Hyperfibrinolysis After Major Trauma: Differential Diagnosis of Lysis Patterns and Prognostic Value of Thrombelastometry

Transcription

1 ORIGINAL ARTICLE Hyperfibrinolysis After Major Trauma: Differential Diagnosis of Lysis Patterns and Prognostic Value of Thrombelastometry Herbert Schöchl, MD, Thomas Frietsch, MD, Michaela Pavelka, MD, and Csilla Jámbor, MD Background: The aim of this study was to diagnose hyperfibrinolysis (HF) and its pattern using thrombelastometry and to correlate the diagnosis with mortality. Furthermore, routine laboratory based and the rotational thrombelastometry analyzer (ROTEM)-derived variables were also correlated with survival. Methods: Severe trauma patients showing HF in ROTEM were consecutively enrolled in the study. Three different HF patterns were compared: fulminant breakdown within 30 minutes, intermediate HF of 30 to 60 minutes, and late HF after 60 minutes. Injury severity score (ISS), hemodynamics, hemoglobin, hematocrit, platelet count (PC), fibrinogen, and ROTEM variables at admission were analyzed. The observed mortality was compared with the predicted trauma and injurity severity score mortality. Results: Thirty-three patients were diagnosed with HF. The mean ISS was Fulminant, intermediate, or late HF (n 11 each group) resulted in 100%, 91%, or 73% mortality, respectively, with the best prognosis for late HF (p ). The actual overall mortality of HF (88%) exceeded the predicted trauma and injurity severity score mortality (70%) (p 0.039). Lower PC ( vs ; p 0.034), ROTEM prolonged clot formation time [CFT, 359 (140/632) vs. 82 (14/190); p 0.042], and lower platelet contribution to maximum clot firmness [MCF EXTEM MCF FIBTEM, 34 (20/40) vs. 46 (40/53); p 0.026] were associated with increased mortality. Conclusion: ROTEM-based diagnosis of HF predicted outcome. Further independent predictors of death were combination of HF with hemorrhagic shock, low PC, and prolonged CFT in ROTEM. ROTEM-based point of care testing in the emergency room is thus able to identify prognostic factors such as prolonged CFT and low platelet contribution to clot firmness (MCF EX MCF FIB ) earlier than standard laboratory-based monitoring. Key Words: Multiple trauma, Hyperfibrinolysis, Thrombelastometry, Thromboelastometry, Thrombelastography, ROTEM, Coagulopathy. (J Trauma. 2009;67: ) Submitted for publication February 26, Accepted for publication August 20, Copyright 2009 by Lippincott Williams & Wilkins From the Departments of Anesthesiology and Intensive Care (H.S.) and Surgery (M.P.), AUVA Trauma Hospital, Salzburg, Austria; Clinic for Anesthesiology and Critical Care Medicine (T.F.), University Hospital Giessen, Marburg, Germany; and Clinic for Anaesthesiology (C.J.), University of Munich, Munich, Germany. Address for reprints: Csilla Jámbor, MD, Clinic for Anesthesiology, University of Munich, Max-Lebsche-Platz 32, D Munich, Germany; csilla.jambor@web.de. DOI: /TA.0b013e31818b2483 Severe trauma is the leading cause of death in the first four decades of life. 1 Analysis of a European database on severe trauma (the German Trauma Registry) has indicated that hemorrhage, massive blood transfusion, and coagulopathy are the most important factors affecting outcome. 2 Ineffective coagulation in combination with a high injury severity score (ISS) results in a mortality rate of up to 100%. 3 There are several reasons for coagulation disorders in major trauma. Major blood loss always includes the loss of clotting factors to a variable degree. Dilution-induced coagulopathy during fluid resuscitation results in critical plasma concentrations of coagulation factors. 4 Massive release of tissue factor from the site of injury extensively activates the coagulation cascade and consumes clotting factors, especially fibrinogen. 5 Furthermore, hypothermia, acidosis, hypovolemia, and hypoperfusion are frequent problems in severe trauma patients and lead to further deterioration of the coagulation process. 6 8 The breakdown of fresh clots, a phenomenon termed hyperfibrinolysis (HF), contributes to coagulopathy to an unknown degree. The incidence of HF is still unknown but has been estimated in the range of 15% to 20%. 9 HF may be underdiagnosed because routine coagulation tests are unable to detect it reliably. The gold standard for detection of HF is thrombelastography or thrombelastometry. 10,11 The aim of our study was to analyze the pattern of HF and correlate it to mortality of HF, induced by severe trauma. Furthermore, routine laboratory-based and the rotational thrombelastometry analyzer (ROTEM) parameters in survivors and nonsurvivors of HF were compared. We hypothesized that (i) HF is associated with increased ISS and mortality, (ii) the survival time of the patients is independent of the pattern of HF, and (iii) thrombelastography is superior to routine lab analysis for detection of HF. PATIENTS AND METHODS ROTEM analysis is routinely performed as a part of coagulation monitoring for all trauma alarms that request the full trauma team in the emergency room (ER). Between January 2003 and December 2007, all trauma patients with the ROTEM diagnosis of HF were consecutively enrolled in this study. HF was diagnosed, when the thrombelastography variable maximum lysis (ML) equaled 100% (Fig. 1). This reflects the complete breakdown of the clot in the thrombelastography trace (compared with a normal trace by normal coagulation status). The ROTEM device, a modification of the classical thrombelastography first described by Hartert in 1948, 12 measures the viscoelastic properties of the clot during its formation and subsequent lysis. After recalcification of the blood sample and addition of an activator such as rabbit brain tissue The Journal of TRAUMA Injury, Infection, and Critical Care Volume 67, Number 1, July

2 Schöchl et al. The Journal of TRAUMA Injury, Infection, and Critical Care Volume 67, Number 1, July 2009 Figure 1. The ROTEM technique assesses the initiation of the coagulation process, