Ciprofloxacin versus Tobramycin for Neutropenic Fevers

IN RESPONSE:

Drs. Bodey, Rolston, and Raad question the low antibiotic response rate of our study compared with those of previous series. The difference involves study design, not drug efficacy. In our study, addition of any antifungal or antiviral agent was one of the definitions of failure. Patients were not excluded if their febrile episode proved to be viral or fungal. Had addition of nonantibacterial agents been allowed as suggested by consensus panels (1), response rates of 61% to 69% would have been noted. As we mentioned in the Discussion, studies with similar design have had similar responses (2).

Drs. Bodey, Rolston, and Raad are correct that the definition of fevers of unknown origin is unclear. The Methods section should have stated that patients “required antibiotic therapy for fever and/or presumed infection.” They also question the degree of neutropenia between groups. Our Table 1 showed that both the degree of neutropenia at study onset and the duration were similar between groups. Piperacillin was used rather than piperacillin–tazobactam because the latter was not yet on the market.

With regard to the questions of Drs. Linkin and Fishman, our study was not designed to assess combination versus monotherapy in the management of neutropenic fever. The experimental combination regimen was compared with the most “traditional” regimen used to treat neutropenic fever—an antipseudomonal β-lactam and an aminoglycoside (the gold standard). We, like others, feel that monotherapy has a role in the management of selected patients and may be a regimen of choice depending on local patterns of infection, antibiotic susceptibilities, and patient profiles. Although Drs. Linkin and Fishman question whether combination therapy hinders the development of antimicrobial resistance, the Infectious Diseases Society of America expert panel cites the “minimal emergence of drug-resistant strains during treatment” as one of the advantages of combination therapy (1). The investigation cited by Drs. Linkin and Fishman that failed to show a benefit of combination therapy over monotherapy in suppression of resistance involved patients with bloodstream infection due to a single genus (Enterobacter); only 6% of that study sample had neutropenia (3). It is therefore unclear if those results can be extrapolated to neutropenic patient populations. Regarding relative toxicity, it goes without saying that any antimicrobial agent or combination of agents has the potential for toxicity. We agree that the results of available randomized, controlled trials do not show a difference in efficacy between combination therapy and monotherapy. For clinicians who prefer to use a more traditional combination regimen for the management of neutropenic fever, our study provides an alternative: ciprofloxacin–piperacillin.

• Copyright ©2004 by the American College of Physicians