Pre‐hospital transfusion of red blood cells. Part 1: A scoping review of current practice and transfusion triggers

Abstract Objectives The primary aim of this scoping review is to describe the current use of pre‐hospital transfusion of red blood cells (PHTRBC) and to evaluate criteria used to initiate PHTRBC. The effects on patients' outcomes will be reviewed in Part 2. Background Haemorrhage is a preventable cause of death in trauma patients, and transfusion of red blood cells is increasingly used by Emergency Medical Services (EMS) for damage control resuscitation. However, there are no guidelines and little consensus on when to initiate PHTRBC. Methods PubMed and Web of Science were searched through January 2019; 71 articles were included. Results Transfusion triggers vary widely and involve vital signs, clinical signs of poor tissue perfusion, point of care measurements and pre‐hospital ultrasound imaging. In particular, hypotension (most often defined as systolic blood pressure ≤ 90 mmHg), tachycardia (most often defined as heart rate ≥ 120/min), clinical signs of poor perfusion (eg, prolonged capillary refill time or changes in mental status) and injury type (ie, penetrating wounds) are common pre‐hospital transfusion triggers. Conclusions PHTRBC is increasingly used by Emergency Medical Services, but guidelines on when to initiate transfusion are lacking. We identified the most commonly used transfusion criteria, and these findings may provide the basis for consensus‐based pre‐hospital transfusion protocols.

when pre-hospital healthcare providers arrive at a scene. Therefore, to replace (ongoing) blood loss, liberal volume replacement, typically with readily available crystalloid fluids, has been long advocated. 12 However, this practice is associated with adverse effects including dilutional coagulopathy, acidosis, hypothermia and accelerated blood loss. 13 In contrast, damage control resuscitation (DCR) involves restrictive fluid resuscitation, which avoids crystalloids while accepting some degree of hypotension, and is increasingly preferred until bleeding can be surgically controlled. [13][14][15][16][17][18] Red blood cells (RBCs) provide a more effective volume expansion than crystalloids; the infusion of large volumes of crystalloids or colloids can thus be avoided. The RBCs benefit haemostasis and thrombosis 19 and restore oxygen-carrying capacity, 20 thereby potentially reducing acidosis through tissue hypoxia. Military medical teams have long been transfusing blood products prior to patients' arrival at a surgical unit. 4,21 As proposed by Jansen et al, 22 differences in survival between civilian casualties who require massive transfusion (60%) 23 and military casualties (93%) 22 may be partly explained by this practice.
In recent years, civilian Emergency Medical Services (EMS) are increasingly transfusing RBCs before hospital arrival. 24,25 However, logistic and operational challenges are hampering the widespread implementation of blood transfusions in the pre-hospital setting.
Moreover, consensus regarding pre-hospital indications for blood transfusions is lacking, and evidence regarding the efficacy of this practice is scarce. Therefore, this systematic review consists of two parts. Part 1 is a scoping review in which we systematically gathered the research done in the area of pre-hospital transfusion of red blood cells (PHTRBC), aiming to describe the current challenges of prehospital transfusion and, in particular, to evaluate which criteria are currently used to initiate PHTRBC. This review may serve as guide to derive consensus based pre-hospital transfusion protocols and informed practice guidelines. In Part 2, the effect on patient outcomes will be systematically appraised.

| METHODS
The review was registered at Prospero (website: https://www.crd. york.ac.uk/prospero, identification number: CRD42018084658) and was conducted in accordance with PRISMA-Scr (Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Extension for Scoping Reviews) guidelines. 26 2.1 | Information sources, search strategy and study selection  ((blood transfusion) OR (blood AND transfusion))). Titles and abstracts were screened by two reviewers, and the full texts of all potentially eligible articles were retrieved.
All manuscripts discussing PHTRBC were eligible for inclusion, provided they were written in English, German, French or Dutch.
Since we sought to describe typical practice and identify commonly used indications and transfusion triggers, we did not limit the inclusion to studies allowing comparisons between PHTRBC patients and controls. Reviews and editorials were excluded. Eligibility was assessed independently in a blinded manner by two reviewers (ET & SB). Disagreements about manuscript eligibility were resolved by discussion within the investigator group. Reference lists of suitable articles were screened for additional relevant content.

| Data abstraction
We developed a standardised data-extraction sheet, which was refined after testing with the first 20 articles. In one case, further information was obtained after contacting the authors. 27 The setting and type of transport the EMS used (civilian or military, scene or interfacility), as well as the availability, frequency and volume of PHTRBC transfusions were extracted. Descriptions of problems that arose during PHTRBC are summarised in the text.
Transfusion criteria were recorded and classified as "major criteria" or "minor criteria," depending on whether only one criterion needed to be met to initiate PHTRBC or whether a combination of several criteria was required. Some groups published studies regarding pre-hospital transfusion with the same EMS more than once. In this case, only the most recent and best specified description of transfusion criteria was considered. Criteria are summarised in diagrams ( Figures 2 and 3).

