Bow, E.J., Laverdiere, M., Lussier, N., Rotstein, C., Cheang, M.S. & Ioannou, S. (2002). Antifungal prophylaxis for severely neutropenic chemotherapy recipients: A meta analysis of randomized-controlled clinical trials. Cancer, 94, 3230–3246.

The evaluated treatment was antifungal prophylaxis with azoles (fluconazole, itraconazole, ketoconazole, and miconazole) or an amphotericin B formulation compared with placebo or no prophylaxis controls.

The search used MEDLINE and EMBASE (1966–2000); additional studies were identified from bibliographies/reference lists of articles, topical reviews, and information from the pharmaceutical industry and investigators in the field.

38 randomized, controlled trials

7,014 patients who received cytotoxic therapy for acute leukemia or hematopoietic stem cell transplantation (HSCT) sufficient to result in neutropenia (an absolute neutrophil count [ANC] of less than 1,000) lasting one week or more.

In severely neutropenic patients (ANC less than 1,000 for a week or more), antifungal prophylaxis reduced the use of:

• Parenteral antifungal therapy by 43% (prophylaxis success).
• Superficial fungal infection by 71%.
• Invasive fungal infection by 56%.
• Fungal infection-related mortality by 42%.

In subgroup analyses, superficial fungal infections were not reduced for:

• HSCT recipients overall.
• Patients receiving low-dose amphotericin B formulations.
• Patients in a single, small miconazole trial.

However, superficial fungal infections were reduced in HSCT recipients on azoles.

In subgroup analyses, fluconazole was more effective than itraconazole or low-dose amphotericin B formulations to prevent superficial fungal infections.

In subgroup analyses, a reduction in fungal infection-related mortality was not observed in:

• Pediatric trials.
• Non-HSCT trials.
• Trials comparing azoles with polyene controls.
• Trials comparing low-dose amphotericin B formulations with placebo.
• Trials with itraconzaole, ketoconazole, or miconazole.

There was a reduction in fungal infection-related mortality in trials using fluconazole for antifungal prophylaxis.
Antifungal prophylaxis did not affect:

• Overall mortality, except in subsets of HSCT recipients or patients in which the mean duration of neutropenia was longer than two weeks.
• The incidence of aspergillosis, perhaps because the overall incidence was low (1%) in both groups; therefore, a treatment effect could not be detected.