The REPLACE-2 trial of patients who underwent urgent or elective percutaneous coronary intervention (PCI) demonstrated a significantly lower bleeding risk with bivalirudin plus provisional glycoprotein IIb/IIIa inhibitor compared with unfractionated heparin with planned glycoprotein IIb/IIIa inhibitor. The goal of this analysis was to evaluate whether a hazard existed when unfractionated heparin or low-molecular-weight heparin was administered before study medication in the REPLACE-2 trial. The REPLACE-2 trial randomized 6,010 patients undergoing PCI to receive bivalirudin plus provisional glycoprotein IIb/IIIa inhibitor or unfractionated heparin plus planned glycoprotein IIb/IIIa inhibitor. The present study compared bleeding among patients treated with or without antithrombin therapy in the 48 hours before study treatment. Among patients treated with bivalirudin, there was no difference in protocol-defined major or minor bleeding, bleeding according to Thrombolysis In Myocardial Infarction criteria, or noncoronary artery bypass graft blood transfusions between the patients treated with versus without antithrombin therapy (p=NS). However, in patients treated with unfractionated heparin plus planned glycoprotein IIb/IIIa inhibitor, there was a significant increase in the composite of protocol-defined major or minor bleeding and in noncoronary artery bypass graft blood transfusions (p<0.05 for 3-way comparison vs no unfractionated heparin and for 2-way comparisons of no unfractionated heparin vs unfractionated heparin or low-molecular-weight heparin). In conclusion, in patients treated with bivalirudin, pretreatment with antithrombin therapy was not associated with increased bleeding. In contrast, among patients randomized to receive unfractionated heparin and planned glycoprotein IIb/IIIa, pretreatment with antithrombin therapy was associated with increased protocol-defined composite major or minor bleeding and increased need for transfusion.