| Selection of articles
The search in PubMed and Web of Science yielded 2172 hits after removal of duplicates. Of which, 2024 articles were excluded based on the title and/or abstract, because they did not discuss PHTRBC or were not original research (eg, editorials, reviews). Ninety articles were excluded after screening their full texts because they did not meet the inclusion criteria. For this review, 71 articles were included ( Figure 1).
In total, 57 articles discussed PHTRBC in civilian medical services.
Notably, five articles primarily dealt with a different study topic, but were included as they additionally provided valuable information regarding PHTRBC. 28-32 Table 1 describes studies' characteristics, listing study design, region and period in which the study took place, the primary goal, study group and control group, whether matching occurred, the number of subjects, the type of transport and, in the case of trauma patients, the mechanism of injury and Injury Severity Score.
We included 14 articles discussing PHTRBC in military medical services. Three articles reported pre-hospital transfusion as an additional topic, while primarily discussing another aspect of their study [33][34][35] (Table 1).

| Availability of PHTRBC in civilian services
Four articles report surveys on the availability of pre-hospital transfusion to civilian EMS. In France, 84% of 150 responding Mobile Emergency and Resuscitation Services were able to transfuse their patients with RBCs during the mission. However, the survey did not specify whether teams carried the blood products themselves, or had to order them to be delivered to the scene of injury. 36 Gillon

| Transfusion criteria in civilian services
The criteria used for PHTRBC in children were described by seven groups. Six of these required signs of hypoperfusion to persist after other fluids were administered. 41

| Transfusion criteria in military services
Articles were available about pre-hospital transfusion practices of the  is currently the best that is available to guide pre-hospital transfusion, and our review may aid clinicians to design or refine protocols regarding prehospital transfusion. The protocol for the Amsterdam HEMS has been based on the results of this review, and it is shown in Figure 4.
Our review suggests that there is a broad consensus that hypotension (most often defined as a systolic blood pressure < 90 mmHg), tachycardia (most often defined as a heart rate > 120 beats/min), clinical signs of poor perfusion (in particular changes in mental status) and the type of injury (in particular suspected or confirmed ongoing haemorrhage) are important transfusion triggers in the civilian setting.
Similar criteria were reported in military literature.
However, we believe that the decision to initiate or withhold transfusion should not only be based on "hard" criteria but should also involve the clinical judgement of pre-hospital healthcare providers.
For example, a heart rate > 120/min may be a stress response to pain, and changes in mental status are may be due to traumatic brain injury rather than haemorrhage. Such cases should not automatically prompt a transfusion, rather clinical judgement is needed to assess whether transfusion is the most appropriate therapy. In fact, clinical judgement was frequently allowed to supersede the protocol in published literature, and we explicitly agree with this recommendation.
Furthermore, uniform transfusion criteria may ultimately not be optimal: a 20-year-old healthy patient is likely able to tolerate a degree of hypotension and tachycardia that an 85-year-old is not. Ideally, individual transfusion triggers should be researched. Although not broadly established yet, we suggest studying portable blood-gas analysers to support well-informed RBC-transfusion decisions. Herein, blood-gas analysis quantifies the markers of anaerobic metabolism during haemorrhage (eg, lactate, pH and base excess) and allows for tracking RBC-transfusion effects, for example, by restoration of acid/base homeostasis or increase in haemoglobin concentration. In a recent study, we were able to demonstrate the feasibility of portable bloodgas analysis in the pre-hospital setting, including pre-hospital RBC transfusion. 88 However, further studies are required to study how por- This scoping review also highlights several challenges in the implementation of PHTRBC. EMS operators planning to implement prehospital transfusions need to beware of the pitfalls regarding storage and cooling logistics, return of unused units and traceability of blood products. Practical aspects of administration of blood products in the pre-hospital environment, such as the need for an additional intravenous line, drawing blood samples for cross-matching or warming of blood products before transfusion, can all be additional challenges in the prehospital setting in which severely injured patients need to be treated with limited resources and limited personnel.

| CONCLUSIONS
PHTRBC is increasingly used by civilian EMS, but a consensus on transfusion criteria is lacking. This scoping review summarises current practice and may provide the basis for consensus-based pre-hospital transfusion protocols. Some studies into the effects of PHTRBC on outcomes have been performed, and an overview of this data will be presented in the second part of this systematic review.