Traumatic coagulopathy and massive transfusion: improving outcomes and saving blood

Article history

The research reported in this issue of the journal was funded by PGfAR as project number RP-PG-0407-10036 . The contractual start date was in July 2008. The final report began editorial review in August 2014 and was accepted for publication in June 2017. As the funder, the PGfAR programme agreed the research questions and study designs in advance with the investigators. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The PGfAR editors and production house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Karim Brohi reports that TEM Innovations GmbH (ROTEM^®) provided unrestricted support to the programme in the form of equipment and reagents for the observational study [National Institute for Health Research (NIHR) portfolio ID 5637].

Copyright statement

© Queen’s Printer and Controller of HMSO 2017. This work was produced by Brohi et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Introduction

What we knew about trauma-induced coagulopathy in 2008

In 2003, the discovery of a clotting disorder present immediately on arrival in the emergency department prompted a change in our understanding of the disease and the way it was treated. This acute traumatic coagulopathy (ATC) was first described in our study with London’s Air Ambulance. We looked at > 1000 helicopter patients and found that one in four patients arrived with ATC and that, if present, one in four patients would die. 5 By the start of this study several other groups around the world had confirmed the existence of ATC and we had begun to characterise it and explore its pathophysiology. 6,7

Trauma-induced coagulopathy (TIC; all different possible clotting disorders that a trauma patient could develop) could therefore take multiple forms at different times in their clinical course and several different coagulopathies might exist at the same time (Figure 1). It was becoming clear that TIC was not a single disease entity and therefore probably could not be managed in the same way at all times. 8

FIGURE 1.

Proposed model for the mechanisms and drivers of TIC in 2008. Adapted from Hess JR, Brohi K, Dutton RP, Hauser CJ, Holcomb JB, Kluger Y, et al. The coagulopathy of trauma: a review of mechanisms. J Trauma, vol. 65, iss. 4, pp. 748–54. 2008. 8 URL: http://journals.lww.com/jtrauma/pages/default.aspx. Promotional and commercial use of the material in print, digital or mobile device format is prohibited without the permission from the publisher Wolters Kluwer. Please contact healthpermissions@wolterskluwer.com for further information.

The identification of ATC provided a potential therapeutic focus for new management strategies and therapeutics. This was first approached in retrospective studies, with patients who had received higher doses of plasma during their haemorrhage resuscitation appearing to have significantly better outcomes than those given standard resuscitation regimens. 9 Patients given less crystalloid also seemed to do better. 10 In 2007 a new paradigm of resuscitation was postulated, damage control resuscitation (DCR), a strategy specifically targeting ATC. 11 The evidence for this position paper was weak, but the collective approach had four central tenets: stop bleeding, permissive hypotension until control, avoid fluids and target coagulopathy. The DCR approach was almost the complete opposite of the then current approach to haemorrhage resuscitation. Much more work needed to be carried out to show that this strategy was effective and did not expose patients to unnecessary risks and waste precious blood resources.

What has changed in the management of bleeding trauma patients?

The Royal London Hospital is one of the busiest major trauma centres in the world, with nearly 3000 trauma team activations each year, and treats some of the most severely injured patients in the country. In 2008, patients activating our traditional massive transfusion protocol had a 50% mortality rate. 12

In partnership with the Barts Centre for Trauma Sciences at Queen Mary University of London we have iteratively implemented our research findings into practice with a multispecialty team of clinicians and transfusion specialists. By the end of our programme grant in 2014 we had more than halved the mortality of this group of patients to 26% (Figure 2). We had also halved the number of blood products that they required and almost halved their critical care and hospital stays.

FIGURE 2.

Outcomes of ‘Code Red’ patients at the Royal London Hospital 2007–14. Mortality fell from 57% to 26% and 24-hour PRBC requirements halved from 12 to 6 units.

We have a much deeper understanding of TIC in all its forms and a better understanding of the role of clotting therapy in the management of these conditions. We have new therapeutic agents and others in clinical trial pipelines. These changes have been incorporated in new major haemorrhage ‘Code Red’ protocols, which have been disseminated nationally and internationally and incorporated into clinical guidelines around the world.

In the following sections we describe the programme of work and the central findings that have led to these changes in practice.

What we did

We performed multimodal parallel studies to achieve the overall objectives (Figure 3). Although these studies were discrete, the conduct and results of each informed aspects of the others.

FIGURE 3.

Inter-relationships of the different aims of the research programme. MCE, mass casualty event; POC, point of care; RCT, randomised controlled trial; ROTEM, rotational thromboelastometry.

The core studies are described in the following sections.

