The pharmacokinetics of amphotericin B administered in a conventional 5% dextrose (glucose) (5% D) solution and in a 20% fat emulsion formulation (Intralipid; 20% IL) were compared in 16 patients (mean age, 42 years [range, 18 to 70 years]) who had been hospitalized for hematological malignancies and with proven or suspected fungal infections. All of the patients received 50 mg (approximately 1 mg/kg of body weight per day) of amphotericin B daily in random order, either as a 50-ml lipid emulsion (20% IL) (group I) or in 500 ml of 5% D (group II). Five serum samples were taken during the 24 h after drug administration, and the levels of amphotericin B were measured by high-pressure liquid chromatography. Serum amphotericin B concentrations declined rapidly during the first 6 h, and subsequent measurements revealed a slow terminal elimination phase in both groups. The maximum serum amphotericin B concentration was significantly lower when the drug was administered in 20% IL (1.46 +/- 0.61 versus 2.83 +/- 1.17 micrograms/ml; P = 0.02). The area under the concentration-time curve from 0 to 24 h was also much lower in group I (17.22 +/- 11.15 versus 28.98 +/- 15.46 micrograms.h/ml). The half-life of the distribution phase was approximately three times longer in group I (2.92 +/- 2.34 h versus 0.64 +/- 0.24 h; P = 0.011). Conversely, the half-lives of the elimination phase were approximately equal in the two groups (11.44 +/- 5.18 versus 15.23 +/- 5.25 h). The mean residence times were also similar in both groups (19.41 +/- 11.13 versus 19.65 +/- 7.86 h). The clearance and the steady-state volume of distribution of amphotericin B in group I were about twice as great as those in group II (62.97 +/- 35.51 versus 33.01 +/- 14.33 ml/kg/h and 1,043.92 +/- 512.10 versus 562.32 +/- 152.05 ml/kg [P = 0.034], respectively). Finally, the volume of distribution in the central compartment was greater in group I than in group II (618.17 +/- 231.80 versus 328.19 +/- 151.71 ml/kg; P = 0.013), but there were no differences in the volume of distribution in the peripheral compartment (425.75 +/- 352.87 versus 234.14 +/- 75.92 ml/kg). These results suggest that amphotericin B has a different pharmacokinetic profile when it is administered in 20% IL than when it is administered in the standard 5% D form and that the main difference is due to a clear-cut difference in the steady-state volume of distribution, especially that in the central compartment.