Choosing optimal antifungal agents to prevent fungal infections in nonneutropenic critically ill patients: Trial sequential analysis, network meta-analysis, and pharmacoeconomic analysis

Abstract

The use of antifungal interventions in critically ill patients prior to invasive fungal infection (IFI) being microbiologically confirmed and the preferred drug are still controversial. A systematic literature search was performed to identify randomized controlled trials (RCTs) that compared untargeted antifungal treatments applied to nonneutropenic critically ill patients. The primary outcomes were all-cause mortality and proven IFI rates. A random-effects model was used with trial sequential analyses (TSA), a network meta-analysis (NMA) was conducted to obtain indirect evidence, and a cost-effectiveness analysis using a decision-analytic model was completed from the patient perspective over a lifetime horizon. In total, 19 RCTs involving 2,556 patients (7 interventions) were included. Untargeted antifungal treatment did not significantly decrease the incidence of all-cause mortality (odds ratio [OR] 0.89, 95% confidence interval [95%CI] 0.70 to 1.14), but it did reduce the incidence of proven IFI (OR 0.45, 95%CI 0.29 to 0.71) relative to placebo/no intervention. The TSA showed that there was sufficient evidence supporting these findings. In the NMA, the only significant difference found for both primary outcomes was between fluconazole and placebo/no intervention in preventing proven IFI (OR 0.35, 95%CI 0.19 to 0.65). Based on drug and hospital costs in China, the incremental cost-effectiveness ratios per life-year saved for fluconazole, caspofungin, and micafungin relative to placebo/no intervention corresponded to US$889, US$9,994, and US$10,351, respectively. Untargeted antifungal treatment significantly reduced proven IFI rates in nonneutropenic critically ill patients but with no mortality benefits relative to placebo/no intervention. Among the well-tolerated antifun-gals, fluconazole remains the only one that is effective for IFI prevention and significantly cheaper than echinocandins.