Objective: To evaluate and compare reversal of anticoagulation with different dose regimens of protamine in patients undergoing to CPB (cardiopulmonary bypass), one given according to the heparin dose administered and another calculated according to patient's weight. Patients y Methods: Patients subjected to CPB and receiving a heparin dose greater than 300 IU/kg were enrolled. Exclusion criterias were: preoperative coagulopathy and hypothermic circulatory arrest. The anesthetic technic, drugs given and blood products transfusion were decided by the attending anesthesiologist. Patients were randomized to: Group A or standard dose: Reversal with 0.8 mg of protamine for each 100 IU of heparin given. Group B or reduced dose: Reversal with 2.4 mg of protamine per kilogram of patient's weight, independent of heparin dose used. The protamine was prepared for a person blinded to group allocation, same as the team taking care of the patient. The patient's follow up in the ICU during the first 24 hours was also done by someone blinded to group allocation. Results: There was only one demographic difference at baseline: more women in Group B (p = 0.029). There were no differences among the preoperative: hematocrit, platelets count, oral anticoagulant treatment, heparin administration, aspirin consumption and surgical plan. In the intraoperative course there were no differences in the baseline ACT, hematocrit during CBP, ACT in CBP and CBP duration. The average heparin dose (adjusted per kilogram) was greater in Group B (p = 0.0433). The protamine/heparin ratios were different among groups (Group A 0.81; Group B 0.44), as expected in this study design. The activated coagulation time (ACT) after protamine administration was similar in both groups, with similar amount of supplemental protamine giving during surgery. Transfusion of RBC and the hematocrit at the end of surgery were also similar. At the ICU both groups were nearly identical with respect to accumulated postoperative bleeding, supplemental protamine administration, re-explorations and the hematocrit at 24 hours. The total amount of units of RBC and platelets given was not different among groups. Conclusions: The heparin reversal was efficient in both groups. Reduced protamine dose was not related to more postoperative bleeding or more requirement of blood products administration. Considering the side effects of protamine, it is possible to decrease its dose without compromising effectiveness of heparin reversal.