The analysis of hemodialysis (HD)-related bioincompatibility is focused mainly on phenomena observed in peripheral blood. However, since biocompatibility originates inside the dialyzer, white blood cells (WBC) adhering to the dialyzer are probably most subject to the influence of both dialyzer membrane and dialysate. In order to collect membrane-adherent cells, a reliable and reproducible elution technique was developed. After 3 h of HD, blood was returned to the patient with 0.9% NaCl. Then, dialyzers were eluted by recirculation of phosphate-buffered saline (PBS) or PBS/3 mM EDTA for 20 min, with or without prior flushing with 200 ml PBS. Finally, remaining adherent cells were collected by an afterwash with 10% trypsin. These solutions, as well as blood samples, were analyzed for WBC count, viability and differentiation. Random eluate samples were analyzed by flow cytometry, and the influence of elution on PMN activation was tested in a separate control experiment. WBC numbers decreased by flushing before elution, whereas cell numbers were maximal after elution with PBS/3 mM EDTA (30 x 10(6)). Trypsin afterwash resulted in a further yield of 12 x 10(6) cells. The eluates contained 81% PMN (blood 68%, p < 0.01), with a degranulated appearance, and only 12% lymphocytes (blood 21%, p < 0.05); cell viability in the eluates was > 95%. The eluted cells could be analyzed by flow cytometry, and the procedure itself induced only minimal PMN activation. In conclusion, a maximal number of adherent cells, consisting mainly of PMN, was obtained by direct elution with PBS/3 mM EDTA. The method itself did not induce marked PMN activation, and the cells obtained were suitable for further investigations, including flow cytometry.