Abstracts and Summary of Postlecture Discussions
1 HYPOPERFUSION & ACTIVATION OF PROTEIN C IN TRAUMA
R. Davenport, and K. Brohi. Trauma Sciences, Queen Mary University of London & Royal London Hospital, London, United Kingdom
Up to 25% of severely injured patients arrive at the hospital with significant clotting dysfunction and are four times more likely to die than those with normal hemostasis. Trauma-induced coagulopathy has a multifactorial etiology and classically described by consumption of clotting factors, hemodilution, acidosis, and hypothermia. Trauma-induced coagulopathy has now been shown through experimental and clinical studies to be initiated by an endogenous process—acute traumatic coagulopathy (ATC). This early coagulopathy appears to be driven by the combination of tissue trauma and systemic hypoperfusion and is characterized by global anticoagulation and fibrinolysis, putatively through activation of the protein C (PC) pathway.
Shock and tissue hypoperfusion are strong independent risk factors for poor outcomes in trauma. Mouse models of ATC and large observational studies have shown that patients with a normal base deficit do not have prolonged clotting times, regardless of injury severity or the amount of thrombin generated. However, there is a dose-dependent prolongation of clotting times with increasing systemic hypoperfusion. Shock with tissue injury is associated with increased plasma thrombomodulin (TM) and a reduction in PC. In the presence of tissue hypoperfusion and significant thrombin generation following tissue trauma, the endothelium expresses TM, which complexes with thrombin to divert it to an anticoagulant function. Less thrombin is available to cleave fibrinogen, and thrombin complexed to TM activates PC, which inhibits cofactors V and VIII. In addition, trauma patients with shock have been shown to have a reduction in plasminogen activator inhibitor 1 and elevated tissue plasminogen activity. Activated PC in excess will consume plasminogen activator inhibitor 1 and results in a “de-repression” of fibrinolytic activity. Acute traumatic coagulopathy is thus hypothesized to be driven by systemic hypoperfusion and tissue injury triggering a “thrombin switch,” which diverts thrombin from cleaving fibrinogen to large-scale activation of PC producing early global anticoagulation and hyperfibrinolysis.
High plasma-soluble TM and low PC are associated with increased mortality and greater transfusion requirements. Patients are more likely to develop acute renal injury, ventilator-acquired pneumonia, and multiple organ failure. Novel resuscitation strategies with early high-dose fresh frozen plasma (FFP) are utilized to specifically address ATC. However, large variability exists in the coagulation response to aggressive FFP transfusion, and precisely which patients derive benefit from damage control resuscitation are currently unknown. Theoretically, administration of high-dose FFP in ATC may actually potentiate anticoagulation by augmenting the production of thrombin, accelerating the activation of PC. As well as being involved in coagulation, PC has pivotal role in inflammation, and patients with severe sepsis have low PC levels. Trauma patients receiving massive transfusions have an increased incidence of sepsis, and it is conceivable that the low PC levels seen in both conditions may be the result of systemic hypoperfusion and early activation (and so depletion) of PC. Further studies characterizing the etiology of ATC are required to enable the design of innovative therapeutic strategies able to effectively reverse early traumatic coagulopathy and prevent delayed complications.
First, Voelckel focused on the relationship between increase in circulating activated PC (APC) and frequent incidences of acute lung injury in the studied population of traumatic patients. However, the exact cause(s) of acute lung injury could not be established at this time as the presented data were indicative of only an association between these two events.
Görlinger related the presented evidence to data generated by his own group (PubMed 20929576), by using the clot lysis index (by ROTEM [rotational thromboelastometry]) at the 60-min time point; they were able to accurately discriminate between septic and nonseptic postoperative patients. This was principally exemplified by septic patients with the high risk of death—this cohort appeared not to be able to activate their PC system and failed to produce an effective lysis in the early postseptic period. This was contrasted by a robust shift toward fibrinolysis in virtually all surgical subjects with extensive postoperative trauma.
Another two important points were made by Hunt, who stressed that an effective therapeutic support of the PC system in critical patients can be achieved only with APC (as shown by use of activated drotrecogin α [Xigris] in severe septic patients), because these subjects can neither produce adequate PC quantities, nor ensure an effective conversion of circulating PC zymogen into its active form. She further pointed out that decline of factor V (a major breakdown target for plasmin) has been traditionally correlated with the degree of fibrinolysis much more than breakdown of APC.
Finally, Soerensen indicated the relevance for measurement of thrombin activated fibrinolysis inhibitor (TAFI) in the setting of trauma-induced coagulopathy given the competitive activation between PC and TAFI with thrombin/TM binding. He argued that because the functional assay of maximal clot firmness (by EXTEM) has a very little sensitivity for a low-grade hyperfibrinolysis, the decrease in the maximal clot firmness observed in Davenport and Brohi’s study might have been due to an inappropriate (and undetectable) shift in activation of TAFI rather than PC alone.
2 ANOTHER FORM OF DIC
J. R. Hess. University of Maryland School of Medicine, Baltimore, Maryland
The common form of disseminated intravascular coagulation (DIC) occurs in sepsis. In that situation, endothelial cells are activated by inflammatory cytokines to express tissue factor and lose thrombomodulin. This procoagulant change results in the primary activation of plasma coagulation independent of endothelial disruption. Platelet and fibrin thrombi are formed intravascularly, and because they are not tethered to the subendothelium, they embolize to cause microvascular blockage. Microvascular blockages in the brain, heart, liver, kidney, and muscle lead to the common symptoms of confusion, weak heart action, liver and kidney dysfunction, and muscle enzyme elevations. This process is truly disseminated, intravascular, and coagulant.
Another form of DIC occurs after massive high-energy injury, when vessel walls are disrupted, and tissue factor is exposed on the surfaces of smooth muscle cells and fibroblasts in the bottoms of millions of endothelial microtears between normal thrombomodulin-bearing endothelial cells. Under these circumstances, the consequences of activation of coagulation are primarily hemorrhagic, with consumption of early factors in the coagulation cascade, and activation of fibrinolysis. The pathogenesis of this process is local to the many sites of injury, extravascular (in the sense that it is outside the continuity of endothelial cells), and hemorrhagic.
Both the thrombotic and hemorrhagic forms of DIC were described in the 2001 International Society of Thrombosis and Hemostasis definition. However, the massive injury-related hemorrhagic form is so clinically distinct from conventional sepsis-associated DIC that a separate literature developed in the trauma community calling the hemorrhagic form of DIC the acute coagulopathy of trauma or the acute coagulopathy of trauma and shock (ACoTS). ACoTS was associated with increased injury severity and the presence of shock, recognized by prolonged prothrombin and partial thromboplastin times or abnormalities in all phases of the thromboelastogram, and led to a high mortality. For individuals who do not have a highly sophisticated understanding of injury and cell-based coagulation, it is probably useful to continue to think of them as two separate but often temporally linked syndromes. It makes treatment easier.
Thrombomodulin, the molecule whose cellular expression is different in these two clinical situations, is normally on the surface of healthy endothelial cells. There, it binds thrombin made with the activation of plasma coagulation, alters thrombin’s conformation and enzymatic specificity to stop activating coagulation factors, and instead activates the anticoagulant molecules protein C and its cofactor protein S. Protein C, in turn, inactivates the activated forms of factors V and VIII (Va and VIIIa) and the fibrinolysis inhibitor, plasminogen activator inhibitor 1. In healthy tissue, thrombomodulin serves as a powerful localizer of the hemostatic response. In the presence of severe soft tissue injury, local activation of factor C on a massive scale leads rapidly to factor V depletion and uninhibited fibrinolysis.
Two physiologic disruptions appear to be the major associates of the acute hemorrhagic form of DIC. These are the extent of endothelial disruption and the onset of shock. Endothelial disruption is clearly a driver of the fibrinolytic process. The extent to which shock decreases the clearance of thrombin and therefore feeds the cycles of factor V activation and inactivation is unknown, but probably substantial. Protection from hemorrhagic injury may well explain why factor V Leiden, which is resistant to protein C degradation, persists in the population. Alternatively, shock may just be a marker of the extent of endothelial disruption.
Whatever the actual mechanisms, ACoTS is the hemorrhagic form of DIC. It is a reminder that the coagulation system is fundamentally weak. Nine grams of fibrinogen and 10 mL of platelets are all we have to make clots in healthy individuals. Large injuries can overwhelm this frail system.
Voelckel inquired whether the almost exponential relationship between Injury Severity Score and incidences of coagulopathy varied between patients with blunt and penetrating trauma. Hess explained that such a comparison was not technically feasible given that the majority of trauma patients in the study suffered from high-velocity/energy gunshot wounds.
Next, Hunt stressed that an optimal diagnosis of DIC, especially in trauma patients, should not solely rely on laboratory findings, but must include a DIC-oriented clinical evaluation by an emergency clinician(s). She additionally appreciated the less known evidence indicating that also trauma patients with high risk of death may display only minor imbalance in coagulation.
By touching on the subject of classification controversy of trauma-induced coagulopathy (TIC), Kashuk rapidly increased the temperature of the subsequent discussion. He pointed out that it is yet to be unequivocally determined whether TIC constitutes an entirely different syndrome type, or the classic one (DIC) that remains largely underdiagnosed. First, it needs to be reliably established whether the early hypercoagulopathic state (present in the classic DIC) indeed occurs in TIC patients, or an alternative (plasminogen activator inhibitor 1–independent) activation is the key trigger.
Hess very graphically argued in favor of an alternative pathophysiology of TIC, thus a complete lack of early hypercoagulation in trauma patients. According to his rationalization, the latter could be due to the default weakness of the procoagulation system that can be easily overwhelmed by even a medium-grade injury and is immediately subjected to a profound consumption of key procoagulative factors. Furthermore, Hess attributed the latter phenomenon to a massive microtearing of endothelium (similar to the one seen in acute lung injury) during traumatic events. This in turn causes an immense exposure of tissue factor that quickly soaks out all available circulating platelets and factor VII (along with other relevant components). In effect, a procoagulative thrombin burst never occurs because the coagulation system is exhausted even before it can be effectively triggered.
Kashuk disagreed, pointing out that a large number of trauma patients who show distinct and immediate coagulation abnormalities in absence of early factor depletion seem to defy the aforementioned notion. Redl, Hess, and Martini speculated that because of a relatively large circulating pool of involved factors, their depletion may not be immediately detectable in the blood, and accuracy of such measurements may be erroneously swayed by the suboptimal selection of tests/end points.
Stalwart counterarguments against the alternative TIC were offered by Soerensen. He claimed that existing data from relevant animal models (e.g., intracerebral injury) actually confirm existence of a transient, yet very robust hypercoagulation that precedes the subsequent hyperfibrinolytic phase. Agreeing with the previously mentioned discussants, Soerensen again stressed the key importance of appropriate choice of tests to determine TIC fluctuations. He specifically alerted the audience to the utter ineffectiveness of activated partial thromboplastin time and prothrombin time measurements for such a monitoring, specifically in the very early phase after traumatic incidents. As a viable alternative, he suggested that the whole-blood clotting assay in corn trypsin inhibitor–stabilized plasma (to inhibit artificial activation of factor XII) offers an excellent sensitivity for detection of posttraumatic hypercoagulation.
The final comments in this debate were offered by Boffard, Soerensen, and Voelckel. The first one claimed that he had observed a very brief period of hypercoagulation in his gunshot wound patients. Yet, he simultaneously stated that the importance of the early and transient phase should not be overestimated in contrast to a much longer hyperfibrinolytic phase. Soerensen fervently disagreed, pointing out that the early hypercoagulative changes, regardless of how brief, may be a decisive trigger of severe hemostatic derangements subsequently occurring in trauma patients. Voelckel ended this heated discussion with a somber concern suggesting a potential need for adjustments of normal/reference values for trauma patients because virtually all classic measurement methods had been typically standardized based on data from septic (and others) but not trauma-induced coagulopathies.
3 IMPACT OF ACIDOSIS AND HYPOTHERMIA
W. Z. Martini. US Army Institute of Surgical Research, Fort Sam Houston, Texas
Hypothermia, acidosis, and coagulopathy constitute the “vicious cycle” in trauma patients. The effects of hypothermia and acidosis on coagulation have been shown to be resulted from inhibitions on enzyme activities and platelet function. Recent studies have revealed the inhibitory mechanisms of hypothermia and acidosis on thrombin generation and fibrinogen availability.
Acidosis and hypothermia inhibit thrombin generation via different mechanisms. Hypothermia primarily inhibits thrombin generation in the initiation phase, whereas acidosis inhibits thrombin generation severely in the propagation phase and moderately in the initiation phase. In fibrinogen embolism, hypothermia inhibits fibrinogen synthesis, whereas acidosis accelerates fibrinogen degradation. Both lead to a potential deficit in fibrinogen availability. Furthermore, in contrast to the reversible effects of hypothermia on coagulation, pH correction cannot immediately reverse the effects of acidosis on coagulation. Thus, acidosis appears to be more detrimental than hypothermia in the development of coagulopathy.
In summary, hypothermia and acidosis impair coagulation process via different mechanisms. Although the impairment of hypothermia can be effectively reversed via rewarming, the detrimental effects of acidosis on coagulation cannot be immediately corrected via pH neutralization. Thus, future studies are warranted to investigate the efficacy of pH neutralization in conjunction with coagulation substrate supplementation in correcting acidosis-induced clotting complications.
Voelckel opened the discussion by inquiring about the clinical importance of the duration and/or severity of respiratory acidosis and its impact upon the length and/or reversibility of coagulation derangements. Martini’s experiments could offer only a partial explanation, given that fixed observation times/severity of acidosis had been applied. Based on the work from other groups, however, it appears that irrespective of the means by which acidosis is reached, coagulation abnormalities are difficult to reverse once they set in.
Next, Kashuk asked whether acidosis occurs by itself before shock, or whether shock is the key trigger, and to what extent these differences may influence coagulation. Recalling her former study based on a model involving severe injury/bleeding (65% blood loss/injury with no intervention for 30 min), Martini stated that recorded coagulation derangements were influenced in a similar fashion, regardless whether shock/acidosis had been combined or acidosis alone had been present. The related issue, whether the impact of acidosis during varying degrees of shock severity is proportional to the changes in the coagulation system, remained open, given the unchanging level of shock severity in all previously mentioned studies.
Görlinger noted that the mechanism responsible, at least partially, for the acidosis-dependent irreversibility of coagulation derangements is related to clustering of glycoprotein 1b receptors on platelets and their subsequent elimination in the liver—an active process that continues even after acidosis had been corrected.
Gerner shared results of his own in vitro ROTEM (rotational thromboelastometry) study, in which effects of hypothermia/acidosis on tissue plasminogen activator–induced hyperfibrinolysis had been investigated. His group observed that acidosis aggravated hyperfibrinolysis, whereas hypothermia notably decreased its rate. Given this interesting discrepancy noted in vitro, Gerner inquired whether Martini’s group observed similar phenomenon in their in vivo models. Unfortunately, this question remained unanswered at this time, given that the group of Martini had not yet acquired reliable data on the impact of hypothermia/acidosis upon fibrinolysis. Similarly, the effect of varying levels of hypothermia on coagulation/fibrinolysis end points, a question voiced by Maegele, is yet to be investigated in subsequent studies.
4 CROSS TALK BETWEEN COAGULATION AND INFLAMMATION
M. Huber-Lang. Clinical and Experimental Trauma-Immunology, Dept. of Orthopedic Trauma, Hand-, Plastic, and Reconstructive Surgery, University Hospital Ulm, Ulm, Germany
Severe tissue injury and systemic inflammation result in generation, release, and exposure of danger- and pathogen-associated molecular patterns (DAMPs and PAMPs). In response, there is an early activation of the “serine protease system” including the coagulation and complement cascade as major fluid phase arsenal of innate immunity. There is increasing evidence that both, the coagulation and the complement system, are capable of sensing, transmitting, translation, and clearance of DAMPs and PAMPs. However, whereas interactions between both systems have been proposed, the underlying molecular mechanisms remain largely unknown.
Recently, we have reported multiple novel links for various factors of the coagulation and fibrinolysis cascades with key components of the complement system driving together the inflammatory response. Coincubations of C3 or C5 with thrombin, factor XIa (factor XIa), FXa, FIXa, and plasmin were all found to generate the potent inflammatory mediators C3a and C5a, respectively. Furthermore, mass spectrometric analyses identified the C3- and C5-cleavage products as the native C3a and C5a anaphylatoxin molecules, and chemotaxis assays using human mast cells and neutrophils confirmed full biological function. When FXa was added to normal human serum as a more complex system, reduction in the complement hemolytic activity was found, suggesting systemic complement activation, which is also reflected by enhanced formation of C3a, C5a, and the membrane attack complex. In this context, specific inhibition of FXa by enoxaparin or fondaparinux led to a concentration-dependent suppression of the anaphylatoxin generation. Translational analysis of plasma from patients with extensive activation of the coagulation cascade by multiple injuries (Injury Severity Score >18) revealed an almost synchronic generation of C5a and furthermore an association between enhanced thrombin-antithrombin complexes and C5a. In accord with these findings, others have recently shown that C5a induces tissue factor activity in human endothelial cells and stimulate tissue factor expression on neutrophils, providing a potent procoagulant environment. Additional cross links of inflammation and coagulation have been lately described, especially platelets polyphosphates acting as proinflammatory and procoagulant mediators, as well as fibrin-domain–containing proteins (e.g., L-ficolin, H-ficolin) acting as pattern recognition receptors and being capable of activating complement. Vice versa, thrombin generation has been reported to be promoted by both MASP2 (mannan-binding lectin-associated serine protease 2) and membrane attack complex. Upstream in the coagulation cascade, FXIIa initiates the traditional “intrinsic pathway” but also activates the “classic pathway” of complement and generates bradykinin as a part of a potent inflammatory reaction.
In summary, an intense molecular cross-talk between the coagulation and complement system exists and seems to trigger progress and maintain the inflammatory response after tissue trauma and during systemic inflammation. Consequently, therapeutic modulation of the coagulation response may also change the inflammatory response, and future immune modulation of the DAMP and PAMP response may significantly interfere with the clotting process.
Discussion was started by Gerner, who sought after a link between ischemia-reperfusion injury (IRI), coagulation, and activation of the complement system. Huber-Lang referred to the early work by P. Ward and G. Till, who revealed for the first time the link between IRI and complement activation (specifically C3) using a hind limb IRI model (PubMed 8102031). Yet, studies that simultaneously combine all three elements are lacking, presenting a great investigative avenue to be followed.
The next question by Görlinger inquired about the importance of Toll-like receptors (TLRs) in the context of IRI and potential relationship between both. Huber-Lang indicated that TLR involvement occurs primarily via PAMPs and DAMPs. Furthermore, whereas the first system is clearly related to infectious stimuli, the latter one seems to be strongly associated to IRI. Injury releases vast amounts of DNA/RNA that activate relevant TLRs, and the recent article by Hauser group (PubMed 0203610) demonstrated that mitochondrial DAMPs released after trauma can effectively signal via TLR9. He further stated that emerging data confirm that a cross-talk between TLRs and complement indeed exists.
Finally, Voelckel asked whether a viable therapeutic strategy, by means of manipulation of the complement system components after traumatic events, can be foreseen in the future. Huber-Lang stressed that such an intervention is unlikely to be successful, primarily because of the extremely quick cascade of events in trauma patients. He referred to recent and ongoing studies by his group, in which C5a receptor antagonist intervention diminished development of acute lung injury, but no data are yet available on the coagulation component. He additionally pointed out that complement system is an important defense mechanism; thus, its complete inhibition may cause more harm than benefit. The ultimate goal of an optimal therapeutic intervention, he underlined, is to only inhibit the excessive activation of this system. Conversely, he was much more optimistic about treatment attempts in sepsis. Septic events frequently induce DIC, yet its evolution is more protracted creating a wider window of therapeutic opportunity. Huber-Lang has been currently involved in an early phase clinical trial that examines potential benefits of C5a receptor antagonist in septic patients.
5 PRIMARY FIBRINOLYSIS IS INTERGRAL TO THE PATHOGENESIS OF THE ACUTE COAGULOPATHY OF TRAUMA
J. Kashuk. Denver Health & Hospital Authority, Trauma Services, Denver, Colorado
Background: The existence of primary fibrinolysis (PF) and a defined mechanistic link to the acute coagulopathy of trauma is controversial. Rapid thromboelastography (r-TEG) offers point-of-care comprehensive assessment of the coagulation system. We hypothesized that postinjury PF occurs early in shock, leading to postinjury coagulopathy and ultimately hemorrhage-related death.
Methods: Consecutive patients over 14 months at risk for postinjury coagulopathy were stratified by transfusion requirements into massive (MT), more than 10 U/6 h (n = 32); moderate (Mod), 5 to 9 U/6 h (n = 15); and minimal (Min), less than 5 U/6 h (n = 14). Rapid TEG was performed by adding tissue factor to uncitrated whole blood. Rapid TEG estimated percent lysis was categorized as PF when greater than 15% estimated percent lysis was detected. Coagulopathy was defined as r-TEG clot strength = G < 5.3 dyn/cm2. Logistic regression was used to define independent predictors of PF.
Results: Thirty-four percent of injured patients requiring MT had PF, which was associated with lower emergency department systolic blood pressure and temperature and worse base deficit/pH/lactate (analysis of variance, all P < 0.0001). The risk of death correlated significantly with PF (P = 0.026) (left figure). Primary fibrinolysis occurred early (median, 58 min; interquartile range, 1.2–95.9 min); every 1-U drop in G increased the risk of PF by 30%, and death by greater than 10% (right figure).
