Summary. It has been suggested that a better outcome of neutropenia-associated invasive fungal infections can be achieved when high doses of lipid formulations of amphotericin B are used. We now report a randomized multicentre study comparing liposomal amphotericin B (AmBisome, 5 mg/kg/d) to amphotericin B deoxycholate (AmB, 1 mg/kg/ d) in the treatment of these infections. Of 106 possible patients, 66 were enrolled and analysed for efficacy: nine had documented fungaemia, 17 had other invasive mould infections and 40 had suspected pulmonary aspergillosis. After completion of the course medication, in the AmBisome group (n ¼ 32) 14 patients had achieved complete response, seven a partial response and 11 were failures as compared to 6, 13 and 15 patients (n ¼ 34) treated with AmB (P ¼ 0·09); P ¼ 0·03 for complete responders. A favourable trend for AmBisome was found at day 14, in patients with documented infections and in patients with pulmonary aspergillosis (P ¼ 0·05 and P ¼ 0·096 respectively). Mortality rates were lower in patients treated with AmBisome (adjusted for malignancy status, P ¼ 0·03). More patients on AmB had a >100% increase of their baseline serum creatinine (P < 0·001).The results indicate that, in neutropenic patients with documented or suspected invasive fungal infections AmBisome 5 mg/kg/d was superior to AmB 1 mg/kg/d with respect to efficacy and safety.

Human bronchoalveolar lavage (BAL) has been described to contain, besides a large number of alveolar macrophages (AM) (approximately 95%), small numbers of monocyte-like cells (approximately 2%) and dendritic cells (DC) (approximately 0.4%). To separate AM (high autofluorescence) from DC, we used a fluorescence activated cell sorter (FACS) to separate BAL cells into a low autofluorescent (LAF) fraction and a high autofluorescent (HAF) fraction. Immunocytologic and functional properties of these fractions were investigated. The LAF fraction was composed of acid phosphatase (APh)- and RFD9-negative cells, which were strongly positive for HLA-DR, L25, RFD1, and CD68. A portion of these cells expressed CD1a (22%) and My4 (60%). The marker pattern of these cells is reminiscent to that of intraepithelial bronchial DC and to that of blood DC. The majority of the LAF cells had a monocyte-like morphology, but after overnight culture the percentage of LAF cells with long cytoplasmic extensions (DC morphology) was strongly augmented (from 18 to 51%). The HAF fraction contained 100% AM, strongly positive for APh, HLA-DR, CD68, RFD7, and RFD9. In culture, the LAF cells formed clusters with T cells and vigorously stimulated the proliferation of allogeneic T cells and naive (CD45RO-negative) T cells. BAL and LAF cells produced higher responses in nonsmokers than in smokers. In contrast, HAF cells did not form clusters with T cells and did not stimulate allogeneic T cell proliferation. HAF cells even suppressed mitogen-driven T cell proliferation.(ABSTRACT TRUNCATED AT 250 WORDS)

Mononuclear phagocytes and dendritic cells (DC) play an important role in the immune response in the lung. DC act in the afferent phase of the immune response by presenting antigen to T cells, while macrophages play a role in the efferent phase by exerting phagocytic/cytotoxic functions. We investigated the localization and the marker pattern of these cells in the human lung. Macrophages, identified as large, rounded, acid phosphatase-positive cells, were mainly detected in the alveolar spaces, in the lumen of the bronch(iol)us, and in the bronchoalveolar lavage (BAL). They were positive for major histocompatibility complex (MHC) class II antigens (DR, DQ), CD68, RFD7, RFD9, and partly positive for RFD1. Irregularly shaped cells with a marker pattern comparable to that of blood-derived DC (positive for DR, DQ, L25, RFD1, and CD68) were predominantly observed in the epithelium and subepithelial tissue of the bronch(iol)us and in the bronchus-associated lymphoid tissue. In the epithelium, approximately 30% of these cells were positive for CD1a (OKT6). In the subepithelial tissue, these DC formed characteristic small clusters with T cells. The BAL, the alveolar spaces, and the alveolar walls contained only a small number of DC. These immunohistologic data suggest that the bronch(iol)us is well equipped to initiate immune responses. The high number of macrophages in the alveolar compartment, which have been described to suppress T cell proliferation, together with low numbers of DC, makes the alveolar compartment less suited for mounting an immune response.

Our data suggest that dendritic cells are involved in asthmatic inflammation and that corticosteroids may downregulate the number of dendritic.

Ursodeoxycholic acid has the potency to suppress eosinophilic inflammation outside the GI tract. This potential comprises to alter critical function of DCs, in essence, the effect of UDCA on DCs through the modulation of the DC/T cell interaction.