The epidemiology of trauma haemorrhage in England and Wales

Incidence

Our national epidemiology study (see Appendix 2) identified 442 patients with major haemorrhage (requiring at least 4 units of PRBCs for resuscitation) and 146 cases of massive haemorrhage (requiring at least 10 units of PRBCs) at 22 hospitals (see Appendix 1). 18

Nationally (for England and Wales), this gives an estimated incidence of 4700 trauma patients a year with major haemorrhage, 1300 of whom will have massive haemorrhage. We unexpectedly found that the incidence was much higher in elderly patients (Figure 4), with more than double the incidence of trauma haemorrhage than in the general population (196 per million vs. 83 per million).

FIGURE 4.

National incidence of major and massive haemorrhage due to trauma: (a) distribution of the 22 participating UK hospitals; and (b) population-based incidence by age group. Source: data from Stanworth et al. 18 Reproduced with permission from Stanworth SJ, Davenport R, Curry N, Seeney F, Eaglestone S, Edwards A, et al. Mortality from trauma haemorrhage and opportunities for improvement in transfusion practice. Br J Surg 2016;103:357–6518 © 2016 BJS Society Ltd Published by John Wiley & Sons Ltd.

Mortality

Mortality in our study was high. Nearly half of trauma patients with massive haemorrhage died (and one-third of major haemorrhage patients). This extrapolates to > 1550 people dying of major haemorrhage each year in England and Wales (585 dying of massive haemorrhage). 18

Deaths occurred early, with over half occurring within the first 24 hours of injury and half of these occurring within the first 4 hours of injury (Figure 5). It is clear that improvements in practice have to occur in this critical post-injury window if outcomes from trauma haemorrhage are to improve.

FIGURE 5.

National mortality associated with trauma haemorrhage. One-year mortality approached 50% for patients with massive haemorrhage. Half of all deaths occurred within the first 24 hours and half of these deaths occurred within the first 4 hours of arrival. Source: data from Stanworth et al. 18 Reproduced with permission from Stanworth SJ, Davenport R, Curry N, Seeney F, Eaglestone S, Edwards A, et al. Mortality from trauma haemorrhage and opportunities for improvement in transfusion practice. Br J Surg 2016;103:357–6518 © 2016 BJS Society Ltd Published by John Wiley & Sons Ltd.

Resource utilisation and health-care costs

On average, major haemorrhage patients received 11 units of blood products (PRBCS, plasma, platelets or cryoprecipitate) and massive haemorrhage patients received 26 blood product transfusions in the first 24 hours. Patients spent just under 3 hours in the emergency department and over half required urgent surgery. 19

Over 80% of trauma patients with major haemorrhage required critical care admission and they spent on average 7 days in critical care and 3 days on a ventilator. They spent on average an additional 20 days in hospital.

Intensive care unit (ICU) costs accounted for the largest proportion of health-care costs, at £4600–9500 per patient. The next largest cost utilisation areas were operating theatres and other ward areas. Blood product transfusions cost approximately £2300 per major haemorrhage patient (£4200 for massive haemorrhage). 19

In total, the estimated cost of treating a major haemorrhage patient was £20,600 (£24,000 for massive haemorrhage). Extrapolating nationally this gives an estimated cost of major haemorrhage after trauma of approximately £85M annually. One-third of all costs are accounted for by elderly trauma patients. 19

Evolution of Code Red major haemorrhage protocols

Before 2008, the Royal London Hospital major trauma centre had a massive transfusion protocol in place, based on traditional guidelines such as the British Commitee for Standards in Haematology4 and College of American Pathologists22 recommendations. These massive transfusion protocols were activated for patients likely to require large volumes of red cells, although without clear criteria for activation. Furthermore, they were primarily targeted at the treatment of late complications of large-volume blood and fluid administration, most specifically dilutional coagulopathy. The protocol was associated with a wide variation in the amount of blood products (e.g. plasma) given to patients. There was also a significant amount of waste, especially of platelets. The mortality rate in patients who activated the protocol was 57%. 12

Activation was tightly controlled so that wastage would be reduced. We reviewed the performance of massive transfusion prediction tools and used data from nearly 5700 trauma patients to develop our own prediction algorithm. 23 We determined that there was no specific prediction tool that performs better than another and no algorithm suitable for early (even pre-hospital) activation of a protocol. Additionally, we recognised that activating haemorrhage protocols only on the basis of a specific number of PRBCs transfused precludes improvements in care that reduce PRBC requirements in the future. 23

In late 2008, coinciding with the start of the programme grant, the trauma centre instituted a new major haemorrhage protocol for trauma patients. This Code Red protocol aimed to identify early patients who were bleeding and prevent (rather than treat) dilutional coagulopathy. Treatment followed the principles of DCR based on the limited evidence available at the time. As our research (and that of others) progressed we iteratively updated the protocol to include new research findings and practices.

