Prehospital EMS Blood Transfusion and Fluid Interventions for Hemorrhagic Shock

Le résumé fourni par la source

Objective. To assess the effectiveness and potential harms of blood transfusion and fluid interventions for hemorrhagic shock in the prehospital setting. Data sources. Ovid MEDLINE®, the Cochrane Central Register of Controlled Trials, the Cochrane Database of Systematic Reviews, CINAHL®, and Scopus® from 1990 through August 2025. Review methods. We developed a protocol with input from 14 experts to synthesize evidence about prehospital transfusion or infusion interventions for traumatic and nontraumatic hemorrhagic shock in patients of all ages. We included randomized controlled trials (RCTs) and comparative nonrandomized studies of interventions (NRSIs) of whole blood, packed red blood cells (PRBC), plasma, and crystalloid fluids in prehospital settings, with outcomes reported up to 30 days postinjury. We synthesized findings on implementation barriers and facilitators using qualitative methods, classified evidence gaps found in the included literature, and characterized primary studies according to the gaps that needed to be addressed. Screening and data extraction were conducted by paired investigators with artificial intelligence assistance. Risk of bias was assessed using Cochrane criteria for RCTs and U.S. Preventative Services Task Force criteria for NRSIs. Strength of evidence (SOE) was graded, and pairwise meta-analyses were conducted for mortality outcomes, stratified by study design, intervention, and comparator. Results. Of 6,027 references, we included 119 studies (128 publications) mostly rated moderate-to-high risk of bias, with 44 (49 publications) consisting of 40,277 total patients for Key Questions 1–5. Identified studies indicated: Insufficient SOE for mortality across all timepoints for whole blood compared with usual care or active interventions (Key Question 1). No difference in mortality for PRBC versus usual care or active interventions at 24-hours (SOE low), discharge (SOE moderate) and 30 days (SOE low; Key Question 2). No difference in mortality for plasma compared with usual care across all timepoints (low SOE), but insufficient SOE for mortality up to 30 days compared with active interventions (Key Question 3). Insufficient SOE for 24-hour and 30-day mortality for low-volume compared with high-volume crystalloid fluids and no difference in mortality across all timepoints for crystalloid fluids compared with other types of fluids (low SOE; Key Question 4). Insufficient SOE for the effect of different transfusion strategies on mortality, but a moderate to large reduction in mortality at discharge for the Advanced Resuscitative Care Bundle, consisting of Calcium, tranexamic acid, and PRBC, compared with usual care (SOE moderate; Key Question 5). Harms were reported infrequently across all Key Questions but include coagulopathy and acute respiratory distress syndrome. Future research should focus on increased rigor, generalizability, and system implementation (Key Question 6). Findings from the Contextual Question suggest leverage points for system change including trauma center and blood bank partnerships, organizational buy-in, clinician training, and data evaluation. Conclusions. There is insufficient information from published studies included in this review demonstrating which blood transfusion or fluid interventions provide the greatest benefit to support clinical guidelines or policy recommendations. This underscores the need for research directly comparing patient-centered outcomes with sufficient rigor and adequate representation of diverse patient populations. Continued efforts are needed to identify effective strategies for implementing, sustaining, and evaluating prehospital blood transfusion and fluid resuscitation programs.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.