Departments of Surgery, Center for Translational Injury Research, and Pathology and Laboratory Medicine, University of Texas Health Science Center, Houston, TX; and Gulf Coast Regional Blood Center, Houston, TX.
OBJECTIVES To determine whether resuscitation of severely injured patients with modified whole blood (mWB) resulted in fewer overall transfusions compared with component (COMP) therapy. BACKGROUND For decades, whole blood (WB) was the primary product for resuscitating patients in hemorrhagic shock. After dramatic advances in blood banking in the 1970s, blood donor centers began supplying hospitals with individual components [red blood
cell (RBC), plasma, platelets] and removed WB as an available product. However, no studies of efficacy or hemostatic potential in trauma patients were performed before doing so. METHODS Single-center, randomized trial of severely injured patients predicted to large transfusion volume. Pregnant patients, prisoners, those younger than 18 years or with more than 20% total body surface area burns (TBSA) burns were excluded. Patients were randomized to mWB (1 U mWB) or COMP therapy (1 U RBC+ 1 U plasma) immediately on arrival. Each group also received 1 U platelets (apheresis or prepooled random donor) for every 6 U of mWB or 6 U of RBC + 6 U plasma. The study was performed under the Exception From Informed Consent (Food and Drug Administration 21 code of federal regulations [CFR] 50.24). Primary outcome was 24-hour transfusion volumes. RESULTS A total of 107 patients were randomized (55 mWB, 52 COMP therapy) over 14 months. There were no differences in demographics, arrival vitals or laboratory values, injury severity, or mechanism. Transfusions were similar between groups (intent-to-treat analysis). However, when excluding patients with severe brain injury (sensitivity analysis), WB group received less 24-hour RBC (median 3 vs 6, P = 0.02), plasma (4 vs 6, P = 0.02), platelets (0 vs 3, P = 0.09), and total products (11 vs 16, P = 0.02). CONCLUSIONS Compared with COMP therapy, WB did not reduce transfusion volumes in severely injured patients predicted to receive massive transfusion. However, in the sensitivity analysis (patients without severe brain injuries), use of mWB significantly reduced transfusion volumes, achieving the prespecified endpoint of this initial pilot study.