Immediately after institution we found that we reduced the wastage of products but there was no immediate impact on survival. 12 We progressed, limiting crystalloids, improving the delivery of products and introducing new therapeutics and evidence-based point-of-care coagulation assessments, and we saw a stepwise reduction in mortality. By the end of the programme grant mortality had reduced to 26%, with an associated reduction in all blood product administration and waste, as well as reductions in critical care and hospital stays.

The Royal London Hospital’s Code Red protocol (Figure 6) has been adopted by almost all major trauma centres in the UK (modified for local practice) and has been disseminated internationally. Further refinements continue to occur as more research results become available from this programme of work and the work of others.

FIGURE 6.

Evolution of the Royal London Hospital major trauma centre Code Red protocols: (a) 2008; and (b) 2011. ABG, arterial blood gas; APTT, activated partial thromboplastin time; BP, blood pressure; Ca⁺⁺, calcium; CA5, clot amplitude at 5 minutes; CT, clotting time; DCS, damage control surgery; ED, emergency department; FBC, full blood count; G&S, Group & Save; HEMS, Helicopter Emergency Medical Services; IV, intravenous; ML, Maximum Lysis; O NEG, O negative; O POS, O positive; POC, point of care; PT, prothrombin time; RBC, red blood cell; ROTEM, rotational thromboelastometry; SpR, specialist registrar.

We have subsequently revised our Code Red protocol again based on our subsequent work using point-of-care functional coagulation assessment. In the following sections we detail our work on using these point-of-care devices to diagnose coagulopathy, predict major haemorrhage and individualise therapy to optimise blood product utilisation and patient outcomes. The protocol is currently being implemented and will be evaluated in a randomised controlled trial as part of a subsequent European Union FP7 programme of work. 24

Rapid diagnosis and characterisation of acute traumatic coagulopathy

Aim 3 of the programme grant was specifically focused on the early identification of coagulopathy and, specifically, ATC. By 2008 ATC had been identified and defined as a prolongation of the laboratory prothrombin time (PT) and was thought to be caused by the systemic activation of anticoagulant factors. 25 Data from our ACIT study (see Appendices 3–5) showed that the laboratory PT was not available in a clinically useful time frame, on average becoming available to the treating trauma team only after 90 minutes. 20 We also investigated a hand-held PT device, used for home warfarin monitoring. These devices are not, however, calibrated for low haematocrit levels found in bleeding patients and we found they significantly under-read compared with the laboratory value. 20

We wished to evaluate the potential for a point-of-care functional coagulation analyser – rotational thromboelastometry [ROTEM^® (Tem International GmbH, Munich, Germany)] – to identify patients with ATC. ROTEM measures the strength of a blood clot as it forms over time in a small well of the device. Different activators and inhibitors can be added to assess different parts of the coagulation system.

In our first study we analysed the ROTEM traces of 300 trauma patients. 20 For the first time we were able to fully characterise the functional characteristics of ATC. We found that the primary functional disorder of ATC is a loss of clot strength [reduction in maximum clot firmness (MCF)]. In contrast, prolongation of clot activation (the PT) is a mild and rather late component of the coagulopathy.

Although measurement of the MCF takes half an hour or more, we were able to identify a ROTEM parameter available within 5 minutes that correlated closely with the MCF (Figure 7). The clot amplitude at 5 minutes (CA5) was able to accurately identify ATC and to predict the need for massive transfusion. It also showed an excellent negative predictive value, suggesting that it could also be used to curtail a major haemorrhage protocol to reduce waste or overtransfusion. We have since validated our finding in a later cohort and across the six INTRN partner sites in a study of 808 ACIT patients. 26 The CA5 remains the fastest and most effective point-of-care diagnostic marker of ATC.

FIGURE 7.

Functional characterisation of ATC: (a) averaged ROTEM curves for normal and ATC patients; (b) identification of CA5 of ≤ 35 mm to identify patients with ATC; (c) combined effects of trauma (injury severity score > 15) and shock (base deficit > 6) on the ROTEM trace; and (d) histogram demonstrating the distribution of CA5 among normal and ATC patients. The dotted line represents the new threshold for ATC (CA5 ≤ 35 mm). Source: reproduced with permission from Davenport R, Manson J, De’Ath H, Platton S, Coates A, Allard S, Hart D, Pearse R, Pasi KJ, MacCallum P, Stanworth S, Brohi K. Functional definition and characterisation of acute traumatic coagulopathy. Critical Care Medicine, vol. 39, issue 12, pp. 2652–8, URL: http://journals.lww.com/ccmjournal/pages/default.aspx. 20 Promotional and commercial use of the material in print, digital or mobile device format is prohibited without the permission from the publisher Wolters Kluwer. Please contact healthpermissions@wolterskluwer.com for further information.

