Platelet function during cardiopulmonary bypass using multiple electrode aggregometry: comparison of centrifugal and roller pumps.

Blood trauma may be lower with centrifugal pumps (CPs) than with roller pumps (RPs) during cardiopulmonary bypass (CPB), because, unlike RPs, CPs do not compress the tubing, and shear stress is considered lower in CPs than in RPs. However, relative platelet function remains unclear. Using multiple electrode aggregometry (MEA), we compared platelet function with CP and RP. Ten swine underwent CPB for 3 h, with five weaned off using CP and five using RP. Platelet function was measured using MEA, as were hemoglobin concentration and platelet count, before sternotomy, after heparin infusion, 30 min and 3 h after starting CPB, after protamine infusion, and 60 min after stopping CPB. Platelet activation was initiated with adenosine diphosphate (ADP), arachidonic acid (AA), and thrombin receptor-activating protein 6 (TRAP). Fibrinogen, platelet factor 4 (PF4), and β-thromboglobin (β-TG) concentrations were measured before sternotomy and 60 min after stopping CPB. In the CP group and using ADP, aggregation was significantly reduced 30 min (P = 0.019) and 3 h (P = 0.027) after starting CPB, recovering to baseline 60 min after CPB was stopped. In the RP group, aggregation was significantly decreased 30 min (P = 0.007) and 3 h (P = 0.003) after starting CPB and after protamine administration (P = 0.028). With AA, aggregation significantly decreased 30 min after starting CPB in both the CP (P = 0.012) and RP (P = 0.016) groups, slightly increasing 3 h after starting CPB and after protamine infusion, and recovering to baseline 60 min after CPB cessation. With TRAP, aggregation in the CP and RP groups decreased 30 min after starting the pump, although changes were not significant; aggregation gradually recovered after 3 h and returned to baseline 60 min after the pumps were stopped. There were no significant differences at all sampling points of MEA. In both groups, fibrinogen, PF4, and β-TG concentrations were similar 60 min after pump cessation and before sternotomy. Platelet function, evaluated with MEA, was lowest 30 min after CPB was started but did not decrease over time in either group. As assessed by MEA, platelet function using CP and RP did not differ significantly. Platelet dysfunction was caused mainly by initial contact with foreign materials and may not be dependent on type of pump.