Perfluorocarbons (PFCs) and their emulsions (PFCEs) were used in organ preservation before transplantation, but not in organ perfusion. Our purpose was to achieve organ perfusion with a PFCE at room temperature or at 37 degrees C, i. e. with oxygenation, to prevent damages related to reoxygenation after hypoxia. Therefore, we first investigated the effect of such emulsions on endothelial cells, the first cells to be in contact with the emulsion. A stem emulsion was prepared from perfluorooctyl bromide (90% w/v), emulsified with egg yolk phospholipids (2% w/v) and stabilized with a mixed fluorocarbon-hydrocarbon "molecular dowel" (1.4% w/v) (droplets of ca 0.2 micron in diameter). This emulsion was found to be stable when diluted with cell culture media or organ preservation fluids. Endothelial cells from human umbilical vein (HUVECs) were cultured in multiwell plates in M199 medium (with growth factors, 10% foetal calf serum and 5% human serum). Confluent cells were incubated overnight with 51Cr, washed and overlayed with M199 (control) or the above PFCE diluted 2x or 4x with M199 (test). After incubation, the cytotoxicity of the PFCEs was estimated by measuring 51Cr release and observing cell morphology by electron and light microscopy. The percentages of released 51Cr were identical to those of the control cells for the 2x, 3x or 4x diluted PFCEs at 4, 25 or 37 degrees C. After return to the M199 medium, the cells grew and multiplied normally. We conclude that the diluted PFCEs were devoid of cytotoxicity. The 2x diluted PFCE was however partially taken up by the cells: by microscopy, we observed intracellular PFC droplets and by density gradient analysis we found a slight increase in cellular density. The diluted PFCEs were compared to classical organ preservation solutions : HUVECs were incubated with UW (University of Wisconsin) or EC (EuroCollins) solutions at +4 and 37 degrees C (3, 17 or 24 h of incubation). The solutions were observed to be toxic to the cells under these conditions, with cell mortality after return to the M199 medium. This cytotoxicity may be attributed to the high K+ concentration of UW and EC, since similar assays performed on HUVECs with Hank's solution adjusted to 100 mM K+ showed a similar % of 51Cr release. UW and EC are therefore not acceptable as dilution media for PFCEs.