Fungal cholangitis is a potentially life-threatening condition. As amphotericin B (AmB) has a broad antimycotic spectrum, in this study its biliary penetration and activity was determined in two patients treated with liposomal AmB (L-AmB) and in one patient receiving AmB colloidal dispersion (ABCD). Biliary and plasma AmB levels were quantified by high-performance liquid chromatography after purification by solid-phase extraction. For assessment of biliary AmB activity, isolates of Candida albicans, Candida tropicalis, Candida glabrata and Candida krusei were incubated in porcine bile at AmB concentrations of 0.025-5.00 mg/L. In addition, patient bile samples retrieved for AmB quantification were inoculated with the same Candida strains. Biliary AmB concentrations were lower and displayed a slower rise and decline than plasma levels. The highest penetration ratio, as expressed by the ratio between the area under the AmB concentration-time curve in bile and plasma (liberated AmB) over the sampling period (AUC0-n bile/AUC0-n LI plasma), was 0.28. Proliferation of C. albicans and C. tropicalis in bile was similar to that in culture medium, whereas growth of C. glabrata was diminished and proliferation of C. krusei was absent in bile. In comparison with culture medium, AmB activity decreased in spiked porcine bile. In all but one patient bile sample, fungal growth was delayed or lacking even when AmB was not detectable. However, no fungicidal effect was observed in patient bile at AmB concentrations up to 1.28 mg/L. Thus, a reliable response of fungal cholangitis to treatment with L-AmB or ABCD cannot be anticipated. Biliary AMB concentrations were lower and displayed a slower rise and decline than plasma levels. The highest penetration ratio as expressed by the ratio between the area under the time-AMB-concentration curve in bile and plasma (liberated AMB) over the sampling period (AUC 0-n bile /AUC 0-n LI plasma ) amounted to 0.28. Proliferation of C. albicans and C. tropicalis in bile was similar to that in culture medium whereas growth of C. glabrata was diminished and proliferation of C. krusei was absent in bile. In comparison with medium, AMB activity decreased in spiked porcine bile. In all but one patient bile sample, fungal growth was delayed or lacking, even when AMB was not detectable. However, no fungicidal effect was observed in patient bile at AMB concentrations of up to 1.28 mg/L. Thus, a reliable response of fungal cholangitis to treatment with LAMB or ABCD cannot be anticipated.