We wished to determine if reduction in the standard heparin administration for cardiopulmonary bypass could be accomplished safely with the use of membrane oxygenators. An experimental study was designed to evaluate two different heparin administration protocols for cardiopulmonary bypass with hollow-fiber membrane oxygenators. Two groups of six pigs were submitted to hypothermic cardiopulmonary bypass (28 degrees C) for 3 hours, then rewarmed, decannulated, and reassessed after 1 hour. In group I (control) heparin was administered to maintain the activated clotting time in excess of 450 seconds; in group II activated clotting time was maintained between 250 and 300 seconds. The mean total heparin administered was 41,000 units in group I and 25,000 units in group II. Concentration of coagulation factors II, V, and VIII, fibrinogen, and platelet count were determined before, during, and 1 hour after bypass. No significant difference in any of these coagulation parameters was observed between the groups. The performance of the oxygenators was similar in both groups, with no evidence of thrombosis. Thus reduced heparin administration, enough to keep activated clotting time between 250 and 300 seconds, was not related either to major coagulation factors and platelet consumption or to derangements in the oxygenator's performance.