Resuscitation of trauma haemorrhage: effectiveness of high-dose plasma therapy

Early approaches to the treatment of TIC focused on treatment with high-dose FFP. This was based on early assumptions about the mechanism of TIC: at this time it was thought to be caused by a loss or consumption of procoagulant factors. It also appeared to be supported by retrospective military and civilian studies27 that showed improved outcomes when patients were given higher doses of FFP than in standard therapy. 9 The best outcomes appeared to be obtained when 1 unit of FFP was transfused for each 1 or 2 units of PRBCs.

There were significant concerns about this approach, however, which involved transfusion of four to five times the amount of emergency FFP than patients usually received. This would have a major resource impact on blood services worldwide and expose patients to a lot more blood products, with the associated potential risks of immunosuppression and infection. No group had ever been able to study the effect of FFP transfusions on coagulation during the resuscitation of a bleeding trauma patient.

We examined the coagulation system response to high-dose FFP transfusion (Figure 8). 28 We were able to show that giving FFP in high doses certainly appeared to reduce the dilutional coagulopathy that occurred as patients continued to bleed and receive PRBC resuscitation. With low-dose FFP, plasma levels of procoagulant factors such as factors II and X fell to low levels after transfusion of 12 PRBC units, whereas, with high-dose FFP, levels were somewhat protected.

FIGURE 8.

Progression of coagulopathy during bleeding and resuscitation: (a) overall incidence of coagulopathy during haemorrhage and resuscitation; and (b) reduced loss of activity of some procoagulant factors with high-dose plasma administration. FII, factor II; FX, factor X; PTr, prothrombin ratio; U, units. Source: reproduced with permission from Intensive Care Medicine, Damage control resuscitation using blood component therapy in standard doses has a limited effect on coagulopathy during trauma haemorrhage, vol. 41, 2015, pp. 239–47, Khan S, Davenport R, Raza I, Glasgow S, De’Ath HD, Johansson PI, Curry N, Stanworth S, Gaarder C, Brohi K, © Springer-Verlag Berlin Heidelberg and ESICM 2014 with permission of Springer. 26

However, these plasma regimens did not treat an existing coagulopathy. They were protective against further deterioration, but trauma patients presenting with ATC never corrected their coagulation parameters to normal until bleeding had stopped and compensatory mechanisms had taken over. Although high-dose plasma was effective in avoiding dilutional coagulopathy, there was still an opportunity to improve outcomes with focused, targeted anti-ATC therapy early in the clinical course.

We also reviewed transfusion practice nationally (Aim 2) as part of our prospective cohort study. 18 We found that there were a number of opportunities for improvement in practice that could directly translate to improved outcomes. We particularly looked at delays in the delivery of FFP and other blood products (Figure 9) and times when the unavailability of products would leave patients exposed to a risk of dilution with PRBCs or intravenous fluid replacement (e.g. crystalloids).

FIGURE 9.

National patterns of transfusion therapy in trauma haemorrhage: (a) late FFP availability results in < 50% of patients achieving a FFP : PRBC ratio of ≥ 1 : 2 by 4 hours after admission; (b) the FFP : PRBC ratio varies and balanced resuscitation is not delivered consistently during resuscitation; and (c) mortality by hour showing the concentration of deaths when FFP is not available or administered. Source: data from Stanworth et al. 18 Reproduced with permission from Stanworth SJ, Davenport R, Curry N, Seeney F, Eaglestone S, Edwards A, et al. Mortality from trauma haemorrhage and opportunities for improvement in transfusion practice. Br J Surg 2016;103:357–6518 © 2016 BJS Society Ltd Published by John Wiley & Sons Ltd.

Bleeding trauma patients tended to receive PRBCs first and there was a delay of 1–3 hours before receiving the first doses of FFP. During resuscitation, there was variation in the delivery of FFP leading to variation in the ratio delivered. There were times when no blood products were available. A large proportion of deaths occurred in the early phase of care when plasma and other blood product delivery was absent or inefficient. There are therefore important opportunities for practice improvement on a national level based on our current understanding of the optimal management of trauma haemorrhage.

Acute trauma coagulopathy: fibrinogen loss and replacement

We identified reduced clot strength as a central functional component of ATC early in the course of this programme. 20 We investigated the cause of this and rapidly turned our attention to fibrinogen, the substrate for all blood clot formation. 29 We found that trauma patients with ATC arrive with a dramatically reduced level of fibrinogen (Figure 10) – around half of the normal level – and that this level decreases rapidly as bleeding continues. When no fibrinogen-containing blood products were administered, fibrinogen levels were near zero after transfusion of 8 units of PRBCs. Even with high-dose FFP (which contains some fibrinogen), fibrinogen levels continued to decline during bleeding. Only when fibrinogen was replaced with cryoprecipitate were fibrinogen levels maintained. Much closer attention is paid now to fibrinogen levels during haemorrhage and cryoprecipitate usage has increased worldwide over the last few years.

