The hemopump HP 31 is an improved version of a catheter-mounted, transvalvular, left ventricular assist device, which can be placed into the left ventricle through the ascending aorta. The purpose of this study was to examine the influence of hematocrit and afterload on the pump flow. The hemopump was tested using a flow bench model filled with heparinized bovine blood. The measurements were performed at four various hematocrit values: 16%, 24%, 32%, and 40%. The pump flow was measured at each hematocrit value under increasing afterload pressures (40-120 mm Hg), by all pump speed levels (n = 7). The average pump flow at highest pump speed and lowest afterload was 5.1 +/- 0.3 l/min (mean +/- standard deviation). The influence of afterload on the pump flow was statistically significant (p < 0.001). The highest afterload pressure of 120 mm Hg caused a reduction in pump flow of 24 +/- 5%. The alterations of hematocrit values caused no statistically significant influence on the pump flow (p = 0.72). The results of our study enabled the construction of the nomogram for the in vivo determination of the pump flow. The in vivo performances of the hemopump can be improved through the afterload reduction, especially in the weaning phase of treatment. The oxygen delivery can be improved through the increase in hematocrit values without significant impairment of the pump flow.