Disseminated fungal infection causes significant morbidity and mortality in children undergoing hematopoietic stem cell transplantation (HSCT). The widespread use of prophylactic oral triazoles has limitations of poor absorption, interindividual variability in metabolism, and hepatic toxicity. AmBisome (amphotericin B liposomal complex) has a better safety profile than the parent drug amphotericin B and produces higher plasma and tissue concentrations. We hypothesized that once-weekly high-dose AmBisome therapy could provide adequate fungal prophylaxis for immunocompromised children undergoing HSCT. We performed a pharmacokinetic pilot study to determine whether once-weekly high-dose AmBisome administration would result in effective concentrations throughout the dosing interval. A total of 14 children (median age, 3 years, 1 month; range, 4.5 months-9 years, 9 months) undergoing HSCT received once-weekly intravenous AmBisome prophylaxis (10 mg/kg as a 2-hour infusion). Blood samples for pharmacokinetic measurements were drawn around the first and the fourth weekly doses. The concentration of non-lipid-complexed amphotericin in plasma was determined by a validated bioassay. Pharmacokinetic parameters after single doses and during steady state were calculated using standard noncompartmental methods. AmBisome was well tolerated at this dose. Complete pharmacokinetic profiles for weeks 1 and 4 were obtained in 12 patients. The half-life calculated in this pediatric population was shorter on average than reported in adults (45 hours vs 152 hours). The volume of distribution correlated best with body weight (R(2) = .55), and clearance was best predicted by initial serum creatinine level (R(2) = .19). Mean (+/- standard deviation) individual plasma trough concentrations were 0.23 (0.13) mg/L after single doses and 0.47 (0.41) mg/L after multiple doses. Mean steady-state area under the curve was higher at week 4 than after a single dose (P < .05). Single-dose and steady-state pharmacokinetic profiles were similar in 8 patients, whereas in 4 patients the week 4 profile showed nonlinear elimination. However, plasma concentrations at 7 days (Cmin) were not significantly different after the first and fourth doses, suggesting no significant accumulation over the course of therapy. Our data show measurable amphotericin B plasma concentrations 7 days after high-dose infusion of AmBisome. This suggests that once-weekly dosing, as described in this study, may provide useful protection against fungal infections.