FIGURE 10.

Fibrinogen levels during trauma haemorrhage: (a) fibrinogen levels are low on admission and decrease dramatically in the absence of any fibrinogen replacement (FFP or platelets or cryoprecipitate); and (b) cryoprecipitate as administered in the Code Red protocol maintains but does not correct fibrinogen levels. Fg, fibrinogen. Source: data from Khan et al. 26

Although cryoprecipitate administered in standard doses can maintain fibrinogen levels during bleeding, it does not correct fibrinogen to normal levels. 29 Trauma patients who arrive with ATC with insufficient fibrinogen substrate to form clots never achieve normal levels of fibrinogen. We therefore conducted ex vivo spiking experiments to determine how much fibrinogen would be necessary to normalise levels. In parallel with the programme grant we then conducted a pilot randomised controlled trial (CRYOSTAT) of early, high-dose cryoprecipitate in bleeding trauma patients. 30 We wanted to see whether or not we could deliver high-dose cryoprecipitate quickly to trauma patients and, in doing so, normalise fibrinogen levels and hence potentially improve outcomes.

The trial was conducted between the Royal London Hospital and the John Radcliffe Hospital, Oxford. 30 We also recruited patients at the Camp Bastion Military Hospital in Afghanistan. We recruited 40 patients and showed that we could deliver high-dose cryoprecipitate (equivalent to 4 g of fibrinogen supplementation) within 90 minutes to bleeding trauma patients who activated the major haemorrhage protocol. Patients in the early high-dose cryoprecipitate arm maintained their fibrinogen levels throughout the bleeding phase (Figure 11). Furthermore, these patients had a much lower mortality rate than patients in the control arm (less than half of our current Code Red mortality rate), although this was not statistically significant in this small pilot trial.

FIGURE 11.

CRYOSTAT study results: (a) early high-dose cryoprecipitate is able to correct fibrinogen levels during trauma haemorrhage; and (b) survival curves for the high-dose cryoprecipitate (green) and standard therapy (black) groups (not significant). Source: data from Curry et al. 29

We are currently recruiting to the Phase III CRYOSTAT2 trial to determine whether or not early high-dose fibrinogen replacement significantly improves outcomes in trauma haemorrhage.

Acute trauma coagulopathy: fibrinolysis and tranexamic acid

In earlier work we had identified excessive clot breakdown (hyperfibrinolysis) as a key characteristic of ATC. 25 However, there was conflicting evidence on hyperfibrinolysis. In studies using functional coagulation tests such as ROTEM, hyperfibrinolysis was very rare and almost uniformly fatal. In a large randomised controlled trial [Clinical Randomisation of an Antifibrinolytic in Significant Haemorrhage (CRASH-2)] by our close collaborators at the London School of Hygiene & Tropical Medicine, a relatively low dose of an antifibrinolytic agent, tranexamic acid, led to significant improvements in survival. 30 This seeming paradox, together with a potential risk of thromboembolism, had restricted the implementation and uptake of tranexamic acid by the NHS and internationally. Some bodies mandated the use of functional clotting assays to guide therapy, whereas others refused to adopt this treatment.

We explored fibrinolysis in trauma patients by examining the fibrinolytic pathway in our ACIT cohort. 31 We found that > 60% of trauma patients have extreme fibrinolysis even in the first minutes after injury (Figure 12). More importantly, this lytic activity is not detectable in viscoelastic assays until endogenous antifibrinolytic proteins are exhausted. This fibrinolysis in trauma patients is common and occult and, therefore, viscoelastic assays should not be used as criteria for antifibrinolytic therapy.

FIGURE 12.

Thromboelastometry is insensitive to fibrinolysis in TIC: (a) maximum lysis occurs only at extremely high levels of plasmin production; and (b) 60% of patients have occult lysis evidenced by PAP levels of > 10× normal and which is not visible on thromboelastometry. PAP, plasmin–antiplasmin. Source: data from Raza et al. 31

We incorporated tranexamic acid into our Code Red protocol in 2011. In the first civilian study of tranexamic acid in a developed civilian trauma system setting, we showed improved outcomes including survival (Figure 13). 32 Patients presenting in shock had the greatest benefit, with a fourfold reduction in the relative risk of death. We also showed that early tranexamic acid administration not only treats hyperfibrinolysis but also protects clot strength throughout trauma haemorrhage.

FIGURE 13.

