INTRODUCTION This study compared clinical outcomes in patients with acute myelogenous leukaemia (AML) who developed prolonged (≥4 days) febrile neutropenia (FN) and received either empirical or pre-emptive antimould prophylaxis in order to evaluate the need for routine empirical antifungal therapy.

METHODS This retrospective study reviewed adult patients (aged ≥18 years) with AML who developed prolonged FN and received antimould prophylaxis during induction or re-induction chemotherapy at a single centre between September 2016 and December 2020. Patients were categorized into pre-emptive or empirical groups based on whether or not there was clinical evidence of invasive fungal infection (IFI) at the start of antifungal treatment, respectively. Clinical outcomes were compared between the two groups after propensity score matching (PSM).

RESULTS In total, 229 chemotherapy episodes (36 and 193 in the empirical and pre-emptive groups, respectively) were analysed. In the pre-emptive group, broad-spectrum antifungal therapy was administered in 45 (23.3%) episodes. After 1:3 PSM, there were no significant differences between the empirical and pre-emptive groups in terms of the incidence of proven or probable IFI [0/36 (0%) vs 5/97 (5.2%); P=0.323], all-cause mortality [3/36 (8.3%) vs 4/97 (4.1%); P=0.388] and IFI-related mortality [0/36 (0.0%) vs 1/45 (2.2%); P=0.556].

CONCLUSION The differences in clinical outcomes between empirical and pre-emptive antifungal therapy in patients with AML who received antimould prophylaxis were not significant. Therefore, broad-spectrum antifungal therapy in patients receiving antimould prophylaxis may be delayed until there is clear evidence of IFI.