A major goal of new perfusion equipment is minimal trauma to blood elements. This study compares two perfusion systems, quantifies the change in blood components and generation of microemboli, as well as compares the hospital courses of each perfusion system. Forty-four coronary patients were assigned to either Group S, a silicone rubber membrane (SciMed) and centrifugal pump (Bio-Medicus) (N=19) or Group C (our routine equipment), a microporous polypropylene membrane (COBE CML) and roller pump (COBE)(N=25). The rise in plasma hemoglobin (26±14mg* in Group S and 26±17mg* in Group C), the drop in hematocrit (-15.0±3.9* in GroupS and -14.7±3.8* in Group Cat the second post-op day), and the decrease in platelet count (-152,000±78,000* in Group S and -129,000±52,000* in Group C) were similar in both groups. There was no difference in rise in post-op alveolararterial oxygen gradients or debris generated by each system. 27.7% in Group S required red cell transfusions and only 8% required red cell transfusions in Group C. There was no significant difference in clinical endpoints such as ICU stay, hospital stay and complication rate. We found no advantage to more expensive perfusion devices and no improvement upon the extensive CPB damage to formed blood elements.