Tranexamic acid reduces mortality and multiple organ failure (MOF) in trauma patients presenting with shock (which is defined by base deficit of > 6 mmol/l). MI, myocardial infarction; VTE, venous thromboembolism. Source: reproduced with permission from Cole E, Davenport R, Willett K, Brohi K. Tranexamic acid use in severely injured civilian patients and the effects on outcomes: a prospective cohort study. Ann Surg vol. 261, iss. 2, pp. 390–4. 32 2015. URL: http://journals.lww.com/annalsofsurgery/pages/default.aspx. Promotional and commercial use of the material in print, digital or mobile device format is prohibited without the permission from the publisher Wolters Kluwer. Please contact healthpermissions@wolterskluwer.com for further information.

This work has supported the uptake of tranexamic acid into clinical guidelines and major haemorrhage protocols. It is now widely used across the country and is given by pre-hospital teams to patients suspected to have haemorrhage before arrival at trauma centres.

The evolved concepts of damage control resuscitation and trauma-induced coagulopathy

In overview, we have used the principles of DCR as a starting point and refined its principles, practice and treatment, iteratively generating evidence and implementing it into practice (Figure 14).

FIGURE 14.

Interventions associated with Code Red protocol evolution at the Royal London Hospital major trauma centre. Fg, fibrinogen; MHP, major haemorrhage protocol; NIHR, National Institute for Health Research; PGfAR, Programme Grant for Applied Research; TXA, tranexamic acid.

Our new concept of TIC is simpler, with a more mechanistic basis. Essentially, there are two main coagulopathies: the endogenous ATC and a subsequent dilution/consumption phenotype that progresses during haemorrhage and resuscitation.

We now believe that the central features of ATC are fibrinogen loss and hyperfibrinolysis. Tranexamic acid has been introduced into practice for hyperfibrinolysis and early fibrinogen replacement is awaiting a next-phase clinical trial.

Dilutional/consumption coagulopathy is now managed by avoiding crystalloids, even in the pre-hospital phase of care; streamlining the delivery of transfusion products; and delivering plasma in a 1 : 1 ratio with PRBCs.

We believe that we have demonstrated that robust implementation of these principles and practices can have a dramatic impact on outcomes for trauma patients. Nationally, trauma haemorrhage mortality remains very high, but we now have a roadmap to begin to reduce these deaths.

Mass casualty and event response plannings

Blood is one of the most constrained resources in the response to mass casualty events such as the London bombings of 2005. 21,33 Restocking of the hospitals involved in a response is difficult or impossible during the early phases of the event. At least initially, hospitals must rely on the stocks already present before the event occurs (Figure 15). New DCR paradigms requiring transfusion of more blood products per casualty will also increase the strain on these limited resources. 34 There are important implications, therefore, for major incident preparedness, as well as for planning for expected events such as the London Olympics.

FIGURE 15.

Provision of red cell transfusions during mass casualty events: (a) red cell stocks at the Royal London Hospital during the bombings of 7 July 2005; and (b) the best correlation of red cell requirements is with the number of critically injured casualties in a mass casualty event. ISS, Injury Severity Score; RBC, red blood cell; RLH, Royal London Hospital. Source: reproduced with permission from Glasgow S, Davenport R, Perkins Z, Tai N, Brohi K. A comprehensive review of blood product use in civilian mass casualty events. J Trauma Acute Care Surg vol. 75, iss. 3, pp. 468–74. 2013. 21 URL: http://journals.lww.com/jtrauma/pages/default.aspx. Promotional and commercial use of the material in print, digital or mobile device format is prohibited without the permission from the publisher Wolters Kluwer. Please contact healthpermissions@wolterskluwer.com for further information.

We studied the utilisation of blood stocks during mass casualty events through a systematic review of the literature and a detailed examination of blood utilisation during the London bombings and Norway attacks. 21

We were able to correlate PRBC transfusions with the number of critically injured casualties (as opposed to other measures of casualty load). As these numbers are relatively constant for specific types of events, this can be used to plan the numbers of PRBC and blood product stocks that need to be held for specific events. We were able to feed this information into the planning process for the 2012 London Olympics. 34

We then created a model of casualty flow and blood utilisation by hospitals during a mass casualty event. 35 We populated this with the data collected from the literature, our specific investigations and the results of interviews with hospital staff and specialists. Once the model was constructed (Figure 16), we could modify parameters to investigate the benefits of alternative methods of utilising and preserving blood resources during such events.

FIGURE 16.

Overview of the queuing model for examining red cell management during mass casualty events. VBA, Visual Basic for Applications.

Exploring the model we can see how increased casualty loads and stock levels have the most profound effect on how quickly casualties receive the blood products they need and how blood stocks are preserved (Figure 17). These cannot easily be mitigated by increasing the number of technicians, reducing processing time or otherwise trying to increase efficiencies within the system.

FIGURE 17.

Example modelling outputs for red cell provision during mass casualty events: (a) tornado plot showing sensitivity analysis for how changing event variables affects the number of P1s receiving timely red cell transfusions; and (b) waterfall plot showing the effect of casualty numbers and the size of red cell stocks on the timely transfusion of P1 casualties.

