Automated separation of whole blood in top and bottom bags into componentsusing the component G4

Background and Objectives: Whole blood can be separated by hard spin centri fugation into layers of blood components according to their specific gravit y. The aim was to develop a program for an automatic separator to subsequen tly express the various components into their respective satellite bags in top and bottom systems with the following requirements: a red cell concentr ate with a low leukocyte and platelet contamination, a 'cell-free' plasma, and a buffy coat with a volume of about 50 mi with an acceptable loss of re d cells. Materials and Methods: The Compomat G4 possesses an independently moving upper and lower press, to automatically express plasma or red cells to satellite bags of top and bottom systems. The influence of the extension of the lower press was studied by pooling and dividing two units of whole blood, and separating these units after centrifugation (2,960 g, 10 min) ei ther with a program where the lower press was completely extended (program C), or with a program that left approximately 1 mm between the door and the lower press (program D). Results: The program (program D), where the lower press was not completely extended, yielded a buffy coat with a volume of 5 2+/-1 mi (mean +/- SD, n = 36), which contained >75% leukocytes and >90% pl atelets of the original whole blood unit, with a red cell loss in the buffy coat of 21+/-1 mi (10.8+/-0.8% of the original volume). The leukocyte cont ent of the red cell concentrates was 775+/-379x10(6) per unit, whereas the plasma contained 3+/-3x10(6) leukocytes and 4+/-3x10(9) platelets per unit. The pooling experiment indicated that complete extension of the lower pres s (program C) resulted in a significantly higher leukocyte contamination of the red cell concentrate (788+/-431x10(6) vs. 658+/-419x10(6); n = 12; p = 0.03), while there was no difference in the yield of red cells or plasma. The buffy coat produced with program D contained significantly more leukocy tes (2,242 +/-396x10(6) vs. 2,065+/-327x10(6), p = 0.005) and more platelet s (96+/-14x10(9) vs. 92+/-17x10(9), p = 0.02) per unit than with program C, probably because buffy coat cells sticking to the container wall are not e xpressed to the red cell concentrate, and thus remain in the buffy coat bag . Therefore, program D met our specifications for blood products. Conclusio ns: The Compomat G4 war rants reproducible separation of whole blood in top and bottom bags into red cells, buffy coat and plasma meeting our specific ations.