Effectiveness of systemic antifungal prophylaxis in patients with neutropenia after chemotherapy: a meta-analysis of randomized controlled trials

Abstract

Background: Invasive fungal infections are responsible for substantial morbidity and mortality among patients with neutropenia after chemotherapy.

Objective: This meta-analysis was conducted to estimate the effect of systemic antifungal prophylaxis on the frequency of invasive fungal infections in patients with neutropenia and the mortality associated with these infections.

Methods: MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials were searched through September 15, 2010, for English-language publications of randomized controlled trials that compared systemic antifungal prophylaxis with placebo or no treatment in patients with neutropenia. Studies involving nonsystemic antifungal agents, direct comparisons of antifungal agents, and empirical, preemptive, or salvage treatment were excluded. Data were extracted according to European Organization for Research and Treatment of Cancer definitions of proven, possible, or probable fungal infections. Effect measures were reported as odds ratios (ORs) for fungal infection and mortality. Outcomes were pooled using random-effects meta-analysis. Study quality was assessed based on the study setting (single center or multicenter), institutional review board approval/informed consent, randomization, description of allocation concealment, double blinding, and reporting of dropouts.

Results: Twenty-six trials including a total of 3979 patients were included in the meta-analysis. Antifungal prophylaxis was associated with significant reductions in proven fungal infections (OR = 0.43; 95% CI, 0.31-0.60; number needed to treat [NNT] = 20) and mortality attributed to fungal infections (OR = 0.49; 95% CI, 0.30-0.80; NNT = 53). Overall mortality was not affected by antifungal prophylaxis (OR = 0.92; 95% CI, 0.74-1.14). Antifungal prophylaxis was associated with a reduction in risk for proven Candida infections (OR = 0.28; 95% CI, 0.20-0.38), but not for aspergillosis or zygomycosis. Antifungal prophylaxis was also associated with a decreased need for antifungal therapy (OR = 0.64; 95% CI, 0.48-0.86). In an explanatory subgroup analysis of major outcomes, only recipients of a hematopoietic stem cell transplant (HSCT) had a significant likelihood of benefiting from antifungal prophylaxis in terms of reductions in risk for proven infections (OR = 0.27; 95% CI, 0.16-0.44) and mortality attributed to infections (OR = 0.41; 95% CI, 0.21-0.81). A metaregression analysis indicated that a multicenter (vs single-center) design and a double-blind (vs unblinded) design were significant moderators of the effect of antifungal prophylaxis on overall mortality and proven systemic fungal infections, respectively.

Conclusions: Systemic antifungal prophylaxis was associated with a reduction in proven fungal infections and mortality attributed to these infections in patients with neutropenia after chemotherapy. In the setting of HSCT, antifungal prophylaxis was associated with reductions in both proven infections and mortality attributed to infections. There was no significant association between antifungal prophylaxis and overall mortality.