Blood transfusion strategies for major bleeding in trauma.

Background: Trauma is a leading cause of morbidity and mortality worldwide. Research shows that haemorrhage and trauma-induced coagulopathy are reversible components of traumatic injury, if identified and treated early. Lack of consensus on definitions and transfusion strategies hinders the translation of this evidence into clinical practice.

Objectives: To assess the beneficial and harmful effects of transfusion strategies started within 24 hours of traumatic injury in adults (aged 16 years and over) with major bleeding.

Search methods: CENTRAL, MEDLINE, Embase, five other databases, and three trial registers were searched on 20 November 2023. We also checked reference lists of included studies to identify any additional studies.

Selection criteria: We included randomised controlled trials (RCTs) of adults (aged 16 years and over) receiving blood products for the management of bleeding within 24 hours of traumatic injury.

Data collection and analysis: We used standard Cochrane methodology to perform the review and assessed the certainty of the evidence using GRADE.

Main results: We included 18 RCTs with 5041 participants. Comparison 1: Prehospital transfusion strategies Five studies compared use of plasma (fresh frozen plasma (FFP) or lyophilised plasma) versus 'standard of care'. We are uncertain of the effect of plasma on all-cause mortality at 24 hours (risk ratio (RR) 1.05, 95% confidence interval (CI), 0.48 to 2.30; 3 studies, 279 participants; very low certainty evidence). There is probably no difference between plasma and standard of care in all-cause mortality at 30 days (RR 0.95, 95% CI 0.78 to 1.17; 3 studies, 664 participants; moderate-certainty evidence). However, the results of one cluster-RCT that could not be included in our meta-analysis suggested that plasma may be associated with a lower risk of death at 30 days (RR 0.54, 95% CI 0.42 to 0.70; 1 study, 481 participants; low-certainty evidence). There may be no difference between plasma and standard of care in the total number of thromboembolic events in 30 days (RR 1.23, 95% CI 0.67 to.2.27; 4 studies, 586 participants; low-certainty evidence). Comparison 2: In-hospital transfusion strategies Ten studies evaluated this comparison, seven providing usable data. The studies evaluated cryoprecitate (three studies); fixed-ratio blood component transfusion (three studies); fresh frozen plasma (FFP) (one study); lyophilised plasma (one study); leucoreduced red blood cells (one study); and a restrictive transfusion strategy (one study). All-cause mortality at 24 hours For all-cause mortality at 24 hours, there is probably no difference between: • cryoprecipitate plus a major haemorrhage protocol (MHP) versus MHP alone (RR 0.92, 95% CI 0.70 to 1.21; 1 study, 1577 participants; moderate-certainty evidence); and • blood products (plasma:platelets:red blood cells (RBCs)) transfused in 1:1:1 ratio versus 1:1:2 ratio (RR 0.75, 95% CI 0.52 to 1.08; 1 study, 680 participants; moderate-certainty evidence). We are uncertain of the effect on all-cause mortality at 24 hours for: • blood products (RBCs:FFP) transfused in 1:1 ratio versus transfusion according to coagulation and full blood count results (Peto odds ratio (POR) 0.45, 0.17 to 1.22; 1 study, 434 participants; very low certainty evidence); and • lyophilised (FlyP) plasma versus FFP (POR 1.04, 95% CI 0.06 to 17.23; 1 study, 47 participants; very low certainty evidence); All-cause mortality at 30 days For all-cause mortality at 30 days, there is probably no difference between blood products (plasma:platelets:RBCs) transfused in a 1:1:1 ratio versus a 1:1:2 ratio (RR 0.85, 95% CI 0.65 to 1.11; 1 study, 680 participants; moderate-certainty evidence). There may be little to no difference between the following interventions in all-cause mortality at 30 days: • cryoprecipitate plus MHP versus MHP alone (RR 0.77, 95% CI 0.33 to 1.78; 2 studies, 1572 participants; low-certainty evidence); and •leucoreduced RBCs versus standard RBCs (RR 1.20, 95% CI 0.74 to 1.95; 1 study,55 participants; low certainty evidence). We are uncertain of the effect on all-cause mortality at 30 days for: •lyophilised plasma versus FFP (RR 0.75, 95% CI 0.28 to 2.02; 1 study, 47 participants; very low certainty evidence); and • blood products (plasma:platelets:RBCs) transfused in 1:1:1 ratio versus standard MHP (RR 2.25, 95% CI 0.90 to 5.62; 1 study, 69 participants; very low certainty evidence). Total number of thromboembolic events at 30 days There may be little to no difference between the following interventions for total thromboembolic events at 30 days: • cryoprecipitate plus MHP versus MHP alone (RR 0.55, 95% CI 0.08 to 3.72; 2 studies, 1645 participants; low-certainty evidence); and • blood products (plasma:platelets:RBCs) transfused in 1:1:1 ratio versus 1:1:2 ratio (RR 1.03, 95% CI 0.75 to 1.42; 1 study, 680 participants; low-certainty evidence). We are uncertain of the effect on the total number of thromboembolic events at 30 days for: •blood products (plasma:platelets:RBCs) transfused in 1:1:1 ratio versus standard MHP (POR 6.83, 95% CI 0.68 to 68.35; 1 study, 69 participants; very low certainty evidence). Comparison 3: Whole blood versus individual blood products We are uncertain of the effect of modified (leucoreduced) whole blood versus blood products (RBCs:plasma) transfused in a 1:1 ratio on all-cause mortality at 24 hours (RR 1.13, 95% CI 0.37 to 3.49) or 30 days (RR 1.62, 95% CI 0.69 to 3.80) (1 study, 107 participants; very low certainty evidence). Comparison 4: Goal-directed blood transfusion strategy of viscoelastic haemostatic assay (VHA) versus conventional laboratory coagulation tests (CCT) to guide haemostatic therapy There may be little or no difference in all-cause mortality at 24 hours between VHA and CCT (RR 0.85, 95% CI 0.54 to 1.35; 1 study, 396 participants; low-certainty evidence). We are uncertain of the effects on all-cause mortality at 30 days (RR 0.75, 95% CI 0.48 to 1.17; 2 studies, 506 participants; very low certainty evidence). There is probably no difference between VHA and CCT in total thromboembolic events at 30 days (RR 0.65, 95% CI 0.35 to 1.18; 1 study 396 participants; moderate-certainty evidence).

Authors' conclusions: Overall, there was little to no evidence of a difference between blood transfusion strategies for mortality or thromboembolic events. The studies covered a wide range of interventions, and the comparators and standard of care practice varied between trials, thereby limiting the pooling of data. Further research is needed.