Conclusions: Our results confirm the existence of PF as detected by r-TEG in severely injured patients. It occurs early (<1 h) and is associated with massive transfusion requirements, coagulopathy, and hemorrhage-related death. These data warrant renewed emphasis on the early diagnosis and treatment of fibrinolysis in this cohort.
Discussion was started by Maegele and Schöchl, who inquired about the technical setup/reliability of TEG analysis used in the presenter’s hospital and requested explanation on the definition of fibrinolysis he uses. Kashuk explained that because of quality assurance and standardization, all TEG measurements are run in the hospital laboratory of medicine, but both emergency department and/or operating room(s) are provided with real-time TEG tracings, thanks to the online video feed system, whereas PF is defined as the maximum clot firmness reduction of greater than 15% of the maximum amplitude on the TEG tracing. In addition, an intermediate-/low-level fibrinolysis is defined as less than 15% maximum clot firmness reduction but with a demonstrated fibrinolysis.
Next, Görlinger commented on the resuscitation decision making in the intensive care unit (administration of cryoprecipitate vs. platelets [PLTs]) guided by the α angle and the maximum (TEG) amplitude. He pointed out that both fibrinogen deficiency and low PLT count can influence the aforementioned parameters, hence leading to potentially false conclusions. Görlinger revealed that the algorithm his hospital has been using relies on the extrinsically activated test with, and without cytochalasin D—to better define whether observed derangements are due to the shifts in PLTs or fibrinogen.
Kashuk explained that the algorithm in use is not incorrect given that situations, in which an isolated narrow α angle or prolonged K could be detected, are very rare. In addition, the effects of fibrinogen typically occur in the same setting as the PLT or enzymatic dysfunction. He additionally stated that a profoundly narrowed α angle simultaneous with a normal PLT function would be a clear indication for the use of cryoprecipitate.
In a supplementary reply, Görlinger recalled very high correlation (r = 0.86) between α angle and the maximum amplitude based on an analysis of more than 900 patients from his hospital. According to him, this suggests that discrimination between PLT-related disturbances versus fibrinogen-dependent derangements is much better when maximum amplitude tests are performed with and without PLT inhibition.
6 HYPERFIBRINOLYSIS IN TRAUMA
H. Schöchl, C. Solomon, W. Voelckel, H. Redl, and S. Bahrami. Ludwig Boltzmann Institute of Experimental and Clinical Traumatology, Vienna; and AUVA Trauma Centre, Salzburg, Austria
One quarter of all major trauma patients are coagulopathic on arrival at the emergency room (ER) (1). For a long time, it has been assumed that blood loss, consumption of coagulation factors, hypothermia, and acidosis are the major drivers of bleeding disorders following injury. Recently, hyperfibrinolysis (HF) has been identified as another important contributor to coagulopathy (2). The combination of hypoperfusion and tissue trauma activates an endogenous anticoagulant pathway, which also results in a profibrinolytic state. However, the extent to which HF contributes to coagulopathy is still unknown. Because HF is a life-threatening bleeding disorder in major trauma patients, a fast and reliable diagnostic tool is essential to treat patients adequately. Viscoelastic tests (rotational thromboelastometry [ROTEM] and thromboelastography [TEG]) are assumed to be the gold standard for diagnosing HF (3). Hyperfibrinolysis according to ROTEM/TEG test results has been observed in three small studies (4–6). However, the real incidence of HF following massive trauma is still speculative; it depends on patient inclusion criteria, Injury Severity Score (ISS), and severity of hypoperfusion and shock.
In a prospective study, Levrat et al. reported HF in six (7.3%) of 83 trauma patients (4). Those patients suffering from HF who sustained major trauma with an ISS of 75 and no measurable levels of fibrinogen subsequently died (4). In a prospective study, Carroll and coworkers (5) enrolled 161 trauma patients from whom blood samples were drawn at the scene and upon arrival in the ER. Three patients showed HF in blood samples taken at the scene. Two of these patients died, but in the third patient, clot lysis had normalized by the time of arrival at the ER, and this patient recovered. One patient developed HF between the sample being taken at the scene and arrival at the ER; this patient died. All of the patients were classified as being in shock, Overall, HF had an incidence of 2.5%, and the HF mortality rate in the ER was 67% (5). In another study, Schöchl et al. (6) identified 33 cases of HF over a 4-year observation period. Based on the pattern of HF, three patient groups were described: one group with fulminant HF, defined as clot breakdown within 30 min; an intermediate group with clot breakdown between 30 and 60 min; and a late group with clot lysis occurring after 60 min (6). All patients in the fulminant group died either in the ER or soon after arrival in the intensive care unit. Only one patient in the intermediate group survived, whereas the mortality rate was lowest in the late group. These findings suggest that fulminant HF might be irreversible and refractory to immediate therapy with antifibrinolytic agents, even in combination with all other therapeutic efforts. Thus, fulminant HF may be a marker of a nonsurvivable injury and a premortem sign. Mortality rates in all three groups were significantly higher than those predicted by trauma ISS methodology. These results are consistent with the notion that HF independently contributes to poor outcome among major trauma patients (5). Tranexamic acid has been shown to effectively stop even fulminant HF (7). Hyperfibrinolysis degrades not only the fibrin network but also fibrinogen; therefore, it is essential to supplement fibrinogen levels to ensure adequate fibrin formation.
Conclusions: Hyperfibrinolysis is underdiagnosed and underreported. Recent data suggest that HF occurs predominately in major trauma with pronounced shock. Hyperfibrinolysis seems to be a predictor of poor outcome. Viscoelastic tests (ROTEM/TEG) allow early and rapid diagnosis of HF. Immediate therapy with antifibrinolytic agents such as tranexamic acid is crucial. As fulminant HF degrades not only fibrin but also fibrinogen, supplementation of fibrinogen is warranted.
1. Brohi K, Cohen MJ, Davenport RA: Acute coagulopathy of trauma: mechanism, identification and effect. Curr Opin Crit Care 13:680–685, 2007.
2. Brohi K, Cohen MJ, Ganter MT, et al.: Acute coagulopathy of trauma: hypoperfusion induces systemic anticoagulation and hyperfibrinolysis. J Trauma 64:1211–1217, 2008.
3. Johansson, PI, Stissing T, Bochsen L, Ostrowski SR: Thrombelastography and thromboelastometry in assessing coagulopathy in trauma. Scand J Trauma Resusc Emerg Med 17:45, 2009.
4. Levrat A, Gros A, Rugeri L, et al.: Evaluation of rotation thrombelastography for the diagnosis of hyperfibrinolysis in trauma patients. Br J Anaesth 100:792–797, 2008.
5. Carroll RC, Craft RM, Langdon RJ, et al.: Early evaluation of acute traumatic coagulopathy by thrombelastography. Transl Res 154:34–39, 2009.
6. Schöchl H, Frietsch T, Pavelka M, et al.: Hyperfibrinolysis after major trauma: differential diagnosis of lysis patterns and prognostic value of thrombelastometry. J Trauma 67:125–131, 2009.
7. Brenni M, Worn M, Brüesch M, et al.: Successful rotational thromboelastometry-guided treatment of traumatic haemorrhage, hyperfibrinolysis and coagulopathy Acta Anaesthesiol Scand 54:111–117, 2010.
*********Joint Discussion After Schöchl/Kashuk*********
The first comment in this joint debate (after Kashuk/Schöchl talks) was offered by Huber-Lang, who, based on his personal observations, shared his concern about a marked increase in bleeding in patients receiving more than 1 L of hydroxyethyl starch. Schöchl speculated that this is most likely the consequence of resuscitative efforts at the accident site. Because ER physicians use large amounts of fluids and high amounts of colloids (2.2 L median, and up to 4 L total according to data from the German trauma registry), patients on arrival are often heavily diluted with all ensuing consequences observed in the intensive care unit.
The next question came from Kozlov who inquired about local coagulation disturbances, namely, whether coagulopathy observed in circulating blood is correlated to coagulation/fibrinolytic derangements observed locally in the microcirculation and/or tissues. This question was left unanswered for lack of available data, but Kashuk admitted that this phenomenon should not be overlooked with regard to treatment of trauma patients. He observed that local hyperfibrinolysis frequents open heart surgery patients, or those who receive various artificial devices. Likely, a fine balance between local and systemic fibrinolysis exists, and it may be strongly related to the magnitude of endothelial disruption (as indicated earlier by Hess).
Fries inquired whether patients are more prone to hyperfibrinolysis when smaller fluid volumes are administered during resuscitation. Both speakers agreed that the degree of shock and administration of catecholamines/vasopressin rather than the fluid amount play major role in the development of hyperfibrinolysis.
The next discussant, Soerensen, wanted to know what an ideal approach to an early detection of hyperfibrinolysis would be, and whether ROTEM indeed provides satisfactory level of sensitivity and quality in detection of different degrees of fibrinolysis (as advocated by both speakers). In Schöchl’s opinion, the fibrin-based assay is both quick and very indicative for hyperfibrinolysis in trauma patients. Yet, when one wants to rely on such assays (which demonstrate an early clot breakdown), a clear and acceptable initiation cutoff for an aggressive treatment (of trauma patients) needs to be established. Kashuk added that because clinical data strongly suggest hyperfibrinolysis occurs very early post trauma, a point-of-care device detecting this condition in the field, shortly after injury, would be ideal. He further speculated that such an immediate measurement might even allow detection of the transient and early (and highly debatable) phase of posttraumatic hypercoagulability. Kashuk finally stated that the most reasonable (and cost-effective) strategy for use of antifibrinolytics in trauma should concentrate on identifying patients that are at the highest risk for hyperfibrinolysis, rather than on administering these drugs to every single patient including those who are at a relatively small risk (for hyperfibrinolysis).
The latter statement provoked Hunt to a passionate rebuttal: she stated that aforementioned derangements are predominantly provoked by release of tissue plasminogen activator (tPA), which in turn is mainly triggered by tissue hypoxia occurring in trauma. She used the scenario of liver transplantation as a paradigm describing the cascade of events occurring in trauma: upon reconnection, hypoxic endothelium of the transplanted liver abundantly releases tPA, producing a classic hyperfibrinolytic trace. In addition, patients with liver diseases have higher level of tPA due to the inability of this organ to effectively remove tPA from circulation—as it is the case with (e.g., posttraumatic) hypoxia.
The next, much calmer, yet no less meritorious, disagreement came from Görlinger, who observed that in his facility most of the liver transplants (>2,000 performed; without prophylaxis with antifibrinolytics) had developed distinct hyperfibrinolysis before reperfusion, suggesting that mechanisms driving hyperfibrinolysis in liver transplants versus trauma patients are entirely different. He then added that in the context of liver transplant prognostication, not the absence/presence of hyperfibrinolysis itself, but the spontaneous ability/inability to halt this process by the reperfused liver is decisive for a good/poor prognosis (respectively).
Another point in the discussion, how the hypercoagulation can be propagated in the presence of a high posttraumatic tPA burst, was brought up by Eibl. One plausible explanation was offered by Hunt, who speculated that excessive concentration of plasmin (triggered by the high posttraumatic tPA burst) activates platelets and factor V, which in turn facilitate clot formation despite high levels of tPA in the blood. Related process had been already reported in the context of old data from infarct patients, in whom reformation of dissolved clots was observed when administration of heparin was prematurely discontinued/not offered.
Following the latter issue, Hunt drew attention to a more general problem: frequent incompatibility of terminology pertaining to characterization of coagulation/fibrinolytic events among different disciplines. This, she stated, often leads to miscommunication between experts representing different fields (e.g., hematologist vs. trauma surgeons) and creates confusion during the review/publication process. As a valid example, Hunt used the controversy with definitions for primary and secondary hyperfibrinolysis. Hunt further speculated that the secondary hyperfibrinolysis (triggered by thrombin generated on the clot surface) can indeed occur very early after trauma and is simply mislabeled as the primary fibrinolytic event. This provoked immediate rebuttal from Kashuk, who indicated that in the ER situation hematology consultation is of no use, and TEG or ROTEM assays are the only practical means for a rapid diagnosis of fibrinolysis. The above exchange prompted Spinella to put out another fundamental question, namely, whether clear separation of those two events (i.e., primary and secondary hyperfibrinolysis) and use of such a nomenclature are at all practical and indeed improve treatment/outcomes in the ER. Both Kashuk and Schöchl accorded that a typical trauma surgeon does not necessarily utilize the concept of primary and secondary hyperfibrinolysis and that their overall conviction is that the early posttraumatic coagulopathy is caused by the primary, and not secondary, hyperfibrinolysis. Kashuk went on to add that secondary hyperfibrinolysis is a clearly diagnosed late event successfully treated with heparin (in contrast to lytic trauma patients). Finally, he indicated that sheer (and early) generation of thrombin in coagulopathic trauma patients should not be considered as a decisive identification between the primary and secondary hyperfibrinolysis, given that it is of an extremely transient nature, and it is quickly exhausted, while the fibrinolytic process continues to predominate.
7 IMPACT OF FLUID REPLACEMENT ON BLEEDING
D. Fries. Clinical Department of Critical Care Medicine, Medical University Innsbruck, Austria
The question as to the optimal volume replacement for compensating intravascular blood loss is the subject of ongoing controversy. Ever since the results of the Cochrane Injuries Group Albumin Reviewers were published, the use of albumin has markedly declined (1). Dextrans are also hardly used anymore for volume therapy. Their extremely negative effects on the coagulatory system mean they should be used as anticoagulants, if at all, and not for volume therapy (2).
Whether colloid infusions are generally superior to crystalloids has also not been conclusively answered. A meta-analysis by Schierhout and Roberts (3) analyzed 26 studies with 1,622 patients. The authors (3) came to the conclusion that the administration of colloids is associated with enhanced mortality. A different meta-analysis by Choi et al. (4) reviewed 17 studies with 814 patients. Only the subgroup of polytrauma patients showed a lower rate of mortality for crystalloids as compared with colloids (4). Whether these meta-analyses are conclusive for European conditions is questionable. The studies involved were primarily American studies and thus used dextrans and hetastarches (high-molecular hydroxyethyl starch [HES] preparations with a high substitution degree), which have not been used for a long time in Central Europe because of their high potential for adverse effects, particularly involving the coagulatory system.
Crystalloids impair the coagulatory system primarily by means of their dilutional effect. Several studies surprisingly postulated that low doses of crystalloids as well as colloids cause hypercoagulability. These were primarily in vitro studies using nonactivated TEG measurements. Despite the shortened clot formation times and enhanced clot strengths seen for TEG, no change in the activated coagulation markers (thrombin-antithrombin complex) was observed (5–9). It is possible that the measured shortened clot formation times and enhanced clot strengths are the result of an in vitro effect produced by the influence of the sedimentation of red blood cells in diluted specimens subject to long measuring time (10). Measuring with activated TEG and markedly shortened measuring time were, however, not able to confirm this phenomenon (11, 12). Moreover, Petroianu et al. (13) concluded that hemodilution with crystalloids and colloids caused a decrease in the activity of various clotting factors in vitro. Thus, why hemodilution should activate coagulation while activated coagulation markers remain unchanged and the activity of various coagulation factors and the thrombocyte count decrease is not known.
In addition to their dilutional effect gelatin preparations also exert specific effects on the coagulatory system. Above all, they impair fibrin polymerization and disturb the network of the fibrin monomers. Moreover, reduced clot elasticity and clot weight have been reported for gelatin replacement (14–15).
Hydroxyethyl starch has been reported to be associated with an increased tendency to bleed, above all when using solutions with a high molecular weight and a high replacement degree (16). Hydroxyethyl starch solutions cause a von Willebrand type 1–like syndrome characterized by diminished FVIII activity and diminished von Willebrand factor plasma levels (17). In addition, HES impairs fibrin polymerization to a larger extent compared with gelatin. Six percent HES 130/0.4 (Voluven; Fresenius Pharma Austria GmbH), a relatively new preparation with a medium molecular weight and a low substitution degree, was praised by the industry to affect the coagulation system less than other HES preparations. However, the smaller effect on the coagulation system is probably only a consequence of the short plasma half-life time compared with preparations with high molecular weight.
1. Cochraine Injuries Group Albumin Reviewers: Human albumin administration in critically ill patients: systemic review of randomised controlled trials. BMJ 317:235–240, 1998.
2. Mortier E, Ongenae M, Baerdemaeker LD, et al.: In vitro evaluation of the effect of profound haemodilution with hydroxyethyl starch 6%, modified gelatine 4%, and dextran 40 10% on coagulation profile measured by thrombelastography. Anaesthesia 52:1061–1064, 1997.
3. Schierhout G, Roberts I: Fluid resuscitation with colloid or crystalloid solution in critically ill patients: a systemic review of randomized trials. BMJ 316:961–964, 1998.
4. Choi P, Yip G, Quinonez LG, Cook DJ: Crystalloids vs. colloids in fluid resuscitation. A systemic review. Crit Care Med 27:200–221, 1999.
5. Roberts I, Evans P, Bunn F, Kwan I, Crowhurst E: Is the normalisation of blood pressure in bleeding trauma patients harmful. Lancet 357: 385–387, 2001.
6. Ruttmann TG, James MFM, Viljoen JF: Haemodilution induces a hypercoagulable state. Br J Anaesth 76:412–414, 1996.
7. Karoutsos S, Nathan N, Lahrimi A, Grouille D, Feiss P, Cox DJA: Thrombelastogram reveals hypercoagulability after administration of gelatin solution. Br J Anaesth 82:175–177, 1999.
8. Ng KFJ, Lam CCK, Chan LC: In vivo effects of haemodilution with saline on coagulation: a randomized controlled trial. Br J Anaesth 88:475–480, 2002.
9. Ruttmann TG, James MFM, Finlayson J: Effects on coagulation of intravenous crystalloid or colloid in patients undergoing peripheral vascular surgery. Br J Anaesth 89:226–230, 2002.
10. Innerhofer P, Fries D, Klingler A, Streif W: In vivo effects of haemodilution with saline on coagulation. Br J Anaesth 89:934–939, 2002.
11. Fries D, Innerhofer P, Klingler A, Berresheim U, Mittermayr M, Calatzis A, Schobersberger W: The effect of the combined administration of colloids and lactated Ringer’s solution on the coagulation system: an in vitro study using thrombelastograph coagulation analysis. Anesth Analg 94:1280–1287, 2002.
12. Innerhofer P, Fries D, Margreiter J, Klingler A, Kühbacher G, Wachter B, Oswald E, Salner E, Frischhut B, Schobersberger W: The effect of perioperatively administered colloids and crystalloids on primary hemostasis and clot formation. Anesth Analg 95:858–865, 2002.
13. Petroianu GA, Maleck WH, Koettner KP, Liu J, Schmitt A: Effect of in vitro hemodilution with hydroxyethyl starch and dextrans on the activity of plasma clotting factors. Crit Care Med 31:250–254, 2003.
14. Mardel SN, Saunders FM, Allen H, et al.: Reduced clot quality of clot formation with gelatin based plasma substitutes. Br J Anaesth 80:204–207, 1998.
15. Engvall E, Ruoslahti E, Miller EJ: Affinity of fibronectin to collagens of different genetic types and fibrinogen. J Exp Med 147:1584–1595, 1978.
16. Treib J, Baron JF, Grauer MT, Strauss RG: An international view of hydroxyethyl starches. Int Care Med 25:258–268, 1999.
17. Vogt NH, Bothner U, Lerch G, Lindner KH, Georgieff M: Large dose administration of 6% hydroxyethyl starch 200/0.5 for total hip arthroplasty: plasma homeostasis, hemostasis and renal function compared to use of 5% human albumin. Anesth Analg 83:262–268, 1996.
Discussion was started by Scalea, who argued that it is difficult to equate the acute resuscitation phase with the intensive care unit (resuscitation) phase. During emergency room administration of limited crystalloid or hypotensive resuscitation (to avoid detrimental effects of excessive crystalloid resuscitation), one needs to account for development of ischemia, contrasting the intensive care unit scenario, in which hemorrhage had been already halted. Scalea insisted that, given a completely different set of circumstances, one should consider these interventions as two separate resuscitation schemes, when effects of various administered fluids are to be evaluated.
Fries considered such a reasoning (i.e., use of limited resuscitation that allows development of ischemia) as fundamentally wrong, given that data from both septic shock and traumatic (presented here) shock show that acidotic/hypoxic patients die at a much greater rate. Scalea responded by pointing the discrepancy of emergency room restricted fluid resuscitation effects between Fries’ data and findings from the study of Bickell et al. (PubMed 7935634) (the first demonstrating no inferiority, whereas the latter one a markedly worse survival compared with full resuscitation). Fries indicated that this discrepancy may be due to the almost exclusive use of crystalloids for volume replacement in the study of Bickell et al.; an adequate volume replacement will never be achieved in trauma victims with an excessive blood loss who were treated exclusively with crystalloids (due to the small volume effect).
Schöchl pointed out to another controversy, namely, the use of crystalloids versus colloids during resuscitation. He stated that in Europe, colloids have been used for over 30 years, given their superiority in intravascular volume replacement. This approach has not changed even though the 1998 meta-analysis paper by Schierhout and Roberts (PubMed 9550953) demonstrated that use of colloids (vs. crystalloids) increases risk of mortality (by approximately 4%) in trauma patients. He also added that Americans routinely use much cheaper crystalloids, and this approach is at least as effective as treatment with colloids. Clearly, this discrepancy requires a careful reevaluation in the immediate future.
