A novel enzyme immunoassay based on direct detection of C5a by a monoclonal antibody (C17/5) specific for a neoepitope exposed in C5a/C5adesArg was used to measure in vivo and in vitro C5a formation during cardiopulmonary bypass. In vivo, we observed a significant threefold to fourfold increase in patient plasma C5a/C5adesArg levels from baseline values (5.6; 1.6 to 12.9 ng/mL) (median and range) up to 42 hours postoperatively (17.5; 6.5 to 46.0 ng/mL) when two different uncoated cardiopulmonary bypass circuits were used. Coating of the extracorporeal circuit with end-point-attached heparin completely abolished C5a formation in vitro during circulation of blood through the circuit for 120 minutes. The C5a concentration (median and range) was 3.2 (2.6 to 15.9) ng/mL at the start and 3.1 (2.7 to 15.0) ng/mL at the end of the experiment. In the uncoated setups the corresponding C5a concentrations were 10.1 (6.2 to 17.5) and 19.7 (13.1 to 24.3) ng/mL. Finally, heparin-coated cardiopulmonary bypass circuits were examined in vivo. C5a levels did not increase significantly during the cardiopulmonary bypass period in the heparin-coated group in contrast to the uncoated group, but the postoperative increase in C5a levels was similar in the two groups. We conclude that heparin coating improves biocompatibility by completely abolishing C5a formation in vitro. The discrepancy between the in vitro and the in vivo findings is probably related to the complicated biological turnover of C5a.