Study | Number of participants | Inclusion criteria | Outcome measures | Whole blood | Component therapy | Results | Quality of the evidence (GRADE) |
Cotton et al
31 | 107 | >18 years of age, highest level trauma activation, active bleeding requiring uncross matched blood. |
Primary: 24 hours blood product use. Secondary: 24 hours mortality, 30-day mortality, length of stay, transfusion-associated complications, infections. | mWB (leucoreduced)+aPLT for every 6 units | RBC+plasma+aPLT for every 6 units | No statistical difference in 30-day mortality 22% mWB vs 14% BCT (p=0.26). No statistical difference in 24 hours mortality 11% mWB vs 10% BCT (p=0.83). No statistical difference in 24 hours component use (p=0.462). No other statistical differences between groups in ARDS, AKI, MODS. | Moderate ⊕⊕⊕ |
Perkins et al
29 | 369 | Trauma patients who required ≥10 units of blood component transfused within first 24 hours and did not receive both FWB and aPLT. |
Primary: survival at 24 hours and 30 days. Secondary: rates of ARDS, MODS, infection, embolic events. | FWB (non-leucoreduced)+RBC+FFP+cryo | aPLT+RBC+FFP+cryo | No statistical difference in 30-day mortality 43% FWB vs 40% aPLT (p=0.72). Note a large loss to follow-up (20% FWB vs 37.6% aPLT). No statistical difference in 24 hours mortality 19% FWB vs 16% aPLT group (p=0.52). Higher incidence of ARDS in FWB vs aPLT 18.8% vs 7.4% (p=0.002). No statistical difference in MODS, embolic events. Multivariate regression analysis of FWB vs aPLT 24 hours OR 3.38, 95% CI 0.96 to 11.87 (p=0.06) 30 days HR 1.38, 95% CI 0.77 to 2.47 (p=0.28). | Very low ⊕ |
Spinella et al
30 | 354 | Military combat patients in Afghanistan and Iraq who received at least 1 unit RBC and were treated at a level II or level III hospital. Those receiving both WFWB and aPLT were excluded. |
Primary: survival at 24 hours and 30 days. Secondary: blood product administration, adverse effects. | WFWB (non-leucoreduced)+RBC+plasma | aPLT+RBC+plasma | Statistically significant difference for both 24 hours and 30-day mortality 24 hours: WB 4% vs BCT 12% (p=0.018) 30 days: WB 5% vs BCT 18% (p=0.002). Multivariate logistic regression analysis improved 30-day survival in WFWB group vs BCT group OR 12.4, 95% CI 1.8 to 80 (p=0.01). Increased AKI in WB (8%) vs BCT group (3%) (p=0.04). No statistical difference in ARDS or embolic events. Actual blood volume transfused higher in BCT group (5.7 L vs 6.8 L, p=0.03) | Low ⊕⊕ |
Jones and Frazier 25 | 1745 | Age 18–45 years, ISS >25, required hospital admission, received blood transfusion. |
Primary: in-hospital mortality. Secondary: survival ORs. | WB (no additional information available) | RBC+PLT | No statistical difference in mortality 21% WB vs 26% BCT (p=0.27). Logistic regression analysis. BCT patients 3.2 times more likely to die vs WB OR 3.164, 95% CI 1.314 to 7.618 (p=0.01). | Very low ⊕ |
Nessen et al
26 | 488 | Military combat patients in Afghanistan requiring treatment by the six studied US Forward Surgical Teams. |
Primary: mortality determined at inpatient discharge. Secondary: mortality between uncross matched and cross-matched blood, number of products transfused. | FWB+RBC+FFP | RBC+FFP | No statistical difference for unadjusted in-hospital mortality between FWB (5.3%) and BCT (8.8%) (p value not reported). FWB were less likely to die vs BCT continuous variable logistic regression analysis OR 0.096, 95% CI 0.02 to 0.53 (p=0.008) stratified propensity score analysis OR 0.11, 95% CI 0.02 to 0.78 (p=0.03). FWB patients received significantly more units of RBC (12.7 vs 4.7, p≤0.001) and FFP (10 vs 2.6, p≤0.001) FWB patients were more likely to receive MBT (52.1% vs 11.6%, p≤0.001). No statistical difference in those who received type specific FWB or uncrossmatched FWB. | Low ⊕⊕ |
Yazer et al
8 | 192 | Male patients attending a level 1 trauma centre with hypotension secondary to bleeding who received at least 1 unit of WB compared with historical controls. |
Primary: haemolysis (haptoglobin used as marker) and transfusion reactions. Secondary: transfusion volumes, mortality rates (no additional information available). | WB (up to 2 unit, leukoreduced)+RBC+plasma+PLT+cryo | RBC+plasma+ PLT+cryo | No statistical difference in mortality between the groups (WB 36% vs BCT 28%, p=0.27). No statistical difference between the number of blood products received by each group. Median haptoglobin concentration on post-WB transfusion day 1 was 25.1 mg/dL (normal). No transfusion reactions in the WB group. | Very low ⊕ |
AKI, acute kidney injury; aPLT, apheresis platelets; ARDS, acute respiratory distress syndrome; BCT, blood component therapy; cryo, cryoprecipitate; FFP, fresh frozen plasma; FWB, fresh whole blood; GRADE, Grading of Recommendations Assessment, Development and Evaluation; ISS, injury severity score; MBT, massive blood transfusion ≥10 units; MODS, multiple organ dysfunction syndrome; mWB, modified whole blood; RBC, red blood cells; WB, whole blood; WFWB, warm fresh whole blood.