Mass casualty response planning therefore has to focus on methods for restocking supplies. In addition, blood utilisation may need to be restricted during these events. We examined several strategies and found that both overall limitation of total PRBC transfusions and the restriction of emergency type O blood transfusions were effective in providing the best care for the most casualties. However, the clinical acceptability of these strategies still needs to be explored.

Epidemiology

In our national study we used a representative sample of large and small hospitals with an element of selection as we relied on the submission of TARN data to assist with data collection. As this work mostly predated the introduction of regional trauma systems, it is possible that we have underestimated the incidence of trauma haemorrhage because of incomplete data reporting. It is also possible that our national estimate is inaccurate because of the hospitals not being representative of the country as a whole. However, the picture of national mortality and other outcomes and of overall transfusion practice and blood product delivery is unlikely to be affected by these potential variations.

Our health economic analysis was as detailed and accurate as possible based on the cost and resource utilisation information available to us. However, these are cost estimates and there was wide variation across the country and between patients. Nevertheless, the total figures represent the most accurate assessment to date of the costs of treating patients with trauma haemorrhage and help to understand where the main cost burden – and potential savings – may lie.

Diagnosis and treatments

The studies on diagnosis and treatment were based principally on the examination of a large cohort from two major trauma centres. As such, although trends and associations can be investigated and hypotheses generated, it is not possible to determine causality from these studies. However, we have looked deeply into mechanisms to support the clinical research findings and, when possible, we have taken these findings into a randomised controlled trial (CRYOSTAT) or collaborated with other trial groups (e.g. CRASH-2). Our findings on diagnostic test accuracy and thresholds for treatment need external validation and utilisation and these projects are ongoing through the INTRN. 36 Our cohorts were not large enough to look at responses to other blood products such as platelet transfusions and further work is required in this area.

Mass casualty management

As with all models the quality of the output is dependent on the base information available for model development. We gathered as much evidence as possible from the literature and specific event information available to us. However, this may not be representative of all possible events or even the average event. However, the model does allow us to vary the input parameters and assess the effects of such variations within the model response. We also have not been able to develop a model that includes plasma and other blood components and this is an area for future work.

Contributions of authors

Karim Brohi (Consultant, Trauma) was the chief investigator for the programme.

Simon Eaglestone (Research Manager, Trauma) co-ordinated the programme, conducted the analysis of the national incidence and current practice of transfusion in trauma haemorrhage and prepared the results for publication and final reporting.

Contributions of others

Shubha Allard (Consultant, Haematology) co-managed the collaborative observational study to describe the national incidence and current practice of transfusion in trauma haemorrhage.

Helen Campbell (Senior Researcher, Health Economics) conducted the review of health-care costs in trauma and led the analyses employing economic models.

Nicola Curry (Clinical Research Fellow, Haematology) conducted the reviews of coagulopathy, transfusion and randomised controlled trials, led the prospective recruitment to the observational studies at John Radcliffe Hospital and co-led the study of fibrinogen depletion in clinical studies.

Ross Davenport (Clinical Research Fellow, Trauma) led the prospective recruitment to the observational studies at the Royal London Hospital, conducted the identification of ATC by ROTEM and led the analyses of transfusion treatment and outcomes.

Antoinette Edwards (Project Co-ordinator, TARN) conducted the analysis of the national incidence and current practice of transfusion in trauma haemorrhage.

Simon Glasgow (Clinical Research Fellow, Trauma) conducted the review of and model development for blood resource usage in mass casualty events.

Harriet Hunt (Research Fellow, Public Health) conducted the systematic review of diagnostic test accuracy.

Chris Hyde (Professor, Public Health and Clinical Epidemiology) was the study lead for the review of diagnostic test accuracy.

Sirat Khan (Clinical Research Fellow, Trauma) conducted the analysis of major haemorrage protocol practice and transfusion ratio effectiveness in the treatment of trauma haemorrhage.

Charlotte Llewelyn (Manager, NHS Blood and Transplant Clinical Studies Unit) co-managed the collaborative observational study to describe the national incidence and current practice of transfusion in trauma haemorrhage.

Imran Raza (Clinical Research Fellow, Trauma) conducted the biomarker analysis and characterisation of elevated fibrinolysis in trauma patients.

Claire Rourke (Research Assistant, Trauma) co-led the study of fibrinogen depletion in clinical studies and conducted the ex vivo analysis of coagulopathic patient blood ROTEM.

Frances Seeney (Principal Statistician, Transfusion Medicine) led the analysis of data from the observational study of the national incidence and current practice of transfusion in trauma haemorrhage.

Simon Stanworth (Consultant, Haematology) was co-investigator for the programme.

