An animal model of disseminated aspergillosis was used to test the in-vivo activity of itraconazole against four isolates of Aspergillus fumigatus. Two reference isolates of A. fumigatus known to be resistant to itraconazole in vitro and in vivo were used as control isolates, and two new isolates were tested under the same conditions. For each isolate MICs for itraconazole and amphotericin B were determined by an NCCLS-based method. Mice infected intravenously were treated either with itraconazole 100 mg/ kg/day or amphotericin B 4.5 mg/kg/day for 10 days. Amphotericin B showed good in-vivo activity against all four isolates. For one strain, which had a low in-vitro MIC for itraconazole, in-vivo therapy with itraconazole prolonged the survival of mice and reduced fungal burdens in organs compared with untreated controls. In mice infected with a strain with a high MIC of >16 mg/L, itraconazole neither prolonged survival nor reduced fungal load in organs compared with controls. It is concluded that there is a relationship between MIC and treatment outcome in mice for A. fumigatus infection.