OBJECTIVES: Although amphotericin B (AmB) and its lipid formulations are used for the treatment of fungal infections of the CNS, the kinetics of AmB in the CSF after intravenous administration of liposomal amphotericin B (LAmB) are not well characterized. PATIENTS AND METHODS: From 14 paediatric haemato-oncological patients (aged 0.4-19.5 years, median 7.6 years), we obtained 30 CSF samples by means of routine punctures (performed for intrathecal treatment of the underlying diseases) at different timepoints after the prophylactic intravenous infusion of LAmB (AmBisome, 3 mg/kg/day). Concurrent serum samples were obtained to calculate the transfer rates. An HPLC method was used for AmB detection. RESULTS: CSF levels of AmB 1-100 h after the intravenous infusion of LAmB were between 10 and 120 ng/mL, except in one case with a level of 529 ng/mL. Concurrent serum levels were about 1000-fold higher, ranging between 3 and 75 μg/mL. CSF levels did not show a clear time-dependent concentration profile, but remained at a steady-state for longer than 48 h after infusion. The transfer rate ranged from 0.02% to 0.92% (median 0.13%) and correlated significantly (r=0.801, P<0.001) with increasing time after infusion. CONCLUSIONS: After the intravenous administration of LAmB, AmB CSF levels were low, confirming published animal data. CSF levels remained at a steady-state level for longer than 48 h. As indicated by published post mortem data, higher levels in brain tissue, which would be necessary for the successful treatment of CNS infections, might be possible.
OBJECTIVES: Although amphotericin B (AmB) and its lipid formulations are used for the treatment of fungal infections of the CNS, the kinetics of AmB in the CSF after intravenous administration of liposomal amphotericin B (LAmB) are not well characterized. PATIENTS AND METHODS: From 14 paediatric haemato-oncological patients (aged 0.4-19.5 years, median 7.6 years), we obtained 30 CSF samples by means of routine punctures (performed for intrathecal treatment of the underlying diseases) at different timepoints after the prophylactic intravenous infusion of LAmB (AmBisome, 3 mg/kg/day). Concurrent serum samples were obtained to calculate the transfer rates. An HPLC method was used for AmB detection. RESULTS:CSF levels of AmB 1-100 h after the intravenous infusion of LAmB were between 10 and 120 ng/mL, except in one case with a level of 529 ng/mL. Concurrent serum levels were about 1000-fold higher, ranging between 3 and 75 μg/mL. CSF levels did not show a clear time-dependent concentration profile, but remained at a steady-state for longer than 48 h after infusion. The transfer rate ranged from 0.02% to 0.92% (median 0.13%) and correlated significantly (r=0.801, P<0.001) with increasing time after infusion. CONCLUSIONS: After the intravenous administration of LAmB, AmBCSF levels were low, confirming published animal data. CSF levels remained at a steady-state level for longer than 48 h. As indicated by published post mortem data, higher levels in brain tissue, which would be necessary for the successful treatment of CNS infections, might be possible.