OBJECTIVES: There are no pharmacoeconomic studies perform to date comparing voriconazole and liposomal Amphotericin B (LAB) for the treatment of systemic fungal infections. The aim of this study was to perform an economic evaluation of voriconazole versus LAB for the treatment of invasive aspergillosis and candidiasis. METHODS: A cost-minimization analysis was performed from the hospital perspective in 2005, as the same efficacy was assumed. A systematic review of available literature was performed between 1996 and 2005, in order to obtain the efficacy and incidence of drug-related adverse events (AE) for each treatment group. Duration of treatment (intravenous: 15.42 days; oral: 4.49 days; the same for both treatments) and mean weight of patients (68.6 Kg) were obtained from a local study: The Fungcost study (Peiro S. Value in Health 2002; 5:564). Only direct costs per episode were considered; medications (iv and oral) at their hospital selling prices; the cost of monitoring AE; and administration costs (obtained from a national cost database). Voriconazol was the oral treatment in both groups. The most important AE for each treatment (and the way to monitor them) were: hepatotoxicity with voriconazole -13,97%- (two chemistries and hematologic tests during the treatment period); and nefrotoxicity with LAB -12,84%- (a daily creatinine clearance measurement). Mean cost per episode and incremental cost were calculated. RESULTS: Mean cost per episode was €6,073.43 (iv treatment 94.22%) for voriconazole, and €8,794.33 (iv: 95.78%) for LAB in the treatment of invasive aspergillosis, with an incremental cost of €2,486.90 (28.28%) The treatment of candidiasis showed a mean cost of €6,307.43 (iv: 94.22%) and €8,779.92 (95.93%), respectively, with an incremental cost of €2,472.48 (28.16%). Results were robust to the sensitivity analysis. CONCLUSION: Using costs and treatment patterns of fungal infections in Spain, voriconazole is more cost-effective than LAB for the treatment of invasive candidiasis and aspergillosis.