8 STANDARD COAGULATION TESTS ARE MISLEADING
P. I. Johansson. Capital Region Blood Bank, Rigshospitalet, Copenhagen, Denmark
Death due to trauma is the leading cause of lost life years worldwide, with hemorrhage being responsible for 30% to 40% of trauma mortality and accounting for almost 50% of the deaths the initial 24 h. On admission, 25% to 35% of trauma patients present with coagulopathy, which is associated with increase in mortality. Classically, coagulopathy is often monitored by plasma-based routine coagulation tests (RCoT) such as activated partial thromboplastin time (aPTT) and prothrombin time (PT). These assays were developed half a century ago to monitor anticoagulation therapy, but have never been validated for the prediction of hemorrhage in a clinical setting. This lack of correlation with clinically relevant coagulopathies can be explained by that plasma-based assays reflect only the small amount of thrombin formed during initiation of coagulation. Consequently, recent reviews have concluded that RCoT are inappropriate for monitoring coagulopathy or guide transfusion therapy. In 1994, the classic clotting cascade of hemostasis was challenged by the introduction of a cell-based model of hemostasis emphasizing the importance of tissue factor (TF) as the initiator of coagulation and the pivotal role of platelets for intact hemostasis. The poor correlation between RCoT and clinical bleeding in, e.g., trauma and surgery, is hence explained by this new understanding.
Basic principles of viscoelastic hemostatic assays. Viscoelastic hemostatic assay (VHA) is a method to assess the viscoelastic properties of coagulation in whole blood under low shear conditions (1–6) and gives a graphic presentation of clot formation and subsequent lysis. Blood is incubated at 37°C in a heated cup, and within the cup is suspended a pin connected to a detector system. As fibrin forms between the cup and pin, the transmitted rotation is detected at the pin, and a trace generated. The technical stability of the VHA analysis is demonstrated by day-to-day variation (CV%) of 5% to 15% for the different parameters. The three different phases of hemostasis are reflected by the VHA. Our group and others have demonstrated that the thrombin burst is reflected by the α angle (TEG/ROTEM) and determines the clot strength and stability. The ability of VHA to reflect thrombin generation has profound clinical utility because coagulation factor deficiencies secondary to, e.g., massive bleeding, dilution, consumption, and thrombocytopenia/pathy result in impaired thrombin generation and impaired clot strength. Furthermore, the results are available within a short time frame, making them relevant to clinical decision making. Also, VHA is considered gold standard for identifying increased fibrinolysis.
Viscoelastic hemostatic assay in the surgical setting. More than 20 clinical studies reporting on the benefit of using VHA when compared with RCoT to identify coagulopathy and guide transfusion therapy have been published. The studies include three randomized clinical trials and involve more than 4,500 patients undergoing major surgery. All studies performed to date report of the benefit of using VHA when compared with RCoT, evidenced by reductions in transfusion requirements and need for re-exploration and improved ability to predict the need for blood transfusion in patients with VHA-guided therapy.
VHA in trauma. More than 10 studies including more than 700 patients have evaluated VHA in trauma patients. Kaufmann et al. found in patients with blunt trauma that a hypocoagulable VHA was associated with increased Injury Severity Score, and only Injury Severity Score and VHA were predictive for early transfusion. Schreiber and colleagues found that hypercoagulability, as evaluated by VHA, was frequent (62%) in trauma patients upon arrival at the ED and that this correlated with increased thrombin-antithrombin complex generation. Activated PTT, PT, and platelet count where within normal limits and could, hence, not identify a hypercoagulable state. Carroll and colleagues addressed the acute posttraumatic coagulopathy, reported by Brohi et al., by VHA analyses of samples obtained at the scene of accident and upon arrival in the ED in 161 trauma patients. Interestingly, they found that that the clot forming parameters demonstrated hypocoagulability and correlated with fatality, whereas none of the RCoT demonstrated such a correlation indicating that VHA is more sensitive in reflecting clinically relevant coagulopathies than RCoT. Plotkin et al. reported in combat patients with penetrating trauma that VHA was a more accurate indicator of blood product requirements than PT, aPTT, and international normalized ratio. They suggested that VHA aided by platelet count and hematocrit should guide blood transfusion requirements. Carroll et al. and Levrat et al. found that hyperfibrinolysis was identified by VHA only in the most severely injured patients, and this was associated with increased mortality rate, and Kashuk et al. reported that primary fibrinolysis is integral in the acute coagulopathy of trauma.
VHA limitations. Although it is possible to adjust the temperature at which the blood sample is analyzed, VHA is routinely performed at 37°C, and therefore the effect of hypothermia will not be recognized. Also, the endothelial contribution to hemostasis is not displayed in VHA, and therefore, von Willebrand disease cannot be investigated. If these causes of abnormal bleeding can be excluded, then a normal VHA trace along with clinically significant bleeding necessitating blood transfusion is suspect of a surgical cause.
VHA conclusions. Clinical studies including more than 5,000 surgical and/or trauma patients have reported on the benefit of using VHA when compared with RCoT to identify coagulopathy and guide transfusion therapy. However, no VHA-guided transfusion therapy has been randomized prospectively and independently validated in trauma patients, and this is highly warranted.
1. Chandler WL: The thromboelastography and the thromboelastograph technique. Semin Thromb Hemost 21 (Suppl 4) 1–6, 1995.
2. Di Benedetto P, Baciarello M, Cabetti L, Martucci M, Chiaschi A, Bertini L: Thrombelastography. Present and future perspectives in clinical practice. Minerva Anestesiol 69:501–509, 2003.
3. Ganter MT, Hofer CK: Coagulation monitoring: current techniques and clinical use of viscoelastic point-of-care coagulation devices. Anesth Analg 106:1366–1375, 2008.
4. Luddington RJ: Thrombelastography/thromboelastometry. Clin Lab Haematol 27:81–90, 2005.
5. Salooja N, Perry DJ: Thrombelastography. Blood Coagul Fibrinolysis 12:327–337, 2001.
6. Johansson PI, Stissing T, Bochsen L, Ostrowski SR. Thrombelastography (TEG) in trauma. Scand J Trauma Emerg Med 17:45, 2009.
The discussion was opened by Hess, who remarked that mechanical devices currently used for monitoring of coagulopathy will be soon eclipsed by a new generation of machines (e.g., TEG, capillary flow ultrasonography) capable of measuring multiple end points in the real time. He added that this expert audience should not only clearly recognize/acknowledge this trend, but also proactively support/participate in it. This could be, e.g., accomplished by encouraging and aiding private manufacturers in development of new and more effective devices for clinical monitoring of coagulopathy.
Bufford observed that from the practical point of view, high purchase price of a new equipment, such as a new-generation TEG machine, will likely force hospitals to sacrifice/scale down on other analytical/diagnostic measures. This argument was counteracted by Spinella, Hess, and Kashuk who collectively argued that an improved multichannel TEG device would successfully replace many, if not all, of the classic coagulation tests, making such new-generation equipment relatively cost-effective. They additionally pointed out that major savings can be expected from an optimized blood conservation/use because blood products are the most expensive commodity for hospitals.
Along the same line, Hunt observed that the classic old tests should not be hastily abolished given that there is no validated new equipment that could be considered as a viable alternative. Rossaint expressed a desire for more studies on the economic issue, e.g., whether a modern ROTEM diagnostic would be a realistic money saver. The latter discussant additionally stressed that introduction of modern equipment should be simultaneous with emergence of individual, personalized medicine. The greatest need, he stressed, exists for monitoring devices, which will better define the real-time status of a traumatized patient. Thanks to that, emergency room (ER) physicians will have a better idea whether a particular patient does or does not require support at any particular time during resuscitation and beyond (e.g., overuse of fibrinogen concentrates in patients who do not really require it).
Regarding the issue of classic coagulation tests, Schöchl painted an even grimmer picture, stating that no valuable diagnostic measures are available in the acute bleeding situation: the key diagnostic/monitoring element in the ER is the real-time hemostatic/coagulative capacity of a bleeding patient. Thus classic tests (i.e., PTTs and aPTTs) are completely irrelevant given that they provide a past (even if performed within a relatively short period), rather than a real-time evidence. In addition, none of the above classic tests has ever been validated/developed to guide treatment in acute bleeding situations.
Next, Görlinger returned to the initial discussion regarding the new TEG/ROTEM equipment and reconfirmed its cost-effectiveness based on the hands-on experience from his own hospital (e.g., massive transfusion costs in cardiac surgery were halved). He then stressed that upon introduction of their diagnostic system, the most savings indeed have come from the more effective use of the blood and blood-related products. Yet, he pointed out, this process should not be taken for granted, and to effectively lower the overall costs via TEG/ROTEM use, introduction of three critical points is essential:
(1) appropriate/effective diagnostic
(2) variety of products that can be effectively used for therapy and
(3) an algorithm that optimizes treatments by successfully combining diagnostic and therapeutic measures (as the most important component).
This provoked a comment from Kashuk (seconded by Görlinger a moment later), who observed that experts like Görlinger should stress that such a multielement diagnostic/treatment protocol may be at first difficult to introduce, and hospitals that are only starting to utilize this algorithmic approach may even observe a transient increase in used blood products. Boffard followed up with another example confirming that use of appropriate algorithms is the way forward, and this path should not be abandoned despite potential disappointments at initial introductory stages. In the context of the transfusion blood requirements in his hospital, he added that the initial 30% of the total nonused blood (ordered from the local blood bank) has dropped to only 2% after algorithm implementation. Görlinger shared a similar experience with regard to the FFP use (dramatic reduction of FFP ordering/use) after introduction of the point-of-care measurements in the liver transplant unit.
An illustrious comment followed from Boffard, who described the standard bench testing (SBT) as “reactive,” whereas the TEG approach as much more “proactive.” Summing up, Görlinger added that comparison of SBT data (even if performed in the OR) versus TEG and/or ROTEM-based results unequivocally shows that conventional bench testing does not reflect the real-time homeostatic situation in trauma patients. Consequently, the latter diagnostic measures emerge as much more reliable and helpful in the ER setting compared with SBT.
In the closing part of the discussion, Schöchl commented on the 2010 study by Dirks et al. (PMID 21138569), in which an algorithm-based aggressive administration of FFP and PLT did not influence mortality in the enrolled trauma population. He has found the outcome of this study relatively disappointing and speculated that the algorithm used by the authors might have been faulty. Boffard pointed out that the study had enrolled a cohort of the most severe trauma patients, and that the overall number of included patients had been too low to draw definitive conclusions.
9 DIAGNOSTIC CHANGES IN MEASURING FIBRINOGEN AND FIBRIN FORMATION
B. Sørensen, C. Fenger-Eriksen, and C. Solomon. Haemostasis Research Unit, Centre for Haemostasis and Thrombosis, Guy’s and St Thomas’ Hospital, London, United Kingdom
Measurement of plasma fibrinogen is often required in critically ill patients or massively bleeding patients being resuscitated with colloid plasma expander. A series of studies have aimed at evaluating different assays of plasma fibrinogen, as well as fibrin polymerization after in vitro dilution with commonly used plasma expanders. Recently, laboratory measurements have also been conducted in a cohort of patients undergoing cardiac surgery investigating measurement of fibrinogen before and after administration of fibrinogen concentrate.
Laboratory measurements following in vitro hemodilution (1):
Fibrinogen measurements were established in plasma samples each diluted in vitro to 30% or 50% with isotonic saline, hydroxyethyl starch (HES) 130/0.4, and human albumin. Fibrinogen levels were assessed using an antigen determination, three photo-optical Clauss methods, one mechanical Clauss method, a prothrombin-derived method, and viscoelastic measurement through thromboelastometry.
Results: Measurement of fibrinogen levels was significantly different when performed on alternate analytical platforms. By 30% and 50% dilution with HES 130/0.4, coagulation analyzers using the photo-optical Clauss methods significantly overestimated levels of fibrinogen. Dilution with human albumin did not affect fibrinogen levels except from one analyzer by 50% dilution level. Viscoelastic measurement of fibrin polymerization was reduced at both dilution levels and appeared to reflect the impairment of fibrin polymerization induced by HES 130/0.4 and to a lesser extent human albumin.
Conclusions: This study demonstrated that different automated coagulation analyzers revealed significantly different levels of fibrinogen. The presence of colloid plasma expander gave rise to erroneous high levels of fibrinogen returned from some coagulation analyzers using the method of Clauss.
Laboratory measurements on clinical blood samples—before and after administration of fibrinogen concentrate (2). Blood samples were collected after cardiopulmonary bypass (CPB) and after fibrinogen concentrate administration. Fibrin polymerization was measured by thromboelastometry. Furthermore, fibrinogen concentration was measured using photo-optical (CA-7000; Siemens Healthcare Diagnostics), mechanical (KC-10 steel ball, Schnitger and Gross hook; Amelung GmbH), and electromechanical (STA-R; Diagnostica Stago) coagulometers. Assessments included agreement between fibrinogen concentration measurements and correlations between fibrinogen concentration and thromboelastometry fibrin polymerization.
Results: After CPB, mean differences between fibrinogen concentration measurements were as follows: steel ball and hook, −0.05 to 0.2 g/L (P < 0.05); steel ball and STA-R, −0.26 to 0.45 g/L (P < 0.05); steel ball and CA-7000, 0.03 to 0.37 g/L (not significant); hook and STA-R, −0.21 to 0.4 g/L (P < 0.05); hook and CA-7000, 0.07 to 0.36 g/L (not significant); and STA-R and CA-7000, 0.29 to 0.33 g/L (P < 0.05). Correlations were significant (P < 0.001) between thromboelastometry fibrin polymerization and fibrinogen concentration determined by steel ball (r = 0.71), hook (r = 0.73), STA-R (r = 0.81), and CA-7000 (r = 0.82) coagulometers. After fibrinogen concentrate administration, agreement between fibrinogen measurement methods was severely impaired, and correlations with FIBTEM MCF were 0.39 (steel ball), 0.33 (hook), 0.59 (STA-R), and 0.33 (CA-7000).
Conclusions: Agreement between fibrinogen concentration measurement methods decreased considerably after fibrinogen concentrate administration. All methods correlated acceptably with FIBTEM MCF at the end of CPB, but not after hemostatic therapy.
The first comment of the discussion came from Redl, who asked about data validation in the Fenger-Eriksen study (presented during the talk) with regard to the platelet interference upon the reliability of FIBTEM assay. Soerensen confirmed the consistency of measurements in the aforementioned study given that the plasma-based FIBTEM assay correlated perfectly with the whole blood–based FIBTEM with simultaneous inactivation of PLTs.
Next, Soerensen provoked the audience with a blunt question whether they agree or disagree that the TEG and/or ROTEM should now be routinely used in the emergency room/operating room trauma setting. Rossaint was the one to immediately respond bringing a list of valid concerns regarding the potentially routine TEG/ROTEM use. First, he indicated that a correct interpretation of TEG/ROTEM data is rather difficult for clinicians. Next, he pointed out to a dangerous, in his personal view, yet an ongoing practice of testing various algorithms based on the TEG/ROTEM data that lack proper validation. Rossaint insisted that there must be a step-by-step approach that will lead to (i) better interpretation of generated TEG/ROTEM data, (ii) subsequent validation that the data interpretation is indeed correct, finally followed by (iii) a detailed analysis of the algorithms in use to prove or disprove their lifesaving potential/cost-effectiveness. He added that even the first elements (i.e., confirmation of diagnostic/interpretational accuracy of the TEG/ROTEM assays) have not been satisfactorily finalized yet, let alone the most crucial algorithm validation.
Soerensen generally agreed with Rossaint’s point of view and reminded closely related arguments expressed earlier by Hunt and Schöchl; both pointed out that one of the major weaknesses of the current point of care is that available tests have been neither properly standardized nor specifically developed for an acute bleeding situation. In fact, it has never been fully proven, recalling Hunt’s words, that using any algorithm-based approach is the recommended way forward. Consequently, it is also impossible to make a reliable cost-benefit analysis using the tools that have not undergone an appropriate standardization process.
Soerensen observed that this is, in fact, the best time to facilitate/initiate such standardization trials; all individuals who are users/believers of TEG/ROTEM testing should undertake an organized effort to convince manufacturers of TEG/ROTEM systems to support/participate in this process. Yet, validation efforts independent of private sector are also possible, he observed, and cited the recently initiated project with the Food and Drug Administration/National Institutes of Health (with Soerensen as one of the participants) as an encouraging example. Furthermore, Soerensen stressed that the most constructive standardization of TEG/ROTEM assay should be performed in synch with both the terminology and according to the current standards used in clinical pathology to win approval/acceptance of experts from all fields (e.g., trauma surgeons, clinical hematologist). Only such an approach will enable practical introduction of this technology into the point-of-care practice. After standardization completion, manufactures need to be identified who are capable of designing/producing assays that can provide the most robust support to validated treatment algorithms.
10 PLATELET FUNCTION IN TRAUMA
C. Solomon* and H. Schöchl†. *Department of Anaesthesiology, Intensive Care and Perioperative Medicine, Salzburg University Hospital SALK; and †Department of Anaesthesiology and Intensive Care, AUVA Trauma Centre, Salzburg, Austria
Trauma-induced coagulopathy is the result of the complex interaction between defective coagulation substrate and enzymes and the effect of hypothermia, acidosis, and hypoperfusion during shock resuscitation. Early and aggressive hemostatic resuscitation protocols have proven effective in optimizing the outcome of trauma patients with severe bleeding. In most trauma resuscitation protocols, a platelet (PLT) count of 50,000 to 100,000/μL is recommended as transfusion target. In some trauma centers, early PLT transfusion is integral part of clinical practice guidelines. However, little is known about PLT function at patient’s arrival in the hospital and its impact on transfusion and survival.
Platelets play an essential role in the coagulation process. Activation of PLTs by the subendothelial collagen results in expression of glycoprotein receptors on the PLT surface, in a shape change with exposure of pseudopodes and in the release of the content of PLT granules. The old cascade model of coagulation has been replaced by a cell-based system in which PLTs enable thrombin generation and clot propagation. The clinical relevance of PLT function has been mostly investigated in cardiology and cardiovascular surgery, but reports on PLT function following trauma are limited. Because of logistic problems, routine measurement of PLT function appears problematic in the challenging scenario of acute trauma care. Born aggregometry, the current gold standard for the assessment of PLT function, is a labor-intensive technique, which is not available in most trauma centers. A number of point-of-care measurement methods for PLT function are available, among which are PFA-100, Accumetrics VerifyNow, and the newly developed semiautomated PLT function analyzer Multiplate based on multiple electrode impedance aggregometry (MEA); MEA has proven sensitive for PLT inhibition induced by anti-PLT medication and for identifying the effects of extracorporeal circulation, of hypothermia, and of hemodilution on PLT aggregation. A further method to evaluate PLT functionality is represented by the measurement of PLT contribution to the elasticity of the whole-blood clot, as provided by the thromboelastometric analysis. The PLT component is defined as the difference in elasticity between the clot obtained following extrinsic activation with tissue factor, with and without addition of the PLT-inhibiting agent cytochalasin D.
Platelet count belongs to the routine measurement of coagulation at arrival in the emergency room. There are currently no reports on the impact of PLT count on outcome in trauma, except for the field of traumatic brain injury. Of interest, PLT count has most often been reported as normal at this time point. A study using impedance aggregometry to identify PLT function in trauma patients immediately upon arrival in the emergency room has been recently performed (1). A significant difference between survivors and nonsurvivors was observed in the aggregometry measurements performed using ADPtest and TRAPtest, as well as in PLT count and the thromboelastometric PLT component of the whole-blood clot. Furthermore, aggregometry values below the normal range at arrival in the emergency room were a predictor for mortality in these trauma patients. Deficitary PLT function appeared to be a sign of coagulopathy associated with increased mortality in trauma. The usefulness of MEA in the hemostatic management of trauma patients requires further investigation.
1. Solomon C, Traintinger S, Ziegler B, Hanke A, Rahe-Meyer N, Voelckel W, Schöchl H. Platelet function following trauma. A multiple electrode aggregometry study. Thromb Haemost 106:322–330, 2011.
Rossaint opened the discussion inquiring whether it is possible to compensate low fibrinogen with high PLT load and/or vice versa. Solomon confirmed such a possibility and disclosed her recent clinical experimental study, in which low PLT count was compensated with high concentration of fibrinogen. Solomon performed a similar study in the clinical setting; to compensate for decrease in the PLT count, hemostatic therapy in patients after major cardiovascular surgery was performed mainly with fibrinogen concentrates (despite the low to normal FIBTEM readouts).
The next question by Rossaint related to the appropriateness of desmopressin use (to improve coagulation via PLT activation) in trauma/surgical patients who are on aspirin. Solomon could not either recommend or discourage desmopressin use, given that its effectiveness and/or potential adverse effects had not been evaluated in a conclusive way in these patients. She mentioned a single study that demonstrated a slight decrease in fibrinogen concentration concurrent with desmopressin use, but there is no clear explanation for this effect. It is clear, however, that advantages of continuous aspirin therapy distinctly outweigh the risk of increased transfusion in trauma/surgical patients. Even though they appear to be at a slightly higher risk for bleeding, so far the transfusion rates in patients on aspirin have not been reported to be significantly higher (compared with nonaspirin patients). Solomon pointed out that, based on existing data, only combined therapy of clopidogrel and aspirin poses a factual risk for hemostatic disturbances.
