Empirical antifungal therapy in patients with neutropenia and persistent or recurrent fever of unknown origin.

Persistent or recurrent fever of unexplained origin (PFUO) in neutropenic patients receiving antibiotic therapy is commonly treated with empirical antifungal therapy (EAFT). EAFT was established as an adequate management of PFUO around 20 years ago with conventional amphotericin B deoxycholate (c-AmB), despite its high rate of infusional and systemic toxicities. In recent years, EAFT trials for PFUO have used less toxic agents, such as the lipid formulations of AmB, the new azoles, and the echinocandin, caspofungin. In clinical trials, the lipid formulations of AmB [especially liposomal AmB (L-AmB)] provided similar efficacy with lower toxicity but at a much higher cost. Although rarely used in clinical practice, fluconazole is equivalent to c-AmB, provided patients at high risk of Aspergillus infections are excluded. Intravenous itraconazole was shown to be equivalent to c-AmB, with a lower toxicity. Voriconazole did not meet non-inferiority criteria when compared with L-AmB. Caspofungin was shown to be non-inferior to L-AmB and more effective in treating baseline invasive fungal infections. To date, alternatives to AmB have shown less toxicity, but improved efficacy is less clear. This is probably because of the weakness of the indication and to the consequent difficulty in establishing objective and reproducible endpoints for comparisons. The new challenge for physicians in this field is probably presumptive antifungal therapy, an approach based on patient risk-group stratification for developing invasive candidiasis or aspergillosis and/or the use of new diagnostic techniques to identify patients at a very early stage of infection.