Tools to predict acute traumatic coagulopathy in the pre-hospital setting: a review of the literature

Introduction: Recognising acute traumatic coagulopathy (ATC) poses a significant challenge to improving survival in emergency care. Paramedics are in a prime position to identify ATC in pre-hospital major trauma and initiate appropriate coagulopathy management. Method: A database literature review was conducted using Scopus, CINAHL and MEDLINE. Results: Two themes were identified from four studies: prediction tools, and point-of-care testing. Prediction tools identified key common ATC markers in the pre-hospital setting, including: systolic blood pressure, reduced Glasgow Coma Score and trauma to the chest, abdomen and pelvis. Point-of-care testing was found to have limited value. Conclusion: Future research needs to explore paramedics using prediction tools in identifying ATC, which could alert hospitals to prepare for blood products for damage control resuscitation.

that might establish future research. Throughout October 2019, the databases MEDLINE and CINAHL (accessed via EBSCOhost) and Scopus were searched for relevant articles. For keywords, a search filter derived by Olaussen et al. (2017) was used, consisting of Ambulances OR Emergency Medical Technicians OR Air Ambulances OR paramedic* OR ems OR emt OR prehospital OR pre-hospital OR first responder* OR emergency medical technicians OR emergency services OR Ambulance* OR HEMS OR field triage. Medical subject headings identified topic-specific keywords for trauma, including: injuries OR wounds and injuries; and coagulopathy.
Inclusion and exclusion criteria were developed based on established systematic review methodology, focusing on publication quality, study design, population, intervention, comparison and outcome (Eriksen & Frandsen, 2018;Meline, 2006) (see Table 1). Specific patient populations, including children, people with traumatic brain injury and pregnant women, were excluded in acknowledgement that the mechanism of coagulopathy in these groups may be different to that of the general adult population (Attard et al., 2014;Conti et al., 2016;Leeper et al., 2018;Moon and Sappenfield, 2016).
A PRISMA method was used to identify and report on eligible articles (Moher et al., 2009) (see Figure 1). The appropriate Critical Appraisals Skills Programme (CASP) (2017) checklist was used to assess and appraise the quality of the research. Themes were then identified from the results of each study jointly by the authors in a concept matrix.

Introduction
Traumatic injury causes approximately 2 million deaths annually and is recognised as the leading cause of preventable mortality in people aged 44 and under (Trauma Audit and Research Network, 2017). It is thought that approximately 33% of major trauma patients develop coagulopathy, 76% of which will have acute traumatic coagulopathy (ATC) prior to hospital arrival (Cohen & Christie, 2017). ATC occurs at the point of trauma, where tissue damage is sufficient that inflammation, fibrinolysis (clot breakdown) and systemic hypoperfusion develop into a failure of the coagulation system to sustain adequate haemostasis (Davenport, 2013). It is thought to have a mortality rate of 50% (Brohi & Eaglestone, 2017).
While paramedics are trained to triage major trauma with the aid of major trauma triage tools (Thompson et al., 2017), identifying major trauma remains a complex process, particularly in patients over the age of 55 years (Durham, 2017). In addition, ATC is not considered in existing pre-hospital triage tools, despite guidance from the National Institute of Health and Care Excellence (NICE) that early recognition of ATC is a key goal of trauma management (NICE, 2016). Timely treatment of ATC is vital in improving patient outcomes (Cohen & Christie, 2017). If paramedics could reliably identify patients at risk of ATC, and implement specific care pathways (such as accelerated blood transfusion activation), this might improve patient outcomes from trauma complicated by ATC.
This literature review aims to identify if pre-hospital screening tools can recognise ATC.

Method
An integrative literature review methodology was decided as a pragmatic approach in order to explore key themes Population: any patient not in a pre-hospital setting, such as in a hospital or laboratory, and studies that focused on animals Intervention: ATC assessment initiated on arrival at hospital in the emergency department Comparison: non-trauma presentations, pregnancy, children or traumatic brain injury Outcome: coagulopathy not the primary aim signs and type of injury, Mitra et al. (2011) formulated the Coagulopathy of Severe Trauma (COAST) score, which was further validated within the study with a prospective comparison cohort of 1225 patients. The authors concluded that specific vital signs were more likely to be apparent in ATC, forming a score ranging from 0 to 7 (see Table 3). A score ≥ 3 was found to have 60% sensitivity and 96.4% specificity. Interestingly, the specificity of the COAST score rose as the score increased (100% when ≥ 5), and sensitivity increased as the COAST score decreased (100% ≥ 0). Mitra et al. (2011) elected not to include other vital signs, including the Glasgow Coma Score (GCS) (p = 0.03) as part of COAST, possibly due to the need to conduct an additional value, which could complicate the scoring process. Further criticisms of the study are the decision to exclude chest trauma as part of the score (n = 119, p = 0.001), and the lack of penetrating trauma in either comparison group.
further excluded due to ongoing research with unpublished results (Tonglet et al., 2019), a focus on assessing variation in pre-and in-hospital coagulation (Floccard et al., 2012;Theusinger et al., 2015), retrospective exploration of traumatic induced coagulopathy from medical treatment rather than ATC (David et al., 2017), using decision tools for blood transfusion rather than ATC recognition  and ATC point-of-care testing commencing in hospital (Gauss et al., 2014;Goodman et al., 2015;Mistral et al., 2017). Four articles were found suitable for discussion on themes identified as prediction tools and point-of-care testing (see Table 2).

