Experience in reducing the hemolysis of an impeller assist heart.

Blood trauma has been one of the main problems of centrifugal pumps. The difficulties in reducing hemolysis are many, and all the factors causing excessive hemolysis always act together, making them difficult to discover and distinguish. Furthermore, error could occur at many points during hemolysis testing, making it difficult to repeat results. In developing the low hemolysis pulsatile and nonpulsatile impeller pumps the authors established an experimental method for investigating and searching for the hemolysis factors. In this study two pumps with only one differing factor were compared or only one factor on one pump was changed in the middle of the test period. In this way the effect of the individual factor on pump hemolysis could be seen and some factors have been thus confirmed as important reasons for hemolysis: 1) the drift of the pump output (including the volume and efficiency) from the design point; 2) impeller vane angles, i.e., the radial logarithmic spiral angle and the axial helical spiral angle; 3) roughness of vane surface and other blood contacting surfaces of the sealing box and pump housing; 4) vibration of the rotor resulting from dynamic disequilibrium; and 5) prerotation swirl at the inlet of the pump. The blood pressure to be pumped has been shown to have no influence on pump hemolysis. After eliminating the hemolysis factors, the blood trauma of the impeller heart has been reduced remarkably. The index of hemolysis of the nonpulsatile pump is 0.015, about one fifth of a clinically used roller made in Shanghai and two sevenths of one Sarns 7,000 Roller; the index of hemolysis of the pulstile pump is 0.020, about one sixth of a self-made diaphragm pump and one thirteenth of the Polystan pulsa tile pump.