Next, Maegele inquired about the shape of PLT count trajectory in her patients on arrival because his intensive care unit patients demonstrate a substantial PLT drop (approximately 50%) at admission. Solomon stated that the PLT count decrease and its critical level on arrival are most likely a consequence of hemostatic interventions and of simultaneous PLT consumption during trauma. Martini followed by asking for the best way to asses PLT function exclusively with ROTEM. Solomon explained that ROTEM is a suboptimal test because it can only describe a general potential of PLTs present in the tested specimen (given that thrombin will cause activation of the entire PLT bound in the ROTEM sample).
Görlinger reaffirmed Solomon’s statement about ROTEM as a poor tool for detection of PLT dysfunction in the intensive care unit scenario. Furthermore, he observed that desmopressin use can increase PLT activity; the same was true when tranexamic acid was administered to patients on dual anti-PLT therapy as indicated in his recent paper (PMID 20962655). The main difference: the PLT effect of tranexamic acid occurs in 20 min compared with 1 to 2 h in case of desmopressin. Görlinger observed that it is currently impossible to precisely define a minimum amount/cutoff level of PLT and fibrinogen (or the combination thereof), at which a respective therapy should be initiated. Such a hypothetical cutoff value is also very much dependent on the monitoring assays and/or algorithms utilized (because they may be more fibrinogen or PLT-oriented). The above was followed by a general statement by Solomon, who mentioned that therapeutic options for reversal of PLT dysfunction are rather limited and PLT transfusion is not very effective. Limited in vitro data show that neither clopidogrel nor aspirin-inhibited PLTs samples can benefit from addition of fresh PLTs, making the clinical question even more challenging.
Next, Rossaint asked a very practical question on the recommended course of action to be followed in patients on both aspirin and clopidogrel, and whether an early administration of PLT in such a bleeding patient is erroneous. Solomon’s recommendation was that an artificial boost of the PLT count in aspirin + clopidogrel patients should be considered as a last resort intervention. This is chiefly due to the lack of strong data confirming clinical effectiveness/efficiency of PLT transfusion in those patients; limited data showed that in patients on clopidogrel, only direct administration, or nonguided administration, of PLT concentrates improved coagulation status and/or the transfusion requirements. Solomon specified that, in the first line, all other coagulation-related parameters (e.g., fibrinogen concentration) should be improved including normalization of the clotting time (that may be prolonged because of a deficit of factors that lead to thrombin generation). Improvement of PLT count and function through transfusion should be performed after all other elements of the coagulation homeostasis had been maximally normalized. Administration of desmopressin could be also considered before transfusion of PLT. Clearly, an appropriate course of action in this cohort of patients requires further in-depth investigation.
11 HAEMOSTATIC THERAPY AND MASSIVE BLEEDING—THE ROLE OF PCC (PROTHROMBIN COMPLEX CONCENTRATE)
D. Fries. Department for General and Surgical Critical Care Medicine, Medical University Innsbruck, Austria
The aim of any hemostatic therapy is to minimize blood loss and transfusion requirements, while the amount of blood products is associated with the degree of morbidity and mortality. In trauma patients with identical Injury Severity Scores, mortality virtually doubles simply as a result of coagulopathy (1). Massive bleeding or massive transfusion in multitraumatized patients is associated with impaired coagulation. In simple terms, to achieve adequate hemostasis, a sufficient amount of thrombin and sufficient coagulable substrate are required. Key elements in coagulation are the formation of thrombin on the platelet surface and the cleavage of fibrinogen by thrombin to form fibrin (2). If sufficient thrombin is formed, it converts fibrinogen to stable fibrin, which determines the firmness of the developing clot in the presence of factor XIII (FXIII) (3).
In the event of marked blood loss, fibrinogen reaches critical values as a function of its original concentration (4) and, as a rule, more so than any other procoagulatory factor or platelets (2, 5). Already small quantities of colloids (>1,000 mL) impair, first of all, fibrin polymerization and thus clot strength (6–8). But which is the next clotting factor to reach critical values in the presence of massive blood loss and dilutional coagulopathy? Al Dieri et al. (9) investigated the relationship between clotting factor concentrations, parameters of thrombin generation, and the amount of blood loss in patients with various congenital coagulation factor deficiencies. The authors demonstrated that bleeding tendency was directly associated with the amount of thrombin generation, which varied linearly to the FII concentrations (9). Fenger-Erikson et al. pointed out that, in the case of dilutional coagulopathy, FII, FX, and FXIII besides fibrinogen decrease beyond the expected level following a 32% dilution with hydroxyethyl starch solution (10).
Prothrombin complex concentrate (PCC) is often administered in clinical practice in the case of prolonged clotting time in critically ill patients, as well as in massive bleeding situations (11). Prothrombin complex concentrate usually contains factors II, VII, IX, and X, as well as protein C and trace amounts of heparin and has been used for years to treat hereditary coagulation deficiencies and to replace coagulation factor deficiency following administration of warfarin-like anticoagulants. A further indication for the administration of PCC is the acquired factor deficiency, but only limited animal data are available on the use of modern PCC preparations in pigs exhibiting acquired coagulation factor deficiencies caused by massive blood loss and administration of HES (12–14). Staudinger et al. (15) investigated the effect of PCC on plasma coagulation in critically ill patients and pointed out that a dose of 2,000 factor IX units of PCC (mean 30 IU/kg body weight) normalized prothrombin time (PT) by raising the plasma level of coagulation factors II, VII, IX, and X in patients with moderately reduced coagulation activity. However, PCC induced a stronger and much longer lasting effect with regard to thrombin generation than did recombinant activated coagulation factor VII (14), which might be associated with an increased risk for thromboembolic complications, especially in patients with acquired coagulation defects (16–18).
In a prospective observational study in 57 patients undergoing cardiopulmonary bypass, FII decreased significantly from baseline, FVII was unchanged, and FIX was increased 2 h after surgery. However, both FVII and FIX are contained in PCC (19). Administration of further clotting factors (FVII, FIX, FX) is probably redundant in normalizing increased clotting times in severe dilutional coagulopathy and may confer an increased risk for developing thromboembolic events.
1. Brohi K, Singh J, Heron M, Coats T: Acute traumatic coagulopathy. J Trauma 54:1127–1130, 2003.
2. Hiippala ST, Myllyla GJ, Vahtera EM: Hemostatic factors and replacement of major blood loss with plasma-poor red cell concentrates. Anesth Analg 81:360–365, 1995.
3. Korte W: Fibrin monomer and factor XIII: a new concept for unexplained intraoperative coagulopathy. Hamostaseologie 26(3 Suppl 1):S30–S35, 2006.
4. Singbartl K, Innerhofer P, Radvan J, Westphalen B, Fries D, Stogbauer R, Van Aken H: Hemostasis and hemodilution: a quantitative mathematical guide for clinical practice. Anesth Analg 96:929–935, 2003.
5. McLoughlin TM, Fontana JL, Alving B, Mongan PD, Bunger R: Profound normovolemic hemodilution: hemostatic effects in patients and in a porcine model. Anesth Analg 83:459–465, 1996.
6. Fries D, Innerhofer P, Klingler A, Berresheim U, Mittermayr M, Calatzis A, Schobersberger W: The effect of the combined administration of colloids and lactated Ringer’s solution on the coagulation system: an in vitro study using thrombelastograph coagulation analysis (ROTEG). Anesth Analg 94:1280–1287, 2002.
7. Innerhofer P, Fries D, Margreiter J, Klingler A, Kuhbacher G, Wachter B, Oswald E, Salner E, Frischhut B, Schobersberger W: The effects of perioperatively administered colloids and crystalloids on primary platelet-mediated hemostasis and clot formation. Anesth Analg 95:858–865, 2002.
8. Mittermayr M, Streif W, Haas T, Fries D, Velik-Salchner C, Klingler A, Oswald E, Bach C, Schnapka-Koepf M, Innerhofer P: Hemostatic changes after crystalloid or colloid fluid administration during major orthopedic surgery: the role of fibrinogen administration. Anesth Analg 105:905–917, 2007.
9. Al Dieri R, Peyvandi F, Santagostino E, Giansily M, Mannucci PM, Schved JF, Beguin S, Hemker HC: The thrombogram in rare inherited coagulation disorders: its relation to clinical bleeding. Thromb Haemost 88:576–582, 2002.
10. Fenger-Eriksen C, Tonnesen E, Ingerslev J, Sorensen B: Mechanisms of hydroxyethyl starch–induced dilutional coagulopathy. J Thromb Haemost 7: 1099–1105, 2009.
11. Schochl H, Forster L, Woidke R, Solomon C, Voelckel W: Use of rotation thromboelastometry (ROTEM®) to achieve successful treatment of polytrauma with fibrinogen concentrate and prothrombin complex concentrate. Anaesthesia 65:199–203, 2009.
12. Dickneite G, Doerr B, Kaspereit F: Characterization of the coagulation deficit in porcine dilutional coagulopathy and substitution with a prothrombin complex concentrate. Anesth Analg 106:1070–1077, 2008.
13. Dickneite G, Pragst I: Prothrombin complex concentrate vs fresh frozen plasma for reversal of dilutional coagulopathy in a porcine trauma model. Br J Anaesth 102:345–354, 2009.
14. Dickneite G, Dorr B, Kaspereit F, Tanaka KA: Prothrombin complex concentrate versus recombinant factor VIIa for reversal of hemodilutional coagulopathy in a porcine trauma model. J Trauma 68:1151–1157, 2009.
15. Staudinger T, Frass M, Rintelen C, Quehenberger P, Wagner O, Stoiser B, Locker GJ, Laczika K, Knapp S, Watzke H: Influence of prothrombin complex concentrates on plasma coagulation in critically ill patients. Intensive Care Med 25:1105–1110, 1999.
16. Bagot CN, Cregg R, Patel RK, Shariff A, Arya R: Perioperative myocardial infarction in a patient receiving low-dose prothrombin complex concentrates. Thromb Haemost 98:1141–1142, 2007.
17. Warren O, Simon B: Massive, fatal, intracardiac thrombosis associated with prothrombin complex concentrate. Ann Emerg Med 53:758–761, 2009.
18. Kohler M, Hellstern P, Lechler E, Uberfuhr P, Muller-Berghaus G: Thromboembolic complications associated with the use of prothrombin complex and factor IX concentrates. Thromb Haemost 80:399–402, 1998.
19. Ternstrom L, Radulovic V, Karlsson M, Baghaei F, Hyllner M, Bylock A, Hansson KM, Jeppsson A: Plasma activity of individual coagulation factors, hemodilution and blood loss after cardiac surgery: a prospective observational study. Thromb Res 126:e128–e133, 2010.
Schöchl indicated that the existing guidelines advise treatment with thrombin generation stimulators at the prolongation of coagulation time (CT) at 150% of normal. However, another study (PubMed 20693180) demonstrated that CT does not correlate with the reduction of coagulation factors up to approximately 35% of normal (from which point CT increases over the threshold value in ∼80 s). The latter data prompted Schöchl’s hospital to initiate treatment whenever CT is longer than 80 s (given that at this time point coagulation factors are in the range of about 30% to 35% and must be substituted). Fries disagreed with Schöchl, stating that even if CT does not really correlate with the factor concentration, it does correlate well to the thrombin generation. He stressed again that PCC should not be administered in minor incidences but in massively bleeding trauma patients under emergency room (ER) conditions. Fries concurred, however, that there is no reliable data on the proper PCC dosage, and no escalation studies have been done to date for the above indication.
Next, the age/comorbidity issue was brought up by Kashuk. He pointed out to the emergence of elderly trauma patients who are frequently prescribed various anti-PLT or anticoagulation agents. He then queried Fries about the recommended state of the art approach for resuscitation/monitoring of those patients (receiving, e.g., coumadin before the traumatic episode), and whether PCC should be the treatment of choice before a lifesaving ER surgery. Fries confirmed that, in the EU, PCC (with calculated dosage) is strongly recommended, and it is typically followed by PT and international normalized ratio monitoring.
Görlinger additionally commented that PCC treatment very much depends on the type of surgery and the underlying problem of the reduced PT value. He explained that, e.g., the liver transplant patients frequently display very low levels of vitamin K (and other factors) and highly increased factor VIII (up to 300% of normal) that are concurrent with an activity of 10% to 20%—a situation entirely different from a trauma patients scenario. Görlinger also shared that in less severe (non-ER) surgeries in his hospital, international normalized ratios are not routinely corrected. Fries ended the discussion, adding that currently the PCC treatment approach depends much more on the physician’s experience rather than on the sound clinical evidence.
12 STOPPING THE BLEEDING: THE ROLE OF RECOMBINANT FVIIA
K. Boffard. Department of Surgery, Charlotte Maxeke Johannesburg Academic Hospital, University of the Witwatersrand, Johannesburg, South Africa
Hemorrhage is a major challenge faced by physicians treating trauma patients and is associated with high morbidity and mortality. In trauma patients, uncontrolled bleeding is the primary cause of early in-hospital mortality, accounting for approximately 40% of patients dying in hospital within 48 h of admission. Effective control of bleeding is therefore of paramount importance. Traditionally, blood transfusion is the mainstay of treatment for bleeding trauma patients. However, there is accumulating evidence to show that blood replacement is associated with an increased risk of infection and complications. In recent years, there has been a move toward reduction of blood transfusion in the management of trauma patients, and a blood replacement strategy should be implemented judiciously to achieve the primary aim of hemorrhage management. Stopping the bleeding (source control) is the key consideration in hemorrhage management.
Coagulopathy associated with traumatic injury arises through several interrelated mechanisms. These include loss and dilution of coagulation factors or platelets and metabolic disorders affecting the coagulation process, e.g., hypothermia. Optimal treatment should avoid large-volume crystalloid resuscitation and entails earlier use of alternative agents, e.g., fresh whole blood and recombinant activated coagulation factor VII (rFVIIa); rFVIIa is a potentially valuable agent for hemorrhage control. It acts at the injury site to enhance thrombin generation, leading to stable fibrin clot formation.
Two recently completed multicenter, randomized, double-blind, placebo-controlled studies (1, 2) examined the efficacy and safety of rFVIIa as an adjunct to standard therapy in critically bleeding patients, who required more than 6 U of red blood cells (RBCs) within 4 h after admission with severe trauma:
Placebo and rFVIIa groups were comparable in blunt and penetrating patient population.
Mortality and morbidity
No significant differences between placebo and rFVIIa in both blunt and penetrating trauma populations
Transfusion requirements for blunt patients (postdosing to 24 h): Significant reduction in RBCs (1.2 U), FFP (2.2 U), and total allogeneic blood (3.6 U); no differences in platelets, fibrinogen concentrate, or cryoprecipitate
Safety: The incidence of adverse and thromboembolic events was evenly distributed between the treatment and placebo groups. No statistical difference was observed for total number of severe adverse events.
The results will be presented. rFVIIa is a viable treatment option as adjunct therapy in specific critically bleeding, severely ill, multitrauma patients.
One of the main challenges in controlling postinjury coagulopathy is the difficulty in its identification and subsequent monitoring of treatment efficacy. Thromboelastography (TEG) represents an improved diagnostic modality for whole-blood analysis. Thromboelastography measures the clotting time (R value), speed of clot formation (α angle), and extent of clot formation (K time) as well as clot strength (mA: maximum amplitude). This method allows quick differentiation between a deficiency of coagulation factors, fibrinogen, and platelets, guiding the use of blood products and coagulation factor concentrates accordingly.
Guidelines for the use of rFVIIa in trauma and massive transfusion: Current transfusion guidelines recommend the use of emergency blood, if necessary. If massive transfusion is expected, leukodepleted and cross-matched blood should always be used. The establishment of a massive transfusion protocol can help optimize the delivery of blood products, reduce mortality, and conserve resources.
Adequate laboratory support should be in place to allow comprehensive evaluation of the patients’ coagulation profile, including full-blood count, platelet count, prothrombin time (PT), activated prothrombin time, and fibrinogen and D-dimer levels. Once laboratory evidence is available to confirm coagulopathy, transfusion packages, consisting of packed RBCs, FFP, and platelets at 2 U each, are issued from the blood bank to improve hemostatic competence of patients. There is a growing consensus in the target ratio of RBC/plasma/platelet transfusions at 1:1:1 (after the initial 2 U of packed red cells) to reconstitute whole blood.
After transfusion of the first 6 U of blood, laboratory tests should be repeated to evaluate treatment efficacy. The most useful adjunct laboratory examination in this situation is the TEG, which will give a specific indication of particular “shortages.”
If major bleeding persists despite the initial 6 U of blood transfusion, TEG should be conducted, and the use of rFVIIa adjunct therapy should be actively considered if the R time is longer than normal. To ensure maximal rFVIIa efficacy, attempts should be made to achieve the following: platelets more than 50 × 109/L, fibrinogen more than 1.0 g/L, pH greater than 7.20, hematocrit greater than 24%, and temperature greater than 32°C.
Ultimately, the use of blood products should be geared toward the following end points: arrest of bleeding, normalized TEG, hemoglobin 6 to 7 mmol/L (8.0–10.00 g/dL), fibrinogen less than 1.5 g/L, clotting less than 1.5 times normal, correction of severe acidosis (pH >7.3), and hypothermia (temperature >35°C).
Conclusions: In the management of massive blood loss, direct source control, attention to maintenance of normal body temperature, and use of coagulation factor concentrates guided by TEG can help bring forward improved outcomes in patients. Adjunct hemostatic agents such as rFVIIa can be an effective approach to control bleeding and reduce blood use.
1. Boffard KD, Riou B, Warren B, Choong PI, Rizoli S, Rossaint R, Axelsen M, Kluger Y, et al.: NovoSeven Trauma Study Group: Recombinant factor VIIa as adjunctive therapy for bleeding control in severely injured trauma patients: two parallel randomized, placebo-controlled, double-blind clinical trials. J Trauma 59:8–15, 2005.
2. Hauser CJ, Boffard K, Dutton R, Bernard GR, Croce MA, Holcomb JB, Leppaniemi A, Parr M, Vincent JL, Tortella BJ, et al.: CONTROL Study Group. Results of the CONTROL trial: efficacy and safety of recombinant activated factor VII in the management of refractory traumatic hemorrhage. J Trauma 69:489–500, 2010.
The first question came from Spinella, who inquired about the proper application of factor VIIa to patients. As explained by Boffard, the correct factor VII dose is 90 μg/kg, but the administration procedure is maximally simplified because of an emergency nature of factor VIIa application; the treatment decision making is based on the monitoring of the (TEG) reaction (R) time (and its decrease) rather than on meticulous calculation of the dose per patient’s body weight.
Görlinger questioned the appropriateness of the recommended FFP dose (2 U). In Görlinger’s opinion, this is highly insufficient and higher doses (i.e., at least 15–20 mL of FFP) should be given to increase thrombin generation and fibrinogen concentration. A very effective standard procedure in his hospital recommends administration of prothrombin complex concentrate whenever a decreased thrombin generation and/or increased clotting time is observed. Boffard pointed out that the patients had received a total of 4 U of FFP (a 1:1:1 treatment scheme) and that the trial distinctly indicated reduction of the FFP amount needed for resuscitation in the context of treatment with factor VIIa. Görlinger returned with yet another argument indicating that hyperfibrinolysis should be excluded before application of factor VIIa, and if this is impossible (due to lack of proper monitoring), tranexamic acid should be the treatment of choice (followed by factor VII only in case of tranexamic acid failure). Boffard fully agreed that, in the scenario of increased hyperfibrinolysis, administration of tranexamic acid would be regarded as the standard of care. Boffard also pointed out to another advantage of the FVIIa use: high cost-effectiveness. In the South African context, approximately 54,000 major trauma patients per year translate to half a million euros in tranexamic acid costs. In case of a maximally targeted factor VIIa therapy, these costs should, in Boffard’s opinion, not exceed €10,000 to €20,000/year (few to several patients/year).
Whereas Soerensen agreed with the potential importance of factor VIIa treatment in the scenario of trauma/hemorrhage, his criticism focused on the assay methodology. He pointed out that the R time used for stratification of recombinant factor VIIa treatment is typically generated in a contact-activated clotting assay (kaolin-based). Such an approach virtually disqualifies any rational assessment of factor VIIa influence upon the R time and, in consequence, any further decision making based on this particular parameter.
Next, Maegele inquired on how, and how often, the effects and success of the utilized algorithm were monitored over time. Boffard explained that they were monitored based on the blood volume transfused and cessation of bleeding in the context of frequency of the recombinant factor VIIa used (maximally three doses) in designated patients. The major advantage of this algorithm, he stressed, was utter elimination of all patients who would not have benefited from the administration of factor VIIa.
At the end of discussion, Kashuk revisited the dosing issue of recombinant factor VIIa. His question related to the appropriateness of the 100 μg/kg dose (original dosing in the trial) that was selected based on the dose approved/used for patients with hemophilia. He speculated that an effective recombinant factor VIIa dose could be in fact much lower from the one administered in the study, dramatically reducing costs of any future trial and/or treatment. Boffard immediately admitted the validity of Kashuk’s comment, adding that it will be extremely difficult to generate enough prospective data to answer this particular question. Soerensen closed the discussion with a helpful suggestion that perhaps data from the cardiac surgery field should be adapted for use in trauma/hemorrhage patients.
