” infection. They may be admitted for something quite different, like a broken limb, and have no infection at all. But within a matter of days this situation can change and, in particular for the more elderly complicated patients with underlying diseases/co-morbidities, the situation can deteriorate very rapidly.


 “…de-escalation makes good sense and should be practiced.”


For example: It would seem unnecessary, foolish even, to commence therapy for a mild chest infection with a very powerful antibiotic when the clinical signs and symptoms do not warrant this. But in some patients what appears to be a mild infection can progress to a more serious clinical situation.

Diagnostic techniques in microbiology have still not advanced to the stage where early detection of a causative pathogen can be easily made. It remains the case that a 48 hour period will lapse before microbiological results can be obtained. So what happens then?

The results may reveal not one, but several bacteria, are present. This then becomes a sort of bacterial “whodunit”? Are we witnessing a genuine polymicrobial infection or are some of those bacteria simply colonizing and not infecting the patient? Which one is the real culprit and how should we target therapy to deal with it? In some cases the microbiological results reveal…nothing!

So the situation is complex.

Consideration of the host (patient history, site of infection), the environment (the level of bacterial resistance in that particular unit) and the suitability of available therapy (efficacy and tolerability of the antibiotics at ones disposal) all play a major part in selecting the most appropriate therapy. Even with microbiological results we may still be dealing with an empiric situation.

In some cases the patient will have already received antibiotic therapy but may not be responding clinically. Studies again suggest that in such a situation it is imperative to select an antibiotic agent from a different class. But how well is this understood or practiced?

Initial therapy with beta-lactam antibiotics, such as penicillin based compounds or cephalosporin is still commonplace in many units. If initial therapy with these agents is failing then the evidence demands a change of antibiotic class. But frequently, in deteriorating clinical situations, clinicians, in pursuit of de-escalation, may turn to carbapenem type agents. However, carbapenems are also beta-lactam antibiotics and so this would appear to go against the tide of evidence.


“…the increase of carbapenemases is a worrying development and resistant strains of klebsiella are becoming a major concern across Europe.”


In recent years we have also witnessed an increasing and disturbing level of resistance to some carbapenem type agents – the increase of carbapenemases is a worrying development and resistant strains of klebsiella are becoming a major concern across Europe. Concerns about increases in carbapenem type resistance are only heightened by the availability of new carbapenem type agents and the possibility of the development of cross-resistance.

The pipeline of new antibacterials in development has diminished over recent years – particularly those agents with activity against gram negative bacteria. The regulatory environment is becoming more demanding, and regulators would prefer to see clear evidence of “superiority” over existing agents, rather than “non-inferiority”. But this is time-consuming and costly to achieve. It may not even be possible.

The question arises whether a new agent, albeit only with evidence of non-inferiority rather than superiority, is still a valuable asset in terms of reducing selection pressure on existing agents? Most clinicians would feel that this is a valid argument.

All of this amounts to a very difficult decision making process and it is perhaps not surprising that de-escalation is easy to agree on, but harder to deliver in practice.

So where do we go from here?

About the author:

Peter West is Senior Marketing Director, Infectious Diseases at Wyeth Europa, based in the UK.

For enquiries, please email westp2@wyeth.com