Discussion
Pre-hospital prediction tools Mitra et al. (2011) sought to develop a tool that can identify ATC in the pre-hospital setting. Based on a retrospective analysis of 1680 trauma patients' pre-hospital vital  Indeed, in a similar study, Peltan et al. (2016) compared the COAST score with their derived Prediction of Acute Coagulopathy of Trauma (PACT) score, finding that the PACT was significantly better at determining patients at risk of ATC than COAST. However, a PACT score had only a moderate sensitivity and specificity (73.1% and 73.8% respectively). Moreover, the PACT score includes a complex scoring system, requiring a calculation of the GCS and the pre-hospital shock index, comprised of heart rate divided by systolic blood pressure, cardiopulmonary resuscitation and advanced airway, where points per variable are not equally weighted. It is unclear why the authors decided to include the shock index when systolic blood pressure was equally significant (p < 0.001) and easier to calculate. Potentially, it was due to the median initial pre-hospital systolic blood pressure being greater than 100 mmHg (119 mmHg) despite confirmed coagulopathy, in contradiction to existing literature. This observational relationship could be explored further in the pre-hospital setting, as it could be that a cohort of patients could be suffering from ATC despite being normotensive.
Additionally, Peltan et al. (2016) did not assess pre-hospital temperature, which may have been different and more relevant when comparing with COAST, thus invalidating their claim that PACT is more discriminative in the pre-hospital setting. Further research would need to validate if PACT is suitable in the pre-hospital environment. Tonglet et al. (2014) investigated pre-hospital application of the Trauma Induced Coagulopathy Clinical Score (TICCS). TICCS comprises three components with a score range of 0-18 (see Table 4). Tonglet et al. (2014) identified that paramedics and doctors who scored TICCS ≥ 10 significantly determined severe ATC patients requiring damage control resuscitation (sensitivity 100% CI: 53.9-100, specificity 95.9% CI: 88.2-99.2, p = 0.0011), a treatment strategy consisting of haemorrhage cessation, permissive hypotension, avoiding unnecessary fluid administration and targeting coagulopathy. The small sample (n = 82) -in which only eight required damage control resuscitation -as well as poor heterogeneity (78.1% of the group were male) and lack Table 3. Coagulopathy of Severe Trauma (COAST) score (Mitra et al., 2011).

Variable
Value Score of reporting the mechanism of injury mean these findings may not be generalisable. Considering that TICCS requires assessment of trauma severity and extent of tissue injury, this type of information would have been useful in assessing the reliability of the tool. Although the three studies share different findings, some common themes within the results and tools emerge. Firstly, vital signs such as GCS and systolic blood pressure, and trauma to the chest, abdomen and pelvis, are key pre-hospital readings that could determine ATC. Unsurprisingly, these are all within the first two steps of the UK pre-hospital major trauma triage tool clinical guidelines (Joint Royal Colleges Ambulance Liaison Committee (JRCALC), 2019) (see Table 5).
Nevertheless, the linearity of the pre-hospital major trauma tool could be omitting patients with ATC, particularly in the presence of polytrauma, where ATC is more likely (Davenport, 2013). Moreover, the existing pre-hospital major trauma tool renders it difficult for pre-hospital clinicians to express through the tool which injury may be a more significant priority until arrival at the MTC. Yet, in identifying ATC at the earliest opportunity, a paramedic could highlight the need to accelerate blood transfusion triggers at hospital (Brohi et al., 2019). This concept is not new among physician-led pre-hospital teams, where studies by Weaver et al. (2016) and Reed et al. (2016) have illustrated how a pre-hospital 'code red' protocol for UK major trauma patients to receive blood transfusion on arrival at hospital improves treatment timeliness and survival. Importantly, physician-led resources remain limited in the pre-hospital setting, where timeliness to definitive care, such as an MTC, remains the priority in major trauma and ATC (Cohen & Christie, 2017;Thompson et al., 2017).
However, 'code red' already comprises similar criteria to TICCS, including systolic blood pressure < 90 mmHg and suspected or active bleeding (Weaver et al., 2016), where TICCS has been further evidenced to reduce mortality in trauma patients when pre-hospital clinicians alert hospitals, and to improve timely preparation of blood products . Therefore, future studies should explore whether paramedics can utilise screening tools to highlight individuals at risk of ATC, and implement specific care pathways such as 'code red', improving care and patient outcomes from ATC.