Elizabeth Stokes (Researcher, Health Economics) conducted the analysis of health-care costs in trauma and the development of the economic models.

Maralyn Woodford (Executive Director, TARN) co-managed the collaborative observational study to describe the national incidence and current practice of transfusion in trauma haemorrhage.

Publications

Input into clinical guidelines and systematic reviews

Spahn DR, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernández-Mondéjar E, et al. Management of bleeding and coagulopathy following major trauma: an updated European guideline. Crit Care 2013;17:R76. https://doi.org/10.1186/cc12685 [European bleeding guidelines (2013)]

Mitra B, O’Reilly G, Cameron PA, Zatta A, Gruen RL. Effectiveness of massive transfusion protocols on mortality in trauma: a systematic review and meta-analysis. ANZ J Surg 2013;83:918–23. https://doi.org/10.1111/ans.12417 (Restricted link: the corresponding author should be contacted for access to the article.)

National Institute for Health and Care Excellence. Detecting, Managing and Monitoring Haemostasis: Viscoelastometric Point-of-Care Testing (ROTEM, TEG and Sonoclot systems). Diagnostics guidance DG13. London: NICE; 2014. URL: www.nice.org.uk/guidance/dg13 (accessed 14 August 2017).

Hunt BJ, Allard S, Keeling D, Norfolk D, Stanworth SJ, Pendry K; British Committee for Standards in Haematology. A practical guideline for the haematological management of major haemorrhage [published online ahead of print 6 July 2015]. Br J Haematol 2015. [British Committee for Standards in Haematology Guidelines (2015)]. (Restricted link: the corresponding author should be contacted for access to the article.)

Klein AA, Arnold P, Bingham RM, Brohi K, Clark R, Collis R. AAGBI guidelines: the use of blood components and their alternatives 2016. Anaesthesia 2016;71(Suppl. 7):829–42. https://doi.org/10.1111/anae.13489 (Restricted link: the corresponding author should be contacted for access to the article.)

Hunt H, Stanworth S, Curry N, Woolley T, Cooper C, Ukoumunne O, et al. Thromboelastography (TEG) and rotational thromboelastometry for trauma induced coagulopathy in adult trauma patients with bleeding. Cochrane Database Syst Rev 2015;2:CD010438. https://doi.org/10.1002/14651858.CD010438.pub2 (Restricted link: the corresponding author should be contacted for access to the article.)

Whiting P, Al M, Westwood M, Ramos IC, Ryder S, Armstrong N, et al. Viscoelastic point-of-care testing to assist with the diagnosis, management and monitoring of haemostasis: a systematic review and cost-effectiveness analysis. Health Technol Assess 2015;19(58). https://doi.org/10.3310/hta19580

National Institute for Health and Care Excellence. Major Trauma: Assessment and Initial Management. NICE guideline NG39. London: NICE; 2016. URL: www.nice.org.uk/guidance/ng39 (accessed 14 August 2017).

Data sharing statement

All available data can be obtained by contacting the corresponding author.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, CCF, NETSCC, PGfAR or the Department of Health. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the PGfAR programme or the Department of Health.

Media

Cohen D. Code Red – Repairing Blood in the Emergency Room. New Scientist, 19 October 2011. URL: www.newscientist.com/article/mg21228352–900-code-red-repairing-blood-in-the-emergency-room/ (accessed 9 September 2015).

BBC World Service and BBC Radio 4. Frontline Medicine: Survival. 2012. URL: www.bbc.co.uk/programmes/b017Ld83 (accessed 28 August 2017).

BBC Television. Trauma Medicine: the Fight for Life. 2014. URL: www.bbc.co.uk/programmes/b04svfs5 (accessed 28 August 2017).

Science and medicine education

Royal Society Summer Science Exhibition, London, 2011.

Moscow Science Fair, Moscow, 2011.

Big Bang Fair, Birmingham, 2012 (Wellcome Trust People Award WT098159A1A).

Centre of the Cell: Trauma Surgery: Science of the Bleeding Obvious, 2011–2014. Shortlisted for a Times Higher Education Widening Participation Award, 2012.

PhDs completed

Ross Davenport. Acute Traumatic Coagulopathy: Functional Characterisation of the Protein C Pathway and Haemostatic Response to Transfusion. PhD Thesis. Glasgow: University of Glasgow; 2011.

Nicola Curry. The Coagulopathy of Trauma Related Major Haemorrhage. PhD Thesis. London: Queen Mary University of London; 2014.

Sirat Khan. Blood Component Therapy in Trauma Haemorrhage. PhD Thesis. Glasgow: University of Glasgow; 2015.

Simon Glasgow. Modelling Red Blood Cell Provision in Mass Casualty Events. PhD Thesis. Glasgow: University of Glasgow; 2015.