13 ROLE OF FIBRINOGEN
B. Sørensen, C. Fenger-Eriksen, N. Rahe-Meyer. Haemostasis Research Unit, Centre for Haemostasis and Thrombosis, Guy’s and St Thomas’ Hospital, London, United Kingdom
Fibrinogen—a potential universal hemostatic agent! A strong and likely provocative statement? However, during the past decade, a series of laboratory and experimental studies have suggested that supplementation of fibrinogen using fibrinogen concentrate may contribute to managing bleeding. Retrospective surveys and a few proof-of-principle clinical studies have indicated that fibrinogen concentrate may have a favorable effect on perioperative bleeding.
Fibrinogen is a 300-kd large protein produced in the liver. The average plasma concentration is 1.8 to 4.3 g/L. Fibrinogen plays a crucial role for regulating hemostasis by (i) facilitating platelet aggregation by bridging of glycoprotein IIb/IIIa receptors, (ii) fibrin polymerization trigger via thrombin generation, and (iii) acting as antithrombin I. In addition, fibrinogen is an acute phase reactant. Acquired deficiency of fibrinogen develops in conjunction with other disorders, such as liver disease, disseminated intravascular coagulation, and excessive bleeding. The role of levels and function of fibrinogen as a hemostatic agent in management of perioperative and traumatic hemorrhage are grossly underestimated. There may be several reasons for fibrinogen concentrate being overlooked as a potent hemostatic intervention. For decades, hematologists have set the lower threshold level of fibrinogen at 1 g/L. Unfortunately, this level has never been clinically validated, and several new publications advocate that the critical level of fibrinogen may be significantly higher. A large proportion of patients experiencing excessive bleeding are treated with colloid plasma expanders for volume substitution. The presence of colloid plasma expanders or high levels of fibrin degradation products have been reported to induce artificial false-high levels of fibrinogen when measured by the Clauss method. These phenomena may further have masked the recognizing and understanding the importance of fibrinogen in management of perioperative/traumatic bleeding. Finally, the use of plasma expanders such as colloids, gelatin, or dextrans is now known to induce a coagulopathy characterized by acquired hypofibrinogenemia and abnormal fibrin polymerization. Moreover, experimental studies, animal studies, retrospective clinical surveys, and a few proof-of-principle prospective randomized clinical studies have demonstrated excellent hemostatic effects of substitution with a fibrinogen concentrate.
This presentation will summarize experimental and clinical efficacy and safety outcome following intervention with fibrinogen concentrate. Finally, it will outline future challenges and pitfalls for discussion.
Kashuk opened the discussion by asking about the current standard protocol used of fibrinogen/fibrinogen concentrates in the presenter’s institution. Soerensen revealed that, unfortunately, there are no established/unified standard protocols in the United Kingdom yet and that fibrinogen treatment is typically recommended at discretion of a consulting hematologist on case-to-case basis.
Next, Kozlov inquired about the possible mechanism/importance of the recorded fibrinogen decrease that was much stronger compared with the magnitude of hemodilution (in trauma patients). Soerensen speculated that this was most likely due to fibrinogen molecules losing their functionality given that, e.g., an exposure of fibrinogen to hydroxyl starch (HES) significantly reduces its polymerization ability. Hess expanded on this stating that existing data reliably indicate long-lasting stability of fibrinogen in blood products and that its loss is essentially linear with hemodilution. Upon administration, however, HES interferes with formation of fibrinogen fibrils due to the presence of electrical charge (on HES) that prevents in turn linear extension of fibrinogen globules (by platelet-induced breakdown of hydrophobic bonds in fibrinogen).
Eibl asked whether the amount of soluble fibrinogen in the administered preparations is known/has been investigated. Soerensen agreed that this is an important issue given that the level of soluble fibrin monomers has been linked to an increased risk of bleeding and decrease in factor XIII activity, thus a general inability to covalently cross-link fibrin. In Soerensen’s opinion, dilutional coagulopathy appears to involve also some forms of dysfibrinogenemia in the fibrin structure. Clearly, more mechanistic studies are needed to convince the critics that it is actually the intervention with fibrinogen that exerts the hemostatic function.
Next two discussants, Martini and Kashuk, observed that it may be necessary to monitor for presence of acidosis and/or hypothermia in these trauma patients, who would be anticipated to receive fibrinogen concentrates. Given that both hypothermia and acidosis decrease availability of fibrinogen (although due to different mechanisms), dose adjustments of fibrinogen in such patients may become necessary. Soerensen again supported this notion, additionally stating that a goal-directed intervention (as formulated by Schöchl in the preceding talks) is a definite way forward, otherwise risking a very imprecise fibrinogen administration. Yet, he stressed, such a monitoring cannot rely on the fibrinogen measurements utilizing the clot-based methods. He finally added that clinical trials data leave no doubt that treatment with fibrinogen (1 g approximately €280 in the EU and $1,000 in the United States) constitutes an extremely potent hemostatic intervention and that its utility will likely eclipse current use of all other hemostatic agents.
14 FIBRINOLYSIS, ANTIFIBRINOLYTICS AND TANGLED TERMINOLOGY
B. J. Hunt. Thrombosis & Haemostasis, King’s College University and Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom
The CRASH-2 study has shown that the use of tranexamic acid (TA), an antifibrinolytic agent, reduces mortality in those with traumatic bleeding. This trial has emphasized the importance of fibrinolytic activation in causing bleeding. The widespread use of thromboelastography in monitoring traumatic coagulopathy has led to the use of a new language around fibrinolysis, which is noticeably different from the classic terminology of fibrinolysis. This author suggests that in future publications, the word “thromboelastography” or ROTEM” is prefixed before the new terminology to prevent confusion.
The largest trial to date of antifibrinolytics: the Clinical Randomisation of Antifibrinolytics in Significant Haemorrhage (CRASH-2) trial assessed the effects of administration of TA within 8 h of injury in trauma patients with, or at risk of substantial bleeding (1); 20,211 trauma patients from 40 countries were randomly assigned within 8 h of injury to either TA (1 g load, then 1 g over 8 h) or placebo. The primary outcome was in-hospital mortality within 4 weeks of injury. All-cause mortality was significantly reduced with TA (14.5% vs. 16%; relative risk [RR], 0.91; 95% confidence interval [CI], 0.85–0.97; P = 0.0035). Bleeding-related mortality was also reduced (4.9% vs. 5.7%, respectively) without an increase in fatal or nonfatal vascular occlusive events. As a consequence of this trial, tranexamic acid has been incorporated into trauma bleeding protocols worldwide. The most recent publication from the CRASH-2 group (2) showed that tranexamic acid should be given as early as possible to bleeding trauma patients. For trauma patients admitted late after injury, tranexamic acid was less effective and could be harmful. This was based on strong evidence that the effect of tranexamic acid on death due to bleeding varied according to time from injury. Treatment within an hour of injury produced an RR of 0.68 (95% CI, 0.57–0.82; P < 0.0001). Treatment given between 1 and 3 h showed an RR of 0.79 (95% CI, 0.64–0.97; P = 0.03). However treatment given after 3 h seemed to increase the risk of death due to bleeding RR of 1.44 (95% CI, 1.12–1.84; P = 0.004).
Despite the reduction in mortality, TA did not reduce transfusion requirements in CRASH-2. Why might this be? The first most obvious conclusion is that the study was not designed to detect a reduction in bleeding. Indeed the management of blood loss during trauma is not fine-tuned to compensate for losses—blood is given empirically and not against measure blood losses. Moreover, investigators were asked to use their normal practice, and the availability of blood components is variable among the 40 countries of the countries involved, and it is widely recognized that transfusion practice varies widely between units (3). Use of blood components is only a very crude measure of bleeding especially when no trigger for transfusion and stopping transfusion were given (4).
It has also been clearly established that antifibrinolytic agents reduce blood loss in patients with surgical and both normal and traumatic injury. A systematic review of randomized trials assessing tranexamic acid in patients undergoing elective surgery identified over 50 studies (5). Tranexamic acid reduced the need for blood transfusion by a third (RR, 0.61; 95% CI, 0.54–0.70). Another systematic review of randomized trials of antifibrinolytic agents given for bleeding during the postpartum period concluded that TA reduced blood loss in postpartum hemorrhage (6).
The utility of antifibrinolytics in reducing blood loss in surgery and death rate in traumatic bleeding implies that activation of fibrinolysis is important in the coagulopathy of bleeding. Fibrinolysis is responsible for clot breakdown. Traditionally, definitions of fibrinolytic activation have been separated into primary and secondary fibrinolysis: primary fibrinolysis represents increased fibrinolytic activity independent of other factors, whereas secondary fibrinolysis is a consequence of activation of coagulation and thus thrombin activation, which stimulates the endothelium to produce increased amounts of tissue plasminogen activator. Hyperfibrinolysis is the term used when fibrinolytic activity is greater than fibrin formation such that clot integrity is threatened, and there is clot breakdown (7).
Tranexamic acid (trans-4-(aminomethyl)cyclohexanecarboxylic acid) is a synthetic derivative of the amino acid lysine that competitively inhibits the activation of plasminogen to the serine protease plasmin via binding to kringle domains. Tranexamic acid is also a competitive inhibitor of tissue plasminogen activator. Tranexamic acid blocks the lysine-binding sites of plasminogen, resulting in an inhibition of plasminogen activation and fibrin binding to plasminogen and therefore to an impairment of fibrinolysis (8).
Perhaps the success of tranexamic acid in reducing mortality in traumatic bleeding should not be surprising as the potent stimuli of tissue plasminogen activator from the endothelium are adrenaline, vasopressin, histamine, hypoxic stress, and thrombin (9, 10)—all present in varying degrees in a bleeding trauma patient. And yet activation of fibrinolysis in trauma and surgery has been surprisingly poorly studied in the past. Increased levels of fibrin degradation products (usually D-dimers, which are the breakdown product of cross-linked fibrin) are measured crudely by near-patient testing to exclude venous thromboembolism and can be measured precisely by enzyme-linked immunosorbent assays. Using D-dimer enzyme-linked immunosorbent assays D-dimer levels have been shown to increase perioperatively especially in cardiac surgery and liver transplantation (11). Brohi et al. (12) showed that D-dimer concentrations are raised in trauma patients at the time of hospital admission (median prehospital time 28 min), with the highest values in the most severely injured patients. A similar study from Japan (13) also noted increased fibrin degradation products in severely injured trauma patients. Future studies to study activation of fibrinolysis in bleeding patients using modern research assays are still required. Instead, over the last few decades, most of our information on surgical and traumatic injury comes from the use of thromboelastography, where assessment has shown that thromboelastographic detection of increased fibrinolytic activation is associated with risk of death (14). However, a new language has arisen in the field in thromboelastography, which uses the old terminology in new ways (Table 1).
With the thromboelastogram (TEG), fibrinolysis is measured by the decrease in maximal amplitude over 30 min after the maximal amplitude has been reached (Lys30). The normal range is less than 7.5% The R, K, α angle, and MA variables can also be incorporated into a coagulation index (CI) as defined by the equation: CI = −0.6516R − 0.3772K + 0.1224MA + 0.0759α − 7.7922 (15). The CI functions as an overall assessment of coagulation, with values less than −3.0 said to signify a hypocoagulable sample and values over +3.0 said to signify a hypercoagulable sample. It is said that a LY30 greater than 7.5% with a CI less than 1.0 represents primary fibrinolysis; whereas a LY30 at greater than 7.5% with a CI greater than 3.0 represents secondary fibrinolysis. However, more recently, the term primary fibrinolysis has been applied when greater than 15% estimated percent lysis is detected in TEG in bleeding trauma patients (16). This was seen in one study on 34% of patients requiring massive transfusion after traumatic injury and significantly correlated with risk of death. Schöchl et al. (14) used the term hyperfibrinolysis for increased lysis on the rotational thromboelastometry (ROTEM), again greater than 15% of maximal amplitude, and such changes were associated with poor prognosis.
Thus, it seems that the trauma world is developing a different language around increased fibrinolytic activity as detected by TEG and ROTEM compared with traditional usage. While Lys30 has been shown to correlate with increased fibrinolysis with an old assay for fibrinolytic activity—the euglobulin lysis time (17), the relevance of these terms to conventional measurements such as tissue plasminogen activator levels at present is unclear, and it remains uncertain how much fibrinolysis is actually occurring undetected when the Lys30 is less than 7.5%. In view of the justified current interest in fibrinolysis and bleeding, and to prevent confusion and retain mutual understanding between the fields, this author would plea that the term “TEG” or “ROTEM” is used when describing results from thromboelastographic traces, e.g., TEG hyperfibrinolysis. Also the term primary fibrinolysis in TEG terminology sits uneasily with the classic term, perhaps the term TEG hyperfibrinolysis can be used instead?
1. CRASH-2 Trial Collaborators, Shakur H, Roberts I, Bautista R, Caballero J, Coats T, Dewan Y, El-Sayed H, Gogichaishvili T, Gupta S, Herrera J, et al.: Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 376(9734):23–32, 2010.
2. CRASH-2 Collaborators, Roberts I, Shakur H, Afolabi A, Brohi K, Coats T, Dewan Y, Gando S, Guyatt G, Hunt BJ, Morales C, et al.: The importance of early treatment with tranexamic acid in bleeding trauma patients: an exploratory analysis of the CRASH-2 randomised controlled trial. Lancet 377(9771): 1096–1101, e1–2e, 2011.
3. Hoyt DB, Dutton RP, Hauser CJ, Hess JR, Holcomb JB, Kluger Y, Mackway-Jones K, Parr MJ, Rizoli SB, Yukioka T, et al.: Management of coagulopathy in the patients with multiple injuries: results from an international survey of clinical practice. J Trauma 65:755–764, 2008.
4. Rosencher N, Zufferey P, Samama CM: Definition of major bleeding in surgery: an anaesthesiologist’s point of view. J Thromb Haemost 8:1443–1444, 2010.
5. Henry DA, Carless PA, Moxey AJ, O’Connell D, Stokes BJ, Fergusson DA, Ker K: Anti-fibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 3:CD001886, 2011.
6. Shakur H, Elbourne D, Gülmezoglu M, Alfirevic Z, Ronsmans C, Allen E, Roberts I: The WOMAN Trial (World Maternal Antifibrinolytic Trial): tranexamic acid for the treatment of postpartum haemorrhage: an international randomised, double blind placebo controlled trial. Trials 11:40, 2010.
7. Hunt BJ, Segal H: Hyperfibrinolysis [published correction appears in J Clin Pathol 1997;50:357]. J Clin Pathol 49(12):958, 1996.
8. Mannucci PM, Levi M: Prevention and treatment of major blood loss. N Engl J Med 356:2301–2311, 2007.
9. Hekman CM, Loskutoff DJ: Fibrinolytic pathways and the endothelium. Semin Haemost Thromb 13:514, 1987.
10. Cesarman-Maus G, Hajjar KA: Molecular mechanisms of fibrinolysis. Br J Haematol 129:307–321, 2005.
11. Segal HC, Hunt BJ, Cottam S, Downing A, Beard C, Francis JL, Potter D, Tan KC: Fibrinolytic activity during orthotopic liver transplantation with and without aprotinin. Transplantation 58:1356–1360, 1994.
12. Brohi K, Cohen MJ, Ganter MT, et al.: Acute coagulopathy of trauma: hypoperfusion induces systemic anticoagulation and hyperfibrinolysis. J Trauma 64:1211–1217, 2008.
13. Sawamura A, Hayahawa M, Gando S, et al.: Disseminated intravascular coagulation with a fibrinolytic phenotype predicts mortality. Thromb Res 1214:608–613, 2009.
14. Schöchl H, Frietsch T, Pavelka M, Jámbor C: Hyperfibrinolysis after major trauma: differential diagnosis of lysis patterns and prognostic value of thrombelastometry. J Trauma 67:125–131, 2009.
15. TEG 5000 User’s Manual. Skokie, IL: Hemoscope Corp.
16. Kashuk JL, Moore EE, Sawyer M, Wohlauer M, Pezold M, Barnett C, Biffl WL, Burlew CC, Johnson JL, Sauaia A: Primary fibrinolysis is integral in the pathogenesis of the acute coagulopathy of trauma. Ann Surg 252:434–442, 2010.
17. Levrat A, Gros A, Rugeri L, Inaba K, Floccard B, Negrier C, David JS: Evaluation of rotation thrombelastography for the diagnosis of hyperfibrinolysis in trauma patients. Br J Anaesth 100:792–797, 2008.
Rossaint asked, after observing that approximately 30% to 40% of trauma patients are hyperfibrinolytic, about potential adverse effects of administration of TA to a nonhyperfibrinolytic patient cohort. Davenport immediately added that this percentage may be in fact much higher given that actual biochemical definition (of hyperfibrinolysis) is based on ROTEM or TEG parameters. Hunt agreed and revealed that her unpublished data from a series of bleeding trauma patients (from Royal London Hospital) demonstrated that virtually all of them had suffered from increased fibrinolysis, and only a relatively severe fibrinolysis can be detected on ROTEM.
Schöchl observed that there were no marked differences in perfusion requirements between the trial groups (both received equal amount of RBCs) and inquired about any other effects of TA in treated patients (compared with controls). Hunt explained that the lack of differences in perfusion requirements were likely due to a typically crude process of blood replacement in emergency trauma patients. Transfusion protocols from all centers had been analyzed, revealing that the decision making for blood replacement was chiefly based on the hemoglobin levels.
Redl brought up another aspect, namely, a potential convulsion adverse effect of TA (γ-aminobutyric acid α receptor agonist), indicating that this hazard may be relatively difficult to discern in any multicenter study. Yet, in his opinion, it is of major importance given that even millimolar amounts of TA applied to the central nervous system were shown to be highly lethal (in preclinical rat models). Hunt fully agreed that the lowest effective doses should be administered and that monitoring for seizures/fits is a must in any clinical study. Yet, she simultaneously observed that because none such complications had been reported during the study, the selected doses appear to be relatively safe.
Next, Kashuk expressed an important concern about an acceptable time window for treatment with TA in trauma patients with varying levels of injury severity. Namely, he indicated that the most severe patients (with the highest mortality) receiving massive transfusions are the ones displaying very early (and transient) hyperfibrinolytic phase. Thus, they are the most desirable target population for treatment with TA. Yet, its administration beyond the time frames of this relatively short hyperfibrinolytic period may be in fact more harmful than beneficial. On the other hand, it is relatively hard to establish potential survival benefits of TA administration in nonmassive transfused patients, whose survival rate approximates 98% to 99%. Hunt explained that this will be the very subject of her upcoming paper, yet she also revealed that preliminary analyses suggest an approximate 4- to 6-h window of opportunity. She added that the use of antifibrinolytics in disseminated intravascular coagulation patients may be detrimental because of an increased accumulation of microclots after fibrinolysis had been switched off (e.g., by TA).
Bufford inquired whether the trial conclusions would have changed had the head trauma cohort been excluded; mortality of patients from developing countries with Glasgow Coma Scale score of 5 or less approaches 100%, and this might have biased the data analysis. He additionally pointed out that it may be relatively difficult to effectively select all potential subjects for treatment given the marked differences in stratification of injury severity and enormous disparity in mortality between developed and undeveloped countries. Hunt explained that following the advice of statistical consultants, subgroup analyses have been discouraged/disallowed because of an excessive “noise” generated by the multicenter design. In effect, the trial investigators have decided to focus on the main end point question (i.e., TA vs. placebo effect), approaching it not in the context of absolute values, but rather the delta (change) value. Hunt finally added that the overall similarities in human physiology warrant a relatively uniform reduction-of-mortality assumption, regardless of the geographical latitude.
15 MASSIVE TRANSFUSION PROTOCOLS ARE THE ANSWER
P. I. Johansson. Section for Transfusion Medicine, Capital Region Blood Bank, Rigshospitalet, Copenhagen, Denmark
Background: Continued hemorrhage remains a major contributor of mortality in massively transfused patients, and controversy regarding the optimal management exists. Recently, the concept of hemostatic resuscitation, i.e., providing large amount of blood products to critically injured patients in an immediate and sustained manner as part of a massive transfusion protocol, reducing the amount of crystolloid administered, has been introduced, with wide implementation of the concept of damage control (1). The rationale behind this hemostatic resuscitation concept is that circulating whole blood contains red cells, plasma, and platelets at a 1:1:1 ratio, and transfusion of plasma and platelets in an appropriate unit-for-unit ratio has been proposed as a way to both prevent and treat coagulopathy due to massive hemorrhage. The aim of the present review, therefore, is to investigate the potential effect on survival of hemostatic resuscitation with proactive administration of plasma and/or platelets in trauma patients with massive bleeding.
Methods: English databases were searched for reports of trauma patients receiving massive transfusion (≥10 red blood cells [RBCs] within ≤24 h from admission) that tested the effects of administration of plasma and/or platelets (PLTs) in relation to RBC concentrates on survival from January 2005 to November 2010. Comparison between highest versus lowest blood product ratios and 30-day mortality was performed.
Results: Nineteen retrospective studies were identified, of which four were excluded, one because of lack of treatment groups, one because of lack of data, and two were redundant publications regarding the same cohort of patients, leaving 15 studies for evaluation encompassing 3,475 patients. There were nine studies that tested the effect on survival in relation to fresh frozen plasma (FFP) or PLT-to-RBC ratio; three investigated FFP and PLT-to-RBC ratios. Three studies evaluated implementation of massive transfusion protocols with preemptive FFP and PLT administration versus historic controls. A meta-analysis of the pooled results found a substantial statistical heterogeneity (I 2 = 58%), and highest ratio was associated with a significantly decreased mortality (odds ratio, 0.49; 95% CI, 0.43–0.57; P < 0.0001) when compared with the lowest ratio.
