1210 Network Meta-Analysis of Associated Mortality with Transfusion Strategies for Major Haemorrhage in Adults

Abstract Aim The aetiology of a large proportion of mortality in trauma patients is attributed to acute major haemorrhage. Damage control resuscitation and surgery have been developed as part of management, which often includes large volume blood product transfusion. The aim of this meta-analysis was to investigate differences in mortality between transfusion strategies used in major haemorrhage in adults (18+ years). Method A frequentist random effects network meta-analysis (NMA) using the netmeta package in R 4.0.5 of the eligible papers in the systematic review “Optimal dose, Timing and Ratio of Blood Products in Massive Transfusion: Results from a systematic review” compared a 1:1:1 ratio of red blood cells (RBC), fresh frozen plasma (FFP) and platelets (P) to a ratio of 2:1:1, whole blood, and laboratory-guided replacement. Homogeneity and heterogeneity were assessed with Cochran’s Q. Results Mortality at one month by group was 21.8% (12/55) for whole blood, 17.0% (58/342) for 2RBC:1FFP:1P, 14.3% (5/35) for laboratory-guided transfusion and 22.3% (96/43) for standard 1RBC:1FFP:1P ratio. The NMA showed no significant differences between groups, with relative risk of death when compared to 1RBC:1FFP:1P by one month of 1.42 (0.631; 3.19, p = 0.398) for whole blood, 0.7643 (0.562; 1.04, p = 0.0871) for 2RBC:1FFP:1P and 0.4396 (0.1741; 1.1099, p = 0.0820) for laboratory-guided transfusion. Conclusions The results demonstrated no significant differences in mortality between transfusion strategies in major haemorrhage - however, some arms of the NMA contain few patients and so may be underpowered to detect differences between groups. On this evidence, current use of component therapy is not inferior to another transfusion strategy.