Empiric Antifungal Therapy with Amphotericin B in the Era of Fluconazole Prophylaxis: a Cohort Study in Adults with Acute Myeloid Leukemia Treated within An Institutional Antifungal Policy.

To determine whether fluconazole prophylaxis was effective in decreasing the need for parenteral empiric antifungal therapy in patients with acute myeloid leukemia (AML) and persistent febrile neutropenia or suspected fungal infection at our center.

Prophylaxis with fluconazole in patients with severe chemotherapy-related neutropenia has been found to be beneficial in decreasing the need for parenteral antifungal therapy, and preventing superficial and invasive fungal infections and fungal infection-related mortality (Bow et al., Cancer 2002;94:3230-3246).

The records of all patients at our hospital who presented with AML from January 1999 to July 2009 were reviewed retrospectively. As of September 2005 we adopted an institutional antifungal policy consisting of routine antifungal prophylaxis with fluconazole followed by amphotericin B as the first line parenteral agent in the event of persistent fever despite broad spectrum antibiotics or suspected fungal infection. The policy included criteria for switching from amphotericin B to a second line agent (caspofungin) for continued empiric therapy or another agent depending on clinical or laboratory data or suspicion of a particular pathogen. Explicit criteria were also developed for switching to a second line agent including baseline renal function or change in renal function while receiving amphotericin B or other adverse effects such as significant infusion reactions or electrolyte disturbances. Fluconazole was given at a dose of 400 mg daily starting with induction chemotherapy and continued until blood count recovery or switch to parenteral antifungal agent.

We identified a total of 170 patients with a median age of 61 years (range 18-89 years), 53 % were female and the median follow-up time was 187 days (range 2-2549 days). Baseline cytogenetics grouped patients into poor risk (40%), standard risk (39%) and favorable risk (10%) categories, with 11% unknown or inconclusive. Two-thirds of patients had de-novo AML. Twenty-four percent of patients did not receive induction chemotherapy and were treated with best supportive care, leaving 130 patients who received induction chemotherapy. Overall median survival for chemotherapy treated patients was 409 days, compared with 44 days for patients treated with best supportive care. The majority of patients (77%) who received chemotherapy were treated with standard induction consisting of 3 days of an anthracycline and 7-10 days of continuous infusion cytarabine. Of the patients treated with induction chemotherapy, 65% received prophylaxis with fluconazole and 32% did not, the remainder received prophylaxis with other antifungal agents. The use of prophylactic fluconazole coincided with implementation of our antifungal policy. Of patients who were treated with fluconazole prophylaxis, 62% required parenteral antifungal therapy and 38% did not. Of patients who did not receive fluconazole prophylaxis 56% required parenteral antifungals and 44% did not. These differences relating to receiving fluconazole prophylaxis were not statistically significantly different. For those patients requiring empiric antifungal therapy, they received a median of 18 days of fluconazole (range 3-156 days). Of the 56 patients who were treated with amphotericin B as empiric therapy, 59% were changed to another agent due to renal effects (42%), fever (27%) or other adverse effects (21%). Switching off amphotericin B occurred after a median of 7.5 days (range 0-59 days). Fifty-six percent of patients received caspofungin as the second line agent while the policy was in effect.

Based on our retrospective analysis of the practical use of antifungal prophylaxis within our institutional antifungal policy, fluconazole prophylaxis did not decrease the need for empiric parenteral antifungal therapy. The majority of patients treated with empiric amphotericin B were switched to a second line agent, mostly due to intolerance or adverse effects.

No relevant conflicts of interest to declare.