Point-of-care testing
The second theme identified the possible use of a handheld coagulometer in detecting coagulopathy at the point of pre-hospital care. Beynon et al. (2015) used a handheld point-of-care device (Coaguchek®) to compare the prehospital international normalised ratio (INR) with hospital laboratory values, followed by a brief survey of the pre-hospital physician. Although 19 of the samples were specifically recognised as trauma, Beynon et al. (2015) identified that all patients with coagulopathy had an INR > 1.3 (sensitivity 100%, specificity 98.7%), where British Paramedic Journal 5 (3) Furthermore, point-of-care devices applied in the emergency department to identify ATC have been shown to have weak correlation with laboratory values (Goodman et al., 2015;Mistral et al., 2017). While NICE (2016) recommends early INR monitoring in ATC, currently this practice remains poorly evidenced and has apparently limited use in improving ATC care or referral in the pre-hospital realm.

Limitations
The paucity of literature, and lack of repeated studies using the same tool within the pre-hospital setting, render very limited evidence. One study investigated point-of-care, with a very small sample that limits any comparison or accurate conclusion. Lastly, studies focused on various countries that may not be representative of the UK population or paramedic practice.

Conclusion
This literature review sought to establish if screening tools could identify ATC in the pre-hospital setting, which could be utilised by paramedics. While studies are few, with various small cohorts across the world and of point-of-care INR significantly correlated with laboratory INR (0.68, p < 0.0001). A median time gain of 69 minutes in using the device to detect coagulopathy was highlighted as beneficial for rapid recognition and treatment escalation. However, 42% of pre-hospital physicians felt that the value of point-of-care INR assessment was low, and no physicians would consider any further treatment at the scene.
Furthermore, the point-of-care assessment relies on a blood sample within two minutes of venepuncture, and cannot function in temperatures lower than 5°C or higher than 35°C. Certainly in the UK, such a device would be unreliable during cold winter months. Another confounding factor would be if an individual is on anti-coagulation medication, requiring enhanced knowledge of the threshold of INR, the role of anticoagulant reversal agents and when delay or aggressive treatment is appropriate (Mullins et al., 2018). Such decisions require expertise currently beyond the scope of a pre-hospital paramedic. Additionally, utilisation of such a device in a difficult situation such as major trauma may not be feasible, especially if multiple treatments or rapid extrication are required, where acquiring a blood sample in less than two minutes may not be a priority.  Table 5. UK pre-hospital major trauma triage tool -steps 1 and 2.
Step 1: Physiological • GCS < 14 • Systolic blood pressure < 90 mmHg Step 2 limited evidence, there is potential for tools such as the COAST score or TICCS to predict ATC, and initiate a care plan such as a code red for blood products. The authors recommend that if this were to be investigated further, it should be incorporated within the pre-hospital major trauma triage tool, such that if multiple steps are flagged between steps 1 and 2, an ATC protocol is initiated -or be utilised by experienced critical care paramedics as part of a holistic management plan.

Key points
• Recognition and management of acute traumatic coagulopathy (ATC) need to improve in the UK. Paramedics are in an optimal position to identify ATC, to initiate damage control resuscitation and to alert receiving hospitals to prepare blood products. • Pre-hospital prediction tools such as the COAST score and TICCS show promise in identifying ATC early, and could be used to initiate an early blood transfusion protocol at the awaiting Major Trauma Centre. This needs to be researched further to see if a particular score or threshold is optimal. • Future research could investigate if ATC screening tools can be adapted into the pre-hospital major trauma tool to flag suspected ATC. • Point-of-care testing is unlikely to be beneficial in the pre-hospital setting.

Author contributions
SR co-wrote the Method section, wrote the Results and Discussion sections and did the overall review and writing of the article. JK wrote the Introduction section, co-wrote the Method section, co-identified themes and reviewed the Discussion section. SR acts as the guarantor for this article.