An in vitro model cardiopulmonary bypass (CPB) circuit consisting ot tubing, oxygenator and venous reservoirs with either a roller or a centrifugal pump, and with either heparin-coated (Carmeda Bioactive Surface, CBAS) or uncoated surfaces, was studied with respect to 'blood activation', using small-scale-based blood volume (450 + 500 ml). Sixteen circuits were tested in each pump group, eight with and eight without heparin-coated surfaces, by circulating heparinized fresh human blood for 72 hours at 30 degrees C. Blood plasma, sampled at defined intervals, was analysed for haemolysis (lactate dehydrogenase and potassium), complement activation (C3bc and C5b-9 (TCC)), complement lytic inhibitors (vitronectin and clusterin), coagulation activation (fibrinopeptide A), granulocyte (lactoferrin and myeloperoxidase) and platelet (beta-thromboglobulin) activation and contaminating endotoxin. The heparin coating significantly reduced the concentrations of C3bc, TCC, fibrinopeptide A, lactoferrin, myeloperoxidase and beta-thromboglobulin. The two pump types did not differ with respect to these parameters, but the roller pump caused significantly higher increases in plasma LDH and potassium and significantly greater reductions in clusterin and vitronectin than the centrifugal pump. Endotoxin concentration was low at the start and after 24 hours in all groups. These results confirm that heparin-coated CPB surfaces reduce blood activation, and suggest that centrifugal pumps cause less haemolysis and less reduction in lytic complement inhibitors than roller pumps.
1996. Vol. 11, no 2, 113-123 p.