Empirical Antifungal Therapy for Persistent Fever in Patients with Neutropenia

Open Access
Publisher Website
Google Scholar
Abstract
Sir—I wish to comment on the section of the recently published “Practice Guidelines for the Treatment of Candidiasis” [1] that discussed the use of empirical antifungal therapy for persistently febrile neutropenic patients. While I fully agree that available evidence justifies the use of empirical antifungal therapy for patients with antibiotic-refractory fever and neutropenia, I was surprised to note that fluconazole was not given any consideration for use as a possible therapeutic agent in addition to amphotericin B. It is well known that yeasts (in particular Candida species) are the common fungal pathogens during the early phase of neutropenia and that molds (usually Aspergillus species) are encountered beyond the second week of neutropenia. If so, why should we not consider using fluconazole, a safe agent with good anticandidal but not antiaspergillus activity, as empirical therapy in patients at low risk for aspergillosis? To my knowledge, there are at least 4 studies (3 published studies and 1 abstract) [2–5] that show that fluconazole has efficacy equal to that of amphotericin B for the empirical treatment of persistent fever during neutropenia. Viscoli et al. [4] excluded patients at high risk for aspergillosis namely those who had an abnormal finding on chest radiography, those with surveillance culture results that were positive for Aspergillus species, or those with a prior history of aspergillosis—in their randomized trial of fluconazole and amphotericin B with patients who had neutropenia with persistent fever that was unresponsive to antibacterial therapy. The response rates were 75% for fluconazole and 66% for amphotericin B, with a significant difference in the frequency of adverse effects (32% vs. 82%). In a similar study by Winston et al. [3], nearly 70% of patients in both the fluconazole and amphotericin B study groups had a good outcome, whereas the frequency of adverse effects was less among those in the fluconazole group (13% vs. 81%). Regarding the dose, the guidelines state that amphotericin B, 0.5–0.7 mg/kg/d, is a reasonable dosage treatment of possible yeast or mold infection in the empirical setting. Such low dosages of amphotericin B are suboptimal for empirical or definitive therapy for aspergillosis. For example, in the study by Winston et al. [3], in which empirical amphotericin B was given in a dosage of 0.5 mg/kg/d, fungal infections occurred with equal frequency in recipients of fluconazole and amphotericin B; mold infections were seen in 4 fluconazole and 3 amphotericin B recipients. Mortality resulting from fungal disease was similar in both study groups. Although invasive aspergillosis is a difficult entity to treat using any dosage of amphotericin B in this patient population, higher dosages (∼1 mg/kg/d) are generally preferred for the treatment of suspected aspergillosis.