ORIGINAL ARTICLES Preoperative Use of Enoxaparin Increases the Risk of Postoperative Bleeding and Re-exploration in Cardiac Surgery Patients Susan B. McDonald, MD,* Maurizio Renna, MD, FRCA,* Edward L. Spitznagel Jr, PhD,† Michael Avidan, MBBCh, FCA,* Charles W. Hogue Jr, MD,* Marc R. Moon, MD,‡ Benico Barzilai, MD,§ Rao Saleem, MD,* Jerome M. McDonald, MD, and George J. Despotis, MD* Objective: The purpose of this study was to investigate if the preoperative use of new platelet inhibitors and low– molecular-weight heparins may contribute to bleeding after cardiac surgery. Design: Retrospective data review. Setting: University teaching hospital. Participants: One hundred eleven patients divided in 5 groups. Interventions: Patients were grouped according to preop- erative antithrombotic regimen: group 1, control, no agents (n  55); group 2, clopidogrel (n  9); group 3, enoxaparin (n  17); group 4, any GP IIb/IIIa inhibitor (n  14); and group 5, any drug combination (n  15). Data included cumulative mediastinal chest tube drainage, allogeneic blood transfu- sions, total blood donor exposures, and re-exploration. Measurements and Main Results: Use of any drug (groups 2-5) resulted in greater total blood transfusions and donor exposure (p  0.0003) than control, especially red cells (p  0.002) and platelets (p  0.006). A greater percentage of patients on enoxaparin required mediastinal re-exploration for nonsurgical bleeding versus control (3/17 v 0/55, p  0.001). The use of enoxaparin was associated with signifi- cantly higher chest tube output after the first 24 hours post- operatively (p  0.048). Conclusion: Newer antithrombotic agents were associ- ated with greater transfusion rates and total donor expo- sures. Enoxaparin use was associated with greater overall blood loss and with higher incidence of mediastinal re-ex- ploration. The relative risk-benefit ratio of reduced peripro- cedure morbidity versus increased bleeding complications has yet to be determined. © 2005 Elsevier Inc. All rights reserved. KEY WORDS: platelets, inhibitors, heparin, surgery, bleed- ing, complications N EW ANTITHROMBOTIC drugs have been used with increasing frequency in the treatment of acute coronary syndromes. Such medications include adenosine diphosphate (ADP)-mediated platelet inhibitors, glycoprotein IIb/IIIa (GP IIb/IIIa) receptor inhibitors, and low–molecular-weight hepa- rins (LMWHs). Clinical trials have shown that many of these drugs, either alone or in combination, may offer better out- comes over conventional treatment with unfractionated heparin to certain patient groups, especially those patients undergoing interventional procedures for acute coronary syndromes. 1–9 These subsets of patients, however, are also at high risk for requiring urgent or emergent cardiac surgery, thus raising the question whether the use of these drugs increases the risk of perioperative bleeding and its consequences that range from the small but finite risk of transfusion reactions and infection transmission to the risk related to mediastinal re-explora- tion. 10,11 Accordingly, the authors sought to evaluate whether the use of these new antithrombotic drugs were associated with risk for increased bleeding and blood transfusion requirements in patients requiring coronary artery bypass grafting surgery (CABG). METHODS After institutional review board approval, data were retrospectively reviewed from adult patients undergoing cardiac surgery at Barnes- Jewish Hospital, St. Louis, MO, from January 1998 through June 1999. Only patients having CABG surgery were included; those undergoing valve operations or combined valve/CABG were excluded from anal- ysis. Patients receiving preoperative intravenous unfractionated hepa- rin, oral coumadin, or aspirin therapy were included in the study. Patients were grouped into 5 categories depending on their medical regimen. Group 1, the control group (n = 55), consisted of randomly selected patients who had surgery during the same time period as the drug groups but who did not receive platelet inhibitors or LMWH. Group 2 (n = 9) received clopidogrel (Plavix; Bristol-Meyers Squibb/ Sanofi, New York, NY); group 3 (n = 17) received the LMWH enoxaparin (Lovenox; Aventis Pharmaceuticals); group 4 (n = 14) received any GP IIb/IIIa receptor inhibitor, either abciximab (Reopro; Eli Lilly), eptifibatide (Integrilin; Millennium Pharmaceuticals, Cam- bridge, MA), or tirofiban (Aggrastat; Merck); and group 5 (n = 15) received a combination of more than one of the above drugs (eg, clopidogrel plus eptifibatide). Bleeding indices were represented by amount of cumulative mediastinal chest tube drainage (in milliliters) at 12, 24, and longer than 24 hours postoperatively, by the number of blood product transfusions (in units), by the total number of donor exposures (eg, a transfusion of cryoprecipitate results in 10 donor exposures), and by the need for mediastinal re-exploration for bleeding or tamponade. Variables included in the analysis included age, gender, From the Departments of *Anesthesiology, †Mathematics and Divi- sion of Biostatistics, ‡Surgery, §Medicine, and Cardiothoracic Sur- gery, Washington University, St. Louis, MO. Address reprint requests to George J. Despotis, MD, Department of Anesthesiology, Box 8054, Washington University School of Medicine, 660 South Euclid, St. Louis, MO 63110. E-mail: despotig@notes. wustl.edu © 2005 Elsevier Inc. All rights reserved. 1053-0770/05/1901-0002$30.00/0 doi:10.1053/j.jvca.2004.11.002 4 Journal of Cardiothoracic and Vascular Anesthesia, Vol 19, No 1 (February), 2005: pp 4-10