Is there a rationale for the use of antimicrobial prophylaxis in neutropenic patients?

Antimicrobial prophylaxis in neutropenic patients has been practised in one form or another for several decades but the goal is no longer clear. From being initially solely an attempt at decontamination, drugs such as co-trimoxazole and later the fluoroquinolones were preferred to non-absorbable regimens because they achieve reliable protection against bacteraemia due to Gram-negative bacilli. Nevertheless, fever still invariably occurs during neutropenia leading to the initiation of traditional empirical therapy. Not only is this approach illogical but it also ignores the flexibility afforded the oral and parenteral formulations of the fluoroquinolones. Instead, it might be as effective and less costly if these agents were given orally until the end of neutropenia unless there was evidence of malabsorption or poor oral intake, in which case treatment would be continued parenterally. Should patients develop fever, an attempt would be made to complement treatment with another anti-microbial agent for microbiologically or clinically defined infection. This would be carried out at diagnosis, before any changes in the prophylactic regimen could be made. Otherwise, treatment with the prophylactic regimen would continue without modification. There is a less compelling need for prophylaxis against candidosis, herpes simplex and cytomegalovirus disease as these would be better managed pre-emptively when there is evidence of yeast carriage or re-activation of viral infection. Similarly, prophylaxis of aspergillosis is a forlorn hope and again a pre-emptive approach might serve us better once there is a screening test available and a safe and effective drug.