Conclusions: Meta-analysis of retrospective studies concerning massively transfused trauma patients demonstrates a significantly lower mortality in patients treated with the highest FFP and/or PLT ratio when compared with the lowest FFP and/or PLT ratio. The optimal FFP:RBC and PLT:RBC ratios remains to be established.
1. Johansson PI, Stensballe J: Hemostatic resuscitation for massive bleeding: the paradigm of plasma and platelets. A review of the current literature. Transfusion 50:701–710, 2010.
2. Snyder CW, Weinberg JA, McGwin G Jr, et al.: The relationship of blood product ratio to mortality: survival benefit or survival bias?. J Trauma 66:358–362, 2009.
3. Riskin DJ, Tsai TC, Riskin L, Hernandez-Boussard T, Purtill M, Maggio PM, Spain DA, Brundage SI: Massive transfusion protocols: the role of aggressive resuscitation versus product ratio in mortality reduction. J Am Coll Surg 209:198–205, 2009.
4. Duchesne JC, Kimonis K, Marr AB, Rennie KV, Wahl G, Wells JE, Islam TM, Meade P, Stuke L, Barbeau JM, et al.: Damage control resuscitation in combination with damage control laparotomy: a survival advantage. J Trauma 69:46–52, 2010.
5. CRASH-2 Trial Collaborators. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 3376: 23–32, 2010.
The result of the present review indicates that early administration of plasma and PLTs, together with RBCs, is beneficial for survival in massively bleeding patients. Of particular interest is the finding that in patients with acute traumatic coagulopathy administration of transfusion packages confers a survival advantage when compared with conventional resuscitation strategies. The timing of the administration of plasma and PLTs are important as reported by Snyder et al. (2) and Riskin et al. (3), both suggesting that a survival bias is part of the association between high ratios and increased survival, and this is likely correct in those cases where prethawed plasma is not available, and rather a “catch up” approach is used. Instead, centers having trauma packages with prethawed plasma together with PLTs and RBC available for immediate use report not only improved survival but also a reduction in the rate sepsis as well as multiple organ failure (4). The present analyses do not enlighten what the appropriate FFP:RBC and PLT:RBC ratios are, although 1:2 or higher appears likely, and this warrants further investigations in prospective randomized controlled trials. None of the studies reported using tranexamic acid routinely, and with the results of the recently reported CRASH-2 study in mind, its role in the most severely injured patients appears indicated (5).
16 FFP TO RBC RATIO IS CRUCIAL: DATA FROM CIVILIAN TRAUMA
M. Maegele. Department of Trauma and Orthopedic Surgery, Cologne-Merheim Medical Center, University of Witten/Herdecke, Cologne, Germany
Uncontrolled hemorrhage is still responsible for 30% to 40% of all trauma-related deaths within the first 48 h after hospital admission. Clinical observations, together with recent research, highlighted the central role of coagulopathy in acute trauma care, and early recognition and aggressive management have been shown to substantially reduce mortality and improve outcome in severely injured bleeding patients. To date, the use of fresh frozen plasma (FFP) is an integral part of massive transfusion (MT) protocols in most trauma centers, and its early use has been advocated. Several retrospective studies have demonstrated a survival benefit for bleeding trauma patients when transfused with an early high packed red blood cell (pRBC):FFP 1:1 ratio both in civilian and military settings (1, 2).
In an extension to these works, we tested whether acute transfusion practice of pRBC:FFP 1:1 would be associated with reduced mortality in acute bleeding multiply injury. A retrospective analysis using the TR-DGU database (Trauma Registry of the Deutsche Gesellschaft für Unfallchirurgie [DGU] 2002–2006) on primary admissions with substantial injury (Injury Severity Score [ISS] >16) and MT (>10 pRBCs) was conducted. Seven hundred thirteen patients were divided into three groups, i.e., (i) pRBC:FFP greater than 1.1, (ii) pRBC:FFP 0.9 to 1.1 (1:1), and (iii) pRBC:FFP less than 0.9, and mortality rates were compared. Four hundred ninety seven (69.7%) of patients were male; the mean age was 40.1 (±18.3) years. Injury characteristics and pathophysiological state upon emergency room (ER) arrival were comparable between all three groups. Of 713 patients, 484 had undergone MT with pRBC:FFP greater than 1.1, 114 with pRBC:FFP 0.9 to 1.1 (1:1), and 115 with pRBC:FFP less than 0.9 ratios. Acute mortality (<6 h) rates for pRBC:FFP greater than 1.1, pRBC:FFP 0.9 to 1.1 (1:1), and pRBC:FFP less than 0.9 ratios were 24.6%, 9.6%, and 3.5% (P < 0.0001); 24-h mortality rates were 32.6%, 16.7%, and 11.3% (P < 0.0001), and 30-day mortality rates were 45.5%, 35.1%, and 24.3% (P < 0.001). The frequency for septic complications and organ failure was higher in the pRBC:FFP 0.9 to 1.1 (1:1) group; ventilator days and length of stays for intensive care unit (ICU) and overall in-hospital were highest in the group with pRBC:FFP less than 0.9 ratio (P < 0.0005).
Obviously, not all patients benefited to the same extent from early transfusion practice of pRBC:FFP 1:1 during acute MT. Moreover, there are several well-established risks together with FFP administration in trauma and other critical states of illness, e.g., transfusion-associated cardiocirculatory overload, acute lung injury (transfusion-related acute lung injury), transfusion-related immunomodulation, and increased susceptibility for infection.
In a second step, we tested if a predictive model for MT can be used to determine which patients after severe trauma would benefit or be harmed by the use of a high FFP:RBC ratio within the early few hours of admission. To stratify patients to their individual risk for MT, the Trauma-Associated Severe Hemorrhage (TASH) score was used. This score was recently developed by our group and has been recognized as an easy-to-calculate scoring system to predict the probability for MT as a surrogate for life-threatening hemorrhage after injury. All variables necessary to calculate the score are available within the first 15 min upon ER arrival.
The TR-DGU was again queried from 2002 to 2007 for all primary admissions (no transfers) 16 years or greater with an ISS of 9 or greater, who had received at least one blood transfusion. To minimize survivorship bias, deaths within 60 min of admission to the ER were excluded, and the amount and ratio of blood products transfused were calculated from the products used in the emergency room and/or operation room only, not including ICU. To focus on the initial resuscitation, an MT was defined as receiving 10 or more units of pRBC within the emergency room and/or operating room. We defined a high FFP:pRBC ratio as receiving above a 1:2 ratio of FFP:pRBC. To determine above what TASH score a high or low ratio was associated with mortality, we calculated the mortality at increasing TASH scores. We arbitrarily divided the population, a priori, into six groups: TASH score 0 to 8 (<10% risk of MT), 9 to 10 (10%–15% risk), 11 to 12 (16%–25% risk), 13 to 14 (26%–39% risk), 15 to 16 (40%–54% risk), and greater than 16 (>54% risk). We hypothesized that there would be a point at which there would be a significant (P < 0.05) improvement in survival when evaluating a high FFP:pRBC ratio at increasing TASH score groups. A total of 2,474 primary admissions entered into the TR-DGU between 2002 and 2007 were identified and included in this study; 1,729 (70%) patients were male; the average age was 43 (±19) years, and 93.2% were blunt injuries. The overall mean ISS was 34 (±15.4), with an overall mortality of 24.5%. The mean time of transfusions within emergency room and/or operating room until ICU admission was less than 5 h. Within the patient group with a TASH score of 15 to 16 (40%–54% predictive of MT), those who had received a high FFP:pRBC ratio (FFP:pRBC >1:2) had a relative in-hospital mortality reduction of 42.5% (P = 0.009). We then divided patients based on a TASH score of 15 or greater (n = 659), the more severely injured group, at increased risk for MT, and patients with a TASH score of less than 15 (n = n = 1,815), the less injured group, at decreased risk for MT. A high ratio of FFP:pRBC (FFP:pRBC >1:2) in the ≥15 TASH group was independently associated with survival, with an odds ratio of 2.5 (1.6–4.0), whereas this ratio in the <15 TASH group was associated with increased multiorgan failure, 47% vs. 38% (P < 0.005).
In conclusion, a predictive model of MT upon admission, e.g., TASH, might be able to rapidly identify which severe trauma patients would benefit or have increased complications from the immediate application of a high ratio of FFP:pRBC.
1. Maegele M, Lefering R, Paffrath T, Tjardes T, Simanski C, Bouillon B: Working Group on Polytrauma of the German Society of Trauma Surgery (DGU). Red-blood-cell to plasma ratios transfused during massive transfusion are associated with mortality in severe multiple injury: a retrospective analysis from the Trauma Registry of the Deutsche Gesellschaft für Unfallchirurgie. Vox Sang 95:112–119, 2008.
2. Borgman M, Spinella P, Holcomb J, Blackbourne L, Wade C, Lefering R, Bouillon B, Maegele M: The effect of FFP:RBC ratio on morbidity and mortality in trauma patients based on transfusion prediction score. Vox Sang 101:44–54, 2011.
Bahrami opened the discussion stating that it is necessary to investigate a number of other important confounders (e.g., storage age of the blood) to establish their influence upon outcomes in treated trauma patients. Maegele agreed, yet observed that it is currently too difficult to take into account all possible variables that may influence treatment and/or outcomes.
Kashuk agreed with Maegele that it is important to identify candidates for preemptive resuscitation (e.g., utilizing the newly presented TASH score), who may benefit from treatment with higher plasma ratio(s). Yet, he simultaneously observed that even a more critical objective is to determine when the treatment should be terminated, and how quickly the goal-directed approach can be initiated given that this will ultimately determine the total amount of plasma administered to the patient. Görlinger supported the latter notion, stressing that such a scoring allows initiation of the point-of-care diagnostics and the goal-directed therapy within 10 to 15 min, without a need for the time-consuming laboratory workup. Yet, he also indicated that therapy with FFP (and other blood products) is not very efficient and carries considerable risks such as increased incidences of lung injury and sepsis (3-fold increase of the latter in FFP treated patients). Görlinger additionally pointed out the extremely long time (approximately 15 h) needed to reduce international normalized ratio below the 1.4 cutoff with the use of FFP (PubMed 17215741). In his experience, he stressed, the same effect can be achieved with factor concentrate within approximately 30 min. Although treatment approaches and availability of products vary among countries, Görlinger strongly recommended the FFP use to be restricted to a surgical bleeding only.
Spinella strongly disagreed with Görlinger’s statement regarding the efficiency of FFP treatment. He stressed that existing data predominantly demonstrate reduction in mortality, even if FFP treatment tends to transiently increase incidences of organ failure. In response, Görlinger underlined that the typical 30-day outcome follow-up is, in his opinion, too short for a reliable long-term assessment of negative FFP effects. Patients with multiorgan failure and/or acute lung injury can be successfully kept alive for prolonged periods, given the current state-of-the-art of the modern ICU. Hence, these patients tend to die beyond the first month of trauma/injury. Redl shortly commented that the late mortality is much more dependent on other factors than the early resuscitation, and such a long-term follow-up may be equally deceptive.
Gabriel shifted discussion in the direction of the quality of blood products: he rhetorically asked whether anyone has ever requested/examined the internal quality data from his/her local blood bank. He then strongly stressed that there are extreme differences in the composition of FFP between blood banks because plethora of factors (such as filtration, centrifugation, thawing/freezing time, type of bags used) heavily influence its quality. Seconding Gabriel’s comment, Görlinger brought up platelets as another example and stressed that these products virtually lack standardization. Gabriel weighed in again, explaining that some quality control guidelines for transfusion products exist, but they have not been updated for decades and are certainly not patient-oriented.
Finally, Hess expressed a contrary point of view regarding the feared differences of FFP depending on location/blood bank center. He pointed out to data from the recent multicenter BEST Collaborative study (PMID 21223296, 21223295): it demonstrated remarkable homogeneity/stability of the plasma (and other) products across tested locations/centers. The only minor recorded differences were due to specimen handling. Görlinger ended the discussion by cautioning that relative homogenous stability due to, e.g., freezing/thawing cycles should not be confused with individual variations frequently observed in blood products.
17 SHOULD COMBAT CASUALTIES WITH MASSIVE BLEEDING BE RESUSCITATED WITH A HIGH RATIO OF PLASMA AND PLATELETS TO RBCs?
P. C. Spinella,*,† L. H. Blackbourne,† and J.B. Holcomb†. *Blood Systems Research Institute, San Francisco, California; and †US Army Institute of Surgical Research, San Antonio; and †University of Texas Health Science Center, Houston, Texas
Introduction: Recent conflicts in Afghanistan and Iraq have reignited interest and study in the optimal resuscitative approach to the trauma patients with massive bleeding. As a result, the concept of damage control resuscitation (DCR) has emerged. In general, DCR advocates for rapid surgical control of bleeding and the use of resuscitative products that aim to prevent or treat severe shock and coagulopathy with the goal of decreasing death from hemorrhagic shock. One aspect of DCR is hemostatic resuscitation, which calls for increased ratios of plasma and platelets to red blood cells (RBCs) for these patients with life-threatening hemorrhagic shock. Increased plasma and platelets-to-RBC ratios can be defined as ranging between 1:1 and 1:2.
Results: Retrospective data supporting that the early and increased use of plasma-to-RBC ratios was associated with improved survival by decreasing death from hemorrhage were initially published in combat casualties and have since been replicated in many civilian trauma reports. There are at least 10 recent reports that upon adjustment for confounding variables report an independent association between high plasma-to-RBC ratios and improved survival. A high ratio of platelets to RBCs has been associated with improved survival in four published reports. The populations studied have included patients of decreased and increased age as well as those with penetrating versus blunt injury mechanisms. The actual ratios of plasma and platelets to RBCs that have been associated with improved survival range from greater than 1:1 to 1:2. The time frame from admission that these ratios were measured ranges from approximately 6 to 24 h. The definition of massive transfusion ranges from 8 U of RBCs in 12 h to 10 U in 24 h.
Discussion: To achieve transfusing increased ratios of plasma to RBCs, the use of thawed plasma that is stored at 4°C to 6°C for up to 5 days has been initiated in combat support hospitals and has increased in frequency in large civilian trauma centers. There has also been an increase in the development of massive transfusion protocols, which “push” packages of plasma and platelets in predetermined ratios upon activation instead of blood products to be individually ordered or “pulled” from the blood bank. The use of thawed plasma and massive transfusion protocols has also been associated with improved survival in historical control retrospective studies. Recent data also indicate that high plasma-to-RBC ratios are associated with improved survival for patient predicted to require a massive transfusion and may or may not be harmful for patients who are not predicted to require a massive transfusion upon admission. The use of increased plasma in severe trauma patients in a few reports has been associated with increased risk of organ failure but always in cohorts that had improved survival. Other studies have reported a decreased association with organ failure.
The use of rapid coagulation monitoring devices of whole-blood samples such as ROTEM (rotational thromboelastometry) or thromboelastography has been advocated to improve upon the concept of hemostatic resuscitation and has been reported in combat casualty patients to be more accurate at predicting who will require a massive transfusion compared with standard coagulation testing. It has been simply termed ROTEM- or thromboelastography-directed hemostatic resuscitation. This concept has potential merit and requires further study to determine its applicability and ability to improve outcomes and decrease potential adverse events from the overuse of plasma or platelets in this population.
The main limitation of the retrospective studies reporting an independent association between increased plasma-to-RBC ratios and increased survival is survivorship bias. The exclusion of patients who died within the first 30 to 60 min of admission should minimize this concern, but it is still possible that the results that have been reported are influenced by survivorship bias. Only a prospective randomized trial can deconstruct this and other sources of bias. The funded large prospective randomized controlled trial by Holcomb and colleagues called the PROspective Plasma and Platelet Ratio (PROPPR) trial aims to answer the question of: does an increased empiric ratio of plasma and platelets to RBCS provided immediately upon admission improve survival in patients with severe traumatic injuries by decreasing death from hemorrhage? This important study will provide us valuable data to help us answer the question: Should combat casualties with massive bleeding be resuscitated with a high ratio of plasma and platelets to RBCs?
Conclusions: Until prospective randomized controlled data are available, the use of high ratios of plasma and platelets to RBCs for patients with severe traumatic injury with life-threatening bleeding is appropriate. It is also essential that there be constant vigilance in preventing its misuse or overapplication, in addition to intense monitoring for potential adverse effects.
The first discussant, Görlinger, agreed with Spinella’s statement that it is better to give FFP rather than to dilute patients with other alternative volume resuscitation (e.g., crystalloids, colloids). He then observed that it would be worthwhile to compare the beneficial effects of the two treatment concepts: the initiation of resuscitation with the plasma-to-blood 1:1 ratio followed by subsequent titration based on the goal-directed therapy approach (first concept presented by Spinella) versus the immediate goal-directed, therapy-based initiation of resuscitation aimed primarily at stopping fibrinolysis, inducing clot stability (with fibrinogen and PLTs) and monitoring of thrombin generation (with prothrombin complex concentrate or recombinant activated coagulation factor VII), subsequently followed by a 1:1 ratio treatment in case of an ongoing surgical bleeding (second approach utilized by Görlinger).
Spinella explained that the start of resuscitation based on the subjective empirical ratio (the first concept) has been established for very practical reasons: delay in availability of laboratory data that would allow an immediate commencement of the goal-directed therapy in qualifying patients. This slow analytical turnaround has been, in the large part, due to a general reluctance of intensivists/emergency room surgeons to apply/rely on the laboratory data combined with an overreliance upon their practical experience. Spinella eagerly admitted that resuscitation initiation with the goal-directed therapy would be his preference as well; unfortunately, such an approach has not been very feasible in his trauma service. He again stressed that his strong overall preference is to avoid the use of crystalloids/colloids, and use the fresh whole blood and/or FFP during the first 5 to 10 min of resuscitation instead.
Kashuk indicated another related issue, specifically, an utter inability of proving the survival benefit from the increased ratios treatment to be truly due to correction of coagulopathy. He largely blamed an excessive “background data noise” in the existing retrospective studies and underlined a pressing imperative for early monitoring that could potentially reveal the true mechanism of action of this treatment approach. Kashuk particularly stressed that this expert group is especially responsible for devising ways of testing/establishing optimal solutions regarding the use of (plasma to blood) ratios.
With regard to the excessive study noise, Krostl added that one has to also take into account the quality of surgical work and surgical protocols utilized in the retrospective studies. Spinella observed that physician/surgical team variations are unavoidable also in prospective studies and are typically very difficult to account for, although it is clear that optimal solutions need to be found to reduce this problem. He then pointed out that another key issue is, before any goal-directed therapy trial is launched, to establish explicit and scientifically based goals and triggers to be tested in such a study. It is not uncommon that the trial end points/parameters are selected more based on the “gut feeling” rather than on (un)available evidence. Scalea optimistically observed that the excessive noise due to quality variations can be preemptively reduced (to a certain extent) by careful selection of participating centers/institutions. In his opinion, if a relatively homogenous approach to the critical care protocol is observed by these sites, the physician-related variability will be dramatically diminished. The large volume of patients enrolled will further reduce this complication.
Next, Gabriel was interested in Spinella’s clinical experience in using the fresh whole blood (FWB), safety of its application, and any potential differences compared with treatment with different ratios. Spinella explained that as in case of the 1:1 ratio treatment concept, the utilization of FWB in the combat scenario has also a very practical root: initial unavailability of platelets (and related products) on the battlefield. Spinella’s personal experience indicated that such an approach (early resuscitation with FWB) appeared to be beneficial (also when compared with stored PLTs once they became available in 2005). The 2009 paper by Spinella et al. (PubMed 19359973) confirmed this by showing that an effective (FWB dependent) reversal of shock and coagulopathy was also associated with improved survival (only US casualties analyzed). Yet, Spinella cautioned that this was not reproduced in a larger follow-up study by Perkins et at. (similar survival in the FWB vs. apheresis PLT group; PubMed 20796254). He attributed the latter effect to an excessive variability introduced by enrolment of civilian Iraqi and enemy combatant casualties (apart from the US military personnel). Regarding the safety, Spinella agreed that limited-range testing of FWB on the field constitutes an obvious risk. To reduce it, pretesting of donors (for FWB and PLT) has been introduced by the military.
Opening the last discussion topic, Gabriel inquired whether it would be worthwhile, from the clinical point of view, to carry out a similar study but in the civilian population. Spinella supported such an idea, indicating that such a study (aiming to compare the whole blood to component-based resuscitation) is in the planning stages by a clinical group in Houston, Tex. Another argument favoring the study is the fact that approximately 15% of the US and Canadian children’s hospitals still use the whole blood for neonatal cardiac surgery (according to the most recent survey) and that the (only) prospective controlled study in pediatric surgical patients showed a significant FWB-dependent decrease in postoperative blood loss compared with component therapy (PubMed 1995100). Turning the tables, Kashuk asked Gabriel whether, from the blood banker’s point of view, the same would be feasible given the expected logistic and technical difficulties of the daily whole-blood support during execution of such a trial. Gabriel vigorously confirmed its feasibility, claiming that with the current state-of-the-art equipment/technology a fully tested and leukocyte-reduced (with minimal PLT loss) whole blood could be realistically issued within 12 h of donation.
18 1:1 Approach Supports Dilution
J. R. Hess. University of Maryland School of Medicine, Baltimore, Maryland
Damage control resuscitation (DCR) is the administration of units of red blood cells, plasma, and platelets in a 1:1:1 ratio. It is designed to limit hemodilution during the resuscitation of patients undergoing massive hemorrhage that requires the administration of blood products faster than laboratory results can be obtained. The logic of DCR is to replace all the components of shed blood simultaneously. The process does ensure that severely injured patients do not get profoundly deficient in any one component. Nevertheless, the process does provide less than full repletion of any of the individual components. The problem is that the 1:1:1 logic does not take into account the additions and losses involved in modern blood component manufacturing.
Whole blood is collected in 500 mL units from the veins of healthy donors who typically have a hematocrit of 38% to 50%, 150 to 400 × 109 platelets per liter, and all their coagulation factors at approximately 100% of their normal concentrations. However, dilution in 70 mL of anticoagulant; losses of constituents associated with the incomplete emptying of bags, tubing systems, and leukoreduction filters; further dilution in additive solutions; and, in some countries, losses of activity associated with pathogen reduction schemes all contribute to the net losses from single donor products. These losses are not immediately apparent as a unit of leukoreduced red blood cells in additive solution has a hematocrit of 55% to 60%, a unit of platelets contains platelets at five times the concentration of whole blood, and plasma is only slightly dilute. The problem comes when these components are added back together in a 1:1:1 ratio. The final solution resulting from the mixture of a unit of red cells, a unit of platelets, and a unit of plasma can have a hematocrit as low as 29, a platelet count of 88 × 109 platelets per liter, and plasma coagulation factors at 65% of their normal concentration (0.65 U/mL). The situation is made worse by the fact that typically only 90% of the RBC and 70% of the platelets circulate. This means that patients are being resuscitated with a blood replacement solution with an effective hematocrit of 26, platelet count of 55 × 109 platelets per liter, and an international normalized ratio of 1.4. These values are essentially at the transfusion triggers for the components in critically injured patients, and any additional amounts of one of the components merely dilute the other two.
Actually, it is potentially possible to do slightly better, because platelets are platelets in plasma. However, practically, platelets are usually in short supply, and using large numbers of such units in trauma patients would disrupt the care of cancer patients whose care is typically collocated in large tertiary medical centers with trauma centers. Nevertheless, considerable data suggest that cancer patients receive more prophylactic platelets than they need and that trauma patients would do better with relatively more of this scarce resource.
Several other alternatives, such as reconstituting freeze-dried units of plasma in smaller volumes of water to produce concentrated solutions or using mixtures of industrially prepared pooled plasma products such as four-factor prothrombin complex concentrates and fibrinogen concentrates, are being tried as well.
Ultimately, the solution to this problem lies in better primary hemorrhage control. If total blood loss is reduced, patients will rarely get into the problems of component-related dilutional coagulopathy.
Scalea opened the discussion with a relatively grim comment that the decision making in resuscitation treatment is frustratingly difficult given a number of seemingly unsurpassable shortcomings: a unit of red blood cells has low hematocrit, a unit of plasma contains only 65% of clotting factors on average, stored platelets are of low quality, and any volume treatment causes detrimental blood dilution. All things considered, one may be at loss regarding the optimal course of intervention in a hemorrhaging trauma patient.
Hess was more optimistic and pointed out that a number of evidence-based treatment recommendations can be made and followed in the clinical emergency practice. Namely, it has been relatively well demonstrated to date that resuscitating patients with large amounts of nonblood fluids and/or plasma poor red cells is not advisable. In contrast, administration of plasma and PLTs very early in the resuscitation appears to be of benefit. Yet, Hess had no choice but to admit severe limitations and a small margin of error of any resuscitation scenario. In his personal opinion, an effective resuscitation treatment is a combination of a wide variety of factors such as appropriate surgical technique, the speed and coordination of mobilizing/administering the blood products (and its components), appropriate monitoring, and many others. At present, the fastest way to a successful treatment leads through a maximal optimization of the emergency room/intensive care unit intervention structure in each hospital/trauma center. All things may help, he stressed, yet none of them is a perfect answer all of the time.
Kashuk observed that the choice of (resuscitative) action appears to be frequently clouded by interfering resuscitation goals or, alternatively, by lack of the clearly defined ones. The goal of volume resuscitation intertwines with the correction of coagulopathy objective, fueling further derangement of the latter upon treatment of the first (and vice versa). Kashuk expressed an opinion that more cooperative efforts should be attempted between intensivists and hematologists to devise effective targeted therapies (primarily based on replacement of specific coagulation/fibrinolysis factors). Whereas Hess agreed that the ideal is targeted therapy, he also pointed out the difficulty of target selection given that some of the potential substances are not available, whereas other ones only at a great expense and in limited quantities. Thus, the setting up of general blueprints must ensure an effective functioning of such a targeted therapy system to make it supportable both on a local and global basis.
Finally, Soerensen offered a controversial comment proposing a concept, based on which an allogeneic blood product would be no longer considered as a generous donation from one human being to another, but rather as a drug for profit. This could, in his opinion, open new commercial avenues and boost a rather meager development in the field of blood products.
19 TIME IS LIVE IN TRAUMA-INDUCED COAGULOPATHY
K. Görlinger. Clinic for Anesthesiology and Intensive Care Medicine, University Hospital, Essen, Germany
Severe bleeding, massive transfusion, and persistent coagulopathy are associated with increased morbidity and mortality in trauma and other clinical settings (1–3). However, efficacy of hemostastic therapy in trauma-induced coagulopathy is limited by timely availability of laboratory test results and availability and efficacy of blood components such as fresh frozen plasma (FFP) and platelets, as well. Shore-Lesserson et al. (6) have shown that results of point-of-care (POC) testing are available within 5 to 15 min (4–7). In contrast, the attending physician has to wait about 30 to 90 min for results from the central laboratory. This has already been shown by a TED survey at the German Anesthesiology Congress in 2007 (Figure 1). Therefore, results of coagulation tests performed in the central laboratory cannot be used for a calculated early goal-directed therapy in severe bleeding. On the other hand, POC tests such as thromboelastometry (ROTEM), thromboelastography, and whole-blood multiple electrode aggregometry (Multiplate) can be very helpful in this clinical setting especially by using early available parameters such as coagulation time and amplitude after 10 min (A10) (4, 5, 8, 9). In combination with a first-line therapy with specific coagulation factor concentrates such as fibrinogen concentrate and four-factor prothrombin complex concentrates, this results in a “bleed-to-treat time” of 20 to 30 min and enables a correction of the coagulopathy and to stop diffuse bleeding within 30 to 90 min (10–12). Furthermore, this transfusion and coagulation management strategy are associated not only with a significant reduction in transfusion requirements but also with a reduction in the incidence of thromboembolic events and mortality in several clinical settings (10, 13, 14). In contrast, therapy with FFP and platelet concentrates in most hospitals starts with a delay of about 90 min after admission at hospital and needs even by using a 1:1:1 transfusion algorithm more than 14 h to correct coagulopathy (1, 15). Furthermore, FFP transfusion is effective only by administering more than 15 mL FFP per kg body weight, which is associated with a 3- to 4-fold increase in the incidence of acute lung injury (2, 3, 16, 17).
In summary, calculated early goal-directed coagulation management based on POC diagnostics (thromboelastometry and multiple electrode aggregometry) and first-line therapy with coagulation factor concentrates (fibrinogen concentrate and prothrombin complex concentrate) and hemostatic drugs (tranexamic acid and DDAVP) seems to be effective in trauma-induced coagulopathy and other clinical settings associated with severe bleeding (18–20).
1. Gonzalez EA, Moore FA, Holcomb JB, et al.: Fresh frozen plasma should be given earlier to patients requiring massive transfusion. J Trauma 62:112–119, 2007.
2. Khan H, Belsher J, Yilmaz M, et al.: Fresh frozen plasma and platelet transfusions are associated with development of acute lung injury in critically ill medical patients. Chest 131:1308–1314, 2007.
3. Watson GA, Sperry JL, Rosengart MR, et al.: Fresh frozen plasma is independently associated with a higher risk of multiple organ failure and acute respiratory distress syndrome. J Trauma 67:221–230, 2009.
4. Görlinger K, Dirkmann D, Dusse F, et al.: Fast interpretation of thromboelastometry in non-cardiac surgery. Anesthesiology 113:A1516, 2010.
5. Goerlinger K, Dirkmann D, Hanke A, et al.: ROTEM-based algorithm for point-of-care coagulation management in visceral surgery and liver transplantation: experience of eight years and 829 LTX. Liver Tranplant 14 (Suppl 1): S203–S204, 2008.
6. Shore-Lesserson L, Manspreizer HE, DePerio M, et al.: Thromboelastography-guided transfusion algorithm reduces transfusion in complex cardiac surgery. Anesth Analg 88:312–319, 1999.
7. Toulon P, Ozier Y, Ankri A, et al.: Point-of-care versus central laboratory coagulation testing during haemorrhagic surgery. A multicenter study. Thromb Haemost 101:394–401, 2009.
8. Bolliger D, Görlinger K, Tanaka KA: Pathophysiology and treatment of coagulopathy in massive hemorrhage and hemodilution. Anesthesiology 113: 1205–1219, 2010.
9. Görlinger K, Dirkmann D, Hanke AA, Kamler M, Kottenberg E, Thielmann M, Jakob H, Peters J: First-line therapy with coagulation factor concentrates combined with point-of-care coagulation testing is associated with decreased allogeneic blood transfusion in cardiovascular surgery: a retrospective, single-center cohort study. Anesthesiology 115:1179–1191, 2011.
10. Pabinger I, Brenner B, Kalina U, et al.: Prothrombin complex concentrate (Beriplex P/N) for emergency anticoagulation reversal: a prospective multinational clinical trial. J Thromb Haemost 6:622–631, 2008.
11. Solomon C, Pichlmaier U, Schoechl H, et al.: Recovery of fibrinogen after administration of fibrinogen concentrate to patients with severe bleeding after cardiopulmonary bypass surgery. Br J Anesth 104:555–562, 2010.
12. Görlinger K, Dirkmann D, Müller-Beiβenhirtz H, et al.: Thromboelastometry-based perioperative coagulation management in visceral surgery and liver transplantation: experience of 10 years and 1105 LTX. Liver Transplant 16(Suppl 1):S86, 2010.
13. Schöchl H, Nienaber U, Hofer G, et al.: Goal-directed coagulation management of major trauma patients using thromboelastometry (ROTEM)–guided administration of fibrinogen concentrate and prothrombin complex concentrate. Crit Care 14:R55, 2010.
14. Snyder CW, Weinberg JA, McGwin G Jr, et al.: The relationship of blood product ratio to mortality: survival benefit or survival bias? J Trauma 66: 358–362, 2009.
15. Abdel-Wahab OI, Healy B, Dzik WH: Effect of fresh-frozen plasma transfusion on prothrombin time and bleeding in patients with mild coagulation abnormalities. Transfusion 46:1279–1285, 2006.
16. Dara SI, Rana R, Afessa B, et al.: Fresh frozen plasma transfusion in critically ill medical patients with coagulopathy. Crit Care Med 33:2667–2671, 2005.
17. Fries D, Innerhofer P, Schobersberger W: Time for changing coagulation management in trauma-related massive bleeding. Curr Opin Anaesthesiol 22:267–274, 2009.
18. Theusinger OM, Spahn DR, Ganter MT. Transfusion in trauma: why and how should we change our current practice? Curr Opin Anaesthesiol 22:305–312, 2009.
19. Waydhas C, Görlinger K. Coagulation management in multiple trauma. Unfallchirurg 112:942–950, 2009.
First discussant inquired about the presence/absence of the learning curve upon introduction of the algorithm(s) utilized in Görlinger’s practice (speculating about, e.g., potential overuse of concentrates). Görlinger left no doubts that an immense learning curve is indeed to be expected. The reduction in transfusion requirements achieved by Görlinger’s hospital was in fact relatively slow and gradual. This partly stemmed from different applications this algorithm method had been initially developed for: at first to determine the need for antithrombolytic treatment in the liver transplant patients, whereas the reduction of transfusion requirements for FFP has been formulated only recently. A potentially lifesaving reduction of PLT administration in the liver transplant population (an approximately 50% increase in absolute mortality due to the development of acute lung injury) is the new contemplated direction. Görlinger distinctively stressed that each specific field/problem requires development of an optimally matched algorithm.
Next, Kashuk asked Görlinger to describe his resuscitation protocol before the initiation of targeted therapy and to give details on the guidelines for selection of trauma patients with and without requirements for massive transfusion. Görlinger explained that their bottom line concept combines complete separation of coagulation management from volume therapy, with simultaneous decision making as to which of these two approaches should be initiated first. Görlinger underlined an effective/rapid communication with surgical teams as the key element for both of the aforementioned issues: e.g., in their practice, a low blood pressure period between 30 and 45 min will be tolerated in a patient, if there is a (high) certainty of stopping the bleeding within approximately 30 min. He additionally added that separation of patients for massive- versus moderate- versus small-volume transfusion has been successfully executed with the support of the TASH score (see the previous talk by Maegele) and relying on the experience of attending surgeons/anesthesiologists.
Kashuk followed up with a question regarding the frequency/experience of the DDAVP (desmopressin) and PLT use. Görlinger stated that because there are no consistent data on the effects of DDAVP use, they avoid it under the assumption that trauma-related stress produces DDAVP-like effects (due to the massive noradrenaline/adrenaline release). Similarly, PLT treatment is avoided in favor of fibrinogen. Görlinger admitted that the latter choice has been significantly influenced by an easy access/availability of fibrinogen in the emergency room situation compared with PLTs. Finally, in response to a question by Huber-Lang, Görlinger revealed that since the targeted, algorithm-based management has been appropriately optimized and followed, the total hospital savings approximate 1 million euros per year.
20 GOAL-DIRECTED COAGULATION MANAGEMENT IN TRAUMA-RELATED BLEEDING
H. Schöchl,*† C. Solomon,† W. Voelckel†. *Department of Anaesthesiology and Intensive Care, AUVA Trauma Hospital, Salzburg; †Ludwig Boltzmann Institute for Experimental and Clinical Traumatology, Vienna; and †Department of Anaesthesiology and Intensive Care, Salzburger Landeskliniken SALK, Salzburg, Austria
Approximately 40% of all trauma-related deaths are caused by exsanguination (1). Data from both military and civilian experience suggest that high-volume administration of fresh frozen plasma and red blood cells improves survival rates in trauma patients with severe bleeding (2). However, there is no universal acceptance of this “formula-driven” therapeutic approach (3). An alternative treatment strategy is based on point-of-care viscoelastic tests (rotational thromboelastometry [ROTEM], thromboelastogram [TEG]), the results of which are used to guide coagulation therapy according to each patient’s specific needs. This “theranostic” approach assesses three components of clot formation:
Improving and maintaining clot quality
Platelets, fibrinogen, and activated FXIII are the major contributors to clot quality. Clot firmness has been identified as an important determinant of bleeding in trauma patients (4, 5). A low maximum amplitude or maximum clot firmness, reported by TEG or ROTEM, respectively, appears to predict an increased requirement for blood product transfusion (4). Current European guidelines recommend supplementation of fibrinogen when plasma concentrations are in the critical range of 1.5 to 2.0 g/L (6).
Improving initiation of the coagulation process
Thrombin generation initially appears adequate in patients with trauma-induced coagulopathy. Dunbar and Chandler (7) reported 15 trauma patients with prothrombin time greater than 18 s or an international normalized ratio greater than 1.5, suggesting possible trauma-induced coagulopathy. Thrombin generation was found to be three-fold higher than in controls (P = 0.01) (7). Nevertheless, prothrombin complex concentrates (PCCs) and recombinant activated factor VII, both of which increase thrombin generation, have been used as coagulation therapy during trauma-related bleeding (8, 9). Randomized controlled studies revealed no survival benefit in trauma patients receiving recombinant activated factor VII (8). However, there have been no such trials with PCCs. The administration of fibrinogen concentrate (n = 128) along with PCCs (n = 98), in trauma patients (n = 131) who required 5 or more units of RBCs within 24 h, resulted in favorable survival rates compared with those predicted by both the trauma Injury Severity Score and the revised injury severity classification score (9). In a recent study, Schöchl et al. (10) compared outcomes among trauma patients receiving coagulation factor concentrates as hemostatic therapy with those among patients from a large trauma registry (TR-DGU) treated with FFP-based therapy. Significantly more patients in the FFP group received RBCs and PC, but there was no difference in mortality between the two treatment strategies. These results indicate the potential for goal-directed administration of clotting factor concentrates to reduce the requirement for allogeneic transfusion in the management of severe trauma-related bleeding.
Improving clot stability
Early primary fibrinolysis (hyperfibrinolysis) may be expected in one third of trauma patients with an Injury Severity Score of greater than 16; this is generally associated with massive transfusion requirements and high mortality rates (5, 11). Administration of tranexamic acid reduced the risk of death due to bleeding in trauma patients significantly when compared with placebo (4.9% vs. 5.7%, P = 0.0077) (12). As such, the use of tranexamic acid should be considered for the treatment of severely bleeding trauma patients alongside the other hemostatic agents discussed previously.
1. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read RA, Pons PT: Epidemiology of trauma deaths: a reassessment. J Trauma 38:185–193, 1995.
2. Borgman MA, Spinella PC, Perkins JG, Grathwohl KW, Repine T, Beekley AC, Sebesta J, Jenkins D, Wade CE, Holcomb JB: The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. J Trauma 63:805–813, 2007.
3. Kashuk JL, Moore EE, Johnson JL, Haenel J, Wilson M, Moore JB, Cothren CC, Biffl WL, Banerjee A, Sauaia A: Postinjury life threatening coagulopathy: is 1:1 fresh frozen plasma:packed red blood cells the answer? J Trauma 65:261–270, 2008.
4. Leemann H, Lustenberger T, Talving P, Kobayashi L, Bukur M, Brenni M, Bruesch M, Spahn DR, Keel MJ: The role of rotation thromboelastometry in early prediction of massive transfusion. J Trauma 69:1403–1409, 2010.
5. Kashuk JL, Moore EE, Sawyer M, Wohlauer M, Pezold M, Barnett C, Biffl WL, Burlew CC, Johnson JL, Sauaia A: Primary fibrinolysis is integral in the pathogenesis of the acute coagulopathy of trauma. Ann Surg 252:434–442, 2010.
6. Rossaint R, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar E, Hunt BJ, Komadina R, Nardi G, Neugebauer E, et al.: Management of bleeding following major trauma: an updated European guideline. Crit Care 14:R52, 2010.
7. Dunbar NM, Chandler WL: Thrombin generation in trauma patients. Transfusion 49:2652–2660, 2009.
8. Hauser CJ, Boffard K, Dutton R, Bernard GR, Croce MA, Holcomb JB, Leppaniemi A, Parr M, Vincent JL, Tortella BJ, et al.: Results of the CONTROL trial: efficacy and safety of recombinant activated factor VII in the management of refractory traumatic hemorrhage. J Trauma 69:489–500, 2010.
9. Schöchl H, Nienaber U, Hofer G, Voelckel W, Jambor C, Scharbert G, Kozek-Langenecker S, Solomon C: Goal-directed coagulation management of major trauma patients using thromboelastometry (ROTEM(R))-guided administration of fibrinogen concentrate and prothrombin complex concentrate. Crit Care 14:R55, 2010.
10. Schöchl H, Nienaber U, Maegele M, Hochleitner G, Primavesi F, Steiz B, Arndt C, Hanke A, Voelckel W, Solomon C: Transfusion in trauma: thromboelastometry (TEM)–guided coagulation factor concentrate-based therapy versus standard FFP-based therapy. Crit Care 15:R83, 2011.
11. Schöchl H, Frietsch T, Pavelka M, Jambor C: Hyperfibrinolysis after major trauma: differential diagnosis of lysis patterns and prognostic value of thrombelastometry. J Trauma 67:125–131, 2009.
12. Shakur H, Roberts I, Bautista R, Caballero J, Coats T, Dewan Y, El-Sayed H, Gogichaishvili T, Gupta S, Herrera J, et al.: Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 376(9734):23–32, 2010.
First, Soerensen inquired about the factual usefulness of coagulation time (CT) measurements by EXTEM as a guidance for administration of PCC and justification of the presented PCC dosing. Schöchl admitted that this assay merely hints a reduction (whenever CT exceeds 80 s) in coagulation factor activity (extrinsic pathway) and has not been yet fully validated in that regard. He further explained that the PCC dose (1,500–1,800 U at initiation of resuscitation; approximately 3,000 U in total/patient) has been largely dictated by the size/availability of manufactured PCC bags, adding that the administered PCC volumes are not strictly correlated with the magnitude of changes recorded by EXTEM. Görlinger commented that the 80-s CT cutoff time is also used in his hospital (as an indication for procoagulative treatment). Yet, he speculated that more precise calculations (e.g., to account for potentially lower concentration of tissue factor) might be more beneficial.
Next, Martini pointed out that, according to Schöchl’s data, pH neutralization alone cannot immediately correct coagulopathy in acidotic patients—a clinical finding that has fully agreed with conclusions from Martini’s preclinical studies (see the Thursday lecture by Martini). Both Schöchl and Martini agreed that while buffer therapy alone is not sufficient, an appropriately combined therapy (e.g., fibrinogen and bicarbonate) simultaneously addressing various deficits/deregulations is likely to be much more effective in treating posttraumatic coagulopathy.
Grottke stressed that CT measurements by ROTEM and R time by TEG are very imprecise at best and can only serve as a very rough estimate of thrombin generation. Schöchl again agreed, yet he observed that in the current clinical environment this method provides the best available approach to quickly estimate generation of thrombin in (poly)trauma patients. Soerensen commented that a new assay with higher sensitivity for coagulation factor deficiency would be of great use, although he immediately added that its extended measurement time would likely constitute a serious disadvantage. Görlinger additionally cautioned that because thrombin generation assays may vary in their chemical composition/protocol, the choice of an assay must be precisely matched to a given clinical question(s).
Finally, Boffard addressed the choice of measurement temperatures in ROTEM: he observed that a ROTEM assay performed at 37°C (typical protocol) does not reflect the true coagulation status of a hypothermic patient. Görlinger commented that the choice of temperature should be dictated by the formulated question: the current clotting status of the patient at his/her body temperature or presence/absence of additional deficiency of coagulation factors (at normal body temperature). Schöchl observed that the latter choice is, in his opinion, preferable, because a hypothermia-induced coagulopathy typically resolves upon rewarming, whereas any other normothermic deficiencies frequently require a more aggressive therapeutic intervention. Thus, a ROTEM assay carried out on the blood of a hypothermic patient at 37°C would enable a preemptive planning of optimal measures against deregulations that are about to occur when the patient’s body temperature returns to normal.
21 MANAGEMENT OF THE TRAUMA-INDUCED COAGULOPATHY: THE AUSTRIAN VIEW
S. Kozek-Langenecker. Evangelical Hospital Vienna, Vienna, Austria
An early goal-directed coagulation management has been recommended in 2009 by the Task Force on Perioperative Coagulation of the Austrian Society of Anaesthesiology, Resuscitation and Intensive Care (öGARI) (1, 2). The Austrian view on the management of trauma-induced coagulopathy is based on the clinical experience of anesthesiologists in many hospitals of the country where coagulation factor concentrates have been licensed for both the inherited and acquired deficiencies and where thromboelastometry has been in clinical use as a point-of-care coagulation monitoring for over 10 years. Basic science on dilutional coagulopathy and means of its correction initiated by Fries et al. (3) further promoted the early goal-directed coagulation management. Clinical Austrian experience indicated a clear benefit of a rapid and targeted correction of coagulopathy in massively bleeding patients with reduced blood loss, transfusion requirements, complications, and mortality rates compared with the conventional approach using blind and delayed delivery of procoagulant substances. Noteworthy, it is not a high quality of evidence behind the Austrian view with large randomized controlled studies but rather the result of a national learning process. Slowly retrospective case series and comparisons to international trauma registries are being published (4).
The new term “theranostic management” describes the Austrian view: potent procoagulant intervention is in “thera” and sensitive coagulation monitoring is in “nostic.” Whereas the correction of confounding factors can be administered to all trauma patients (such as the correction of hypothermia, hypocalcemia, and acidosis), the potent and multimodal procoagulant intervention must be specifically indicated on an individual basis to be efficacious and cost-effective. An individualized management according to the cell-based model of hemostasis (5) can only be put into practice if coagulation testing allows rapid detection of the relevant pathomechanisms for trauma-induced coagulopathy. Rotational thromboelastometry with its commercially available test battery permits diagnosing, e.g., hyperfibrinolysis, deficiency in fibrinogen, platelets, and/or thrombin generation within a few minutes. In case of clinical bleeding, we use, e.g., antifibrinolytic drugs, factor concentrates and hemostyptic wound dressings as indicated. Transfusion and coagulation therapy become driven by pathophysiology when using a thromboelastometry-based algorithm (6). Furthermore, the Austrian strategy emphasizes the importance of correcting coagulopathy in the “golden hour,” that is, the early stage after trauma. In anesthesia, the most potent drugs are used, e.g., for analgesia, sedation, or cardiac support. It is normative for anesthetists to use potent drugs in life-threatening bleeding of their patients. Stable procoagulant substances have a standardized and higher factor content in a small volume and are therefore more potent than the labile blood product of allogeneic plasma with a donor-dependent and lower content of coagulation factors. From a practical point of view, management based on plasma is time-consuming because of the requirements for thawing, order, and transport of plasma bags from the blood bank to the patient and administration of larger volumes through a filter. Transfusion medicine increased national awareness for the manifold complications associated with the use of plasma, thus leading to the rational consequence of using this resource only if it is clearly indicated.
1. OEGARI: Österreichische Gesellschaft für Anästhesiologie, Reanimation und Intensivmedizin. Available at: http://www.oegari.at. Accessed August 16, 2012.
2. Fries D, Innerhofer P, Perger P, Gütl M, Heil S, Hofmann N, Kneifel W, Neuner L, Pernerstorfer T, Pfanner G, et al.: Coagulation management in trauma-related massive bleeding—recommendations of the Task Force for Coagulation (AGPG) of the Austrian Society of Anesthesiology, Resuscitation and Intensive Care Medicine (OGARI). AINS 45:552–561, 2010.
3. Fries D, Krismer A, Klingler A, Streif W, Klima G, Wenzel V, Haas T, Innerhofer P: Effect of fibrinogen on reversal of dilutional coagulopathy: a porcine model. Br J Anaesth 95:172–177, 2005.
4. Schöchl H, Nienaber U, Hofer G, Voelckel W, Jambor C, Scharbert G, Kozek-Langenecker S, Solomon C: Goal-directed coagulation management of major trauma patients using thromboelastometry (ROTEM)–guided administration of fibrinogen concentrate and prothrombin complex concentrate. Crit Care 14:R55, 2010.
5. Hoffman M: A cell-based model of coagulation and the role of factor VIIa. Blood Rev 17 Suppl 1:S1–S5, 2003.
6. Holcomb JB: Traditional transfusion practices are changing. Crit Care 14:162, 2010.
Hess observed that resuscitation with selection of factors (such as I, II, VII, XI, and X) may be successful in some patients but remain ineffective in those whose deregulations are not dependent on substituted factors. He then suggested that a system for capturing and reporting potential failures of homeostasis could be helpful in detecting the latter group of patients. The gathered information could be then utilized to identify more appropriate treatment solutions for this cohort of patients. Kozek explained that to minimize the number of “nonresponders,” tests are run not only at admission but also in a sequence to monitor each patient from the laboratory standpoint to allow for necessary adjustments of the therapeutic protocol (given that coagulation derangements rapidly fluctuate). In addition, she indicated that an existing Web-based communication system with all intensive care units ensures a better insight into the treatment effectiveness (and/or its failures). Yet, Kozek stressed out that the Web system has its obvious weaknesses as it functions only as a clinical application and not a classic study.
Next, Soerensen commented on potential benefits of the substitution with factor V (coresponsible for amplification/propagation of thrombin generation), which is considered as very promising by some. Namely, he expressed his doubts about the suggested role of factor V as an important threshold and rate-limiting factor in the treatment of traumatic coagulopathy given that only minimal amounts of factor V are needed to fully correct a diminished thrombin generation. In line with Hess, Soerensen stressed that treatment of traumatic coagulopathy should not really consider substitutions with a variety of possible factors as a definitive end point, but rather regard any hemostatic intervention as a tool delivering pharmacologically effective correction. He then brought up factor VIIa as a more suitable example of this rationale (see talk by Bufford): the desired hemostatic capacity due to the pharmacological action of factor VIIa is effective only at high levels of this factor. Thus, the goal of such an intervention, he pointed out, is not merely a substitution of factor VIIa to bring it to its normal level, but effectively boosting coagulation regardless of the dose needed. Soerensen admitted that this is an ambitious and challenging goal, yet, because trauma-induced derangements are far from any normal, nontraumatic state, simple normalization of any coagulation competent substances to their physiological (pretrauma) concentrations may not be effective enough.
22 MANAGEMENT OF BLEEDING FOLLOWING MAJOR TRAUMA: AN UPDATED EUROPEAN GUIDELINE
R. Rossaint. Department of Anesthesiology, University Hospital Aachen, Aachen, Germany
Background: Evidence-based recommendations can be made with respect to many aspects of the acute management of the bleeding trauma patient, which when implemented may lead to improved patient outcomes.
Methods: The multidisciplinary Task Force for Advanced Bleeding Care in Trauma was formed in 2005 with the aim of developing a guideline for the management of bleeding following severe injury. An updated version of the guideline published by the group in 2007 was created. Recommendations were formulated using a nominal group process, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) hierarchy of evidence, and based on a systematic review of published literature.
Results: The following 31 recommendations were given and graded based on literature published:
Initial resuscitation and prevention of further bleeding
1. The time elapsed between injury and operation should be minimized for patients in need of urgent surgical bleeding control (grade 1A).
2. Adjunct tourniquet should be used to stop life-threatening bleeding from open extremity injuries in the presurgical setting (grade 1C).
Diagnosis and monitoring of bleeding
3. The physician should clinically assess the extent of traumatic hemorrhage using a combination of mechanism of injury, patient physiology, anatomical injury pattern, and the patient’s response to initial resuscitation (grade 1C).
4. Initial normoventilation of trauma patients should be used if there are no signs of imminent cerebral herniation (grade 1C).
5. Patients presenting with hemorrhagic shock and an identified source of bleeding should undergo an immediate bleeding control procedure unless initial resuscitation measures are successful (grade 1B).
6. Patients presenting with hemorrhagic shock and an unidentified source of bleeding should undergo immediate further investigation (grade 1B).
7. Early imaging (focussed assessment sonography for trauma [FAST] or computed tomography [CT]) for the detection of free fluid should be used in patients with suspected torso trauma (grade 1B).
8. Patients with significant free intra-abdominal fluid and hemodynamic instability should undergo urgent intervention (grade 1A).
9. We recommend further assessment using CT for hemodynamically stable patients who are either suspected of having torso bleeding or have a high-risk mechanism of injury (grade 1B).
10. Do not use single hematocrit measurements as an isolated laboratory marker for bleeding (grade 1B).
11. Use both serum lactate and base deficit measurements as sensitive tests to estimate and monitor the extent of bleeding and shock (grade 1B).
12. Routine practice to detect posttraumatic coagulopathy should include the measurement of international normalized ratio, activated partial thromboplastin time, fibrinogen, and platelets. International normalized ratio and activated partial thromboplastin time alone should not be used to guide hemostatic therapy (grade 1C). We suggest that thromboelastometry also be performed to assist in characterizing the coagulopathy and in guiding hemostatic therapy (grade 2C).
Rapid control of bleeding
13. Patients with pelvic ring disruption in hemorrhagic shock should undergo immediate pelvic ring closure and stabilization (grade 1B).
14. Patients with ongoing hemodynamic instability despite adequate pelvic ring stabilization should receive early preperitoneal packing, angiographic embolization, and/or surgical bleeding control (grade 1B).
15. Early bleeding control of the abdomen should be achieved using packing, direct surgical bleeding control, and local hemostatic procedures. In the exsanguinating patient, aortic cross-clamping may be used as an adjunct (grade 1C).
16. Damage control surgery should be used in the severely injured patient presenting with deep hemorrhagic shock, signs of ongoing bleeding, and coagulopathy. Additional factors that should trigger a damage control approach are hypothermia, acidosis, inaccessible major anatomical injury, a need for time-consuming procedures, or concomitant major injury outside the abdomen (grade 1C).
17. Use topical hemostatic agents in combination with other surgical measures or with packing for venous or moderate arterial bleeding associated with parenchymal injuries (grade 1B).
Tissue oxygenation, fluid, and hypothermia
18. Aim at a target systolic blood pressure of 80 to 100 mmHg until major bleeding has been stopped in the initial phase following trauma without brain injury (grade 1C).
19. Crystalloids should be applied initially to treat the bleeding trauma patient (grade 1B). Additional colloids should be considered within the prescribed limits for each solution in hemodynamically unstable patients (grade 2C).
20. Use measures to reduce heat loss, and warm the hypothermic patient to achieve and maintain normothermia (grade 1C).
Management of bleeding and coagulation
21. Aim at a target hemoglobin of 7 to 9 g/dL (grade 1C).
22. Monitor and use measures to support coagulation and initiate as early as possible (grade 1C).
23. Levels of ionized calcium should be monitored during massive transfusion (grade 1C). Calcium chloride should be administered during massive transfusion if ionized calcium levels are low, or electrocardiographic changes suggest hypocalcemia (grade 2C).
24. Use thawed FFP in patients with massive bleeding (grade 1B). The initial recommended dose is 10 to 15 mL/kg. Further doses will depend on coagulation monitoring and the amount of other blood products administered (grade 1C).
25. Maintain a platelet count greater than 50 × 109/L (grade 1C); however, in patients with multiple trauma who are severely bleeding or have traumatic brain injury (grade 2C), a platelet count greater than 100 × 109/L. We suggest an initial dose of four to eight platelet concentrates or one apheresis pack (grade 2C).
26. Use fibrinogen concentrate or cryoprecipitate if significant bleeding is accompanied by thromboelastometric signs of a functional fibrinogen deficit or a plasma fibrinogen level of less than 1.5 to 2.0 g/L (grade 1C). The initial fibrinogen concentrate dose may be 3 to 4 g or 50 mg/kg of cryoprecipitate, which is approximately equivalent to 15 to 20 U in a 70-kg adult. Repeat doses may be guided by thromboelastometric monitoring and laboratory assessment of fibrinogen levels (grade 2C).
27. Antifibrinolytic agents should be considered in the bleeding trauma patient (grade 2C). Monitor fibrinolysis in all patients, and administer antifibrinolytic agents in patients with established hyperfibrinolysis (grade 1B). Suggested dosages are tranexamic acid 10 to 15 mg/kg followed by an infusion of 1 to 5 mg/kg per hour or ε-aminocaproic acid 100 to 150 mg/kg followed by 15 mg/kg per hour. Antifibrinolytic therapy should be guided by thromboelastometric monitoring if possible and stopped once bleeding has been adequately controlled (grade 2C).
28. The use of recombinant activated coagulation factor VII may be considered if major bleeding in blunt trauma persists despite standard attempts to control bleeding and best practice use of blood components (grade 2C).
29. Prothrombin complex concentrate should be used for the emergency reversal of vitamin K–dependent oral anticoagulants (grade 1B).
30. Desmopressin should not be used routinely in the bleeding trauma patient (grade 2C). Desmopressin may be considered in refractory microvascular bleeding if the patient has been treated with platelet-inhibiting drugs such as acetylsalicylsalicylic acid (grade 2C).
31. Do not use antithrombin concentrates in the treatment of the bleeding trauma patient (grade 1C).
Conclusions: A multidisciplinary approach to management of the traumatically injured patient remains the cornerstone of optimal patient care. As the volume and level of evidence in this field accumulate, the current state-of-the-art as reflected in this guideline will need to evolve accordingly.
Redl opened the discussion stressing the critical importance of active lobbying for inclusion of trauma-related topics in the EU-funded research schemes such as the seventh and eighth European Framework Programme. He pointed out that especially this group should acknowledge their responsibility for initiation of such a promotion, and use the Wiggers-Bernard conference as a meaningful leverage platform. Rossaint fully agreed but additionally observed that apart from the joint “protrauma” lobbying, a similar pressure should be exerted by each participant individually, e.g., by directly contacting research national service bureaus (or analogous EU agencies coordinating research-lobbying activities) in his/her own country.
Next, Hess inquired about the evidence justifying the increase in recommended fibrinogen level from 100 to 150 to 200 mg in the current guidelines. He also pointed out that, in his opinion, the “administration of PLTs in the face of massive trauma” recommendation is too vague and should be explicitly defined. Rossaint admitted that the above guideline change is very approximate, given that it has been essentially based on animal studies (marked improvement of survival with higher fibrinogen concentrations); therefore, these findings must be confirmed in randomized clinical trials. With regard to the latter question, Rossaint explained that this indication entails application of high PLT dose (approximately 100,000 recommended) in a continuously bleeding patient who had failed to respond to all other obtainable hemostatic interventions. He again added that also this recommendation has not been based on indisputable scientific evidence and may be erroneous. Schöchl offered the final comment in support of the 100th PLT threshold recommended by Rossaint. He cited a recent paper by Schnüriger et al. (PubMed 20386283) demonstrating that patients with severe brain injury had a 4-fold increase in mortality whenever the PLT count had dropped below 100th.
23 THE AMERICAN VIEW: A FORMULA DRIVEN APPROACH
T. M. Scalea. R. Adams Cowley Shock Trauma Center, Baltimore, Maryland
Virtually all would agree that preventing coagulopathy after trauma is preferable to treating it and that limiting it to the smallest duration possible is optimal, should it occur. Various strategies have been disposed to try to achieve these goals. Most American trauma centers have adopted a formula-driven approach to achieve these goals.
Care in the American trauma centers is designed around rapid recognition and control of hemorrhage. Virtually all would agree that limiting shock and achieving surgical hemostasis are among the most important principles to prevent coagulopathy. In addition, optimal resuscitation is necessary. “Hypotensive resuscitation” has been shown to be as good as or better than large-volume crystalloid resuscitation that is attempted to restore blood pressure to normal. Once surgical control is achieved, resuscitating patients with a goal of normalizing lactate has been shown to be the best predictor of survival following injury.
The ready availability of blood and blood products is key to optimal resuscitation following injury. Yet, blood products may not be available in quantities necessary to achieve the desired end points. Many American trauma centers have adopted massive transfusion protocols. These protocols ensure that high volumes of blood, plasma, and platelets are continuously available. This autotomizes the system and no longer requires surgeons or anesthesiologists to remember to order blood. Blood is delivered to the operating room or resuscitation unit or intensive care unit in predesigned aliquots until the protocol is terminated. Some centers have included other adjunctive hemostatic agents such as recombinant activated factor VII into their massive transfusion protocol so that a dose is available after a prescribed number of units.
“Hemostatic resuscitation” has also become a popular form of transfusion therapy. This began in the early 2000s, when some centers recognized that badly injured patients who received massive transfusion received a unit of plasma for every unit of blood. This was in contradiction to standard transfusion practice in the United States, where plasma and platelets were not given until approximately 10 U of red blood cells had been transfused. Experience from the US military in the current war seemed to favor a 1:1:1 transfusion practice. Civilian trauma centers adopted this strategy, and this formula-driven approach is now commonly practiced in the United States. However, there is disagreement as to what the optimal ratio of red cells to plasma is. In addition, there is some controversy as to when platelets should be used and in what ratio relative to red cells and plasma.
Unfortunately, virtually all of the data in the United States are retrospective. In addition, much has been generated by single centers. It is not until a good multicenter randomized prospective trial is performed that there will be a convincing data to guide therapy.
At the Shock Trauma Center, we do not use a massive transfusion protocol, nor do we have a strict ratio that we use in badly injured patients. We rely on a clinician judgment, crisp communication between the surgeons and anesthesiologists, ongoing monitoring of hemodynamic parameters, and frequent laboratory values to guide our therapy.
Schöchl inquired about the strength of scientific evidence in support of the proposed hypotensive approach. Scalea strongly reaffirmed his recommendations stating that, in his opinion, patients with blood pressure no lower than 80 mmHg should receive no crystalloid/colloid fluids until the source of bleeding had been identified. Blood and plasma may/should be administered in the presence of ongoing hemorrhage, but only in the amounts necessary to maintain the blood pressure of approximately 80 mmHg. Scalea stressed that resuscitation to the normal blood pressure should be performed only after hemorrhage ceases, and the hypotensive (i.e., approximately 80 mmHg) period should (arguably) not exceed 3 h. He also pointed out that patients with traumatic brain injury and spinal cord injury and older than 65 years should not be subjected to the hypotensive protocol. This approach was seconded by Görlinger, who revealed that immediate full resuscitation in his hospital has been abandoned in favor of restricted volume resuscitation (and correction of coagulopathy). With regard to the latter question by Schöchl, Scalea explained that the hypotensive approach stems from numerous animal studies (demonstrating very consistent results) followed by two randomized controlled trials in human patients.
Another discussant asked whether Scalea uses/recommends catecholamines for preservation of the cardiac index. Scalea negated this method, indicating that his goal of 80 mmHg should be reached exclusively by volume treatment (i.e., blood and/or plasma). To justify it, he again recalled preclinical findings in large animals, which unequivocally demonstrated that recurrent hemorrhage (due to full resuscitation) is much more injurious compared with an even relatively prolonged period of hypotension (i.e., at approximately 80 mmHg).
Finally, Kashuk expressed his frustration about the complexity of potential resuscitation protocols, specifically regarding the dilemma of successfully merging the relatively advanced coagulation-controlling approaches used by Europeans and the hypotensive 1:1 ratio approach utilized in the United States. Scalea agreed that adaptation of all potentially beneficial elements into one optimal resuscitation protocol will be very complicated, regardless whether in Europe or the United States. He underlined, however, that the existing evidence reveals a number of principles that should be incorporated/accounted for in any resuscitative procedure: (i) elimination/minimal volumes of crystalloids; (ii) maintaining patients at the low blood pressure is acceptable, whereas an immediate full resuscitation is inappropriate; (iii) rapid reversal of posttraumatic coagulopathy is a must; and finally (iv) adaptation and use of the point-of-care testing followed by individualized resuscitation in all trauma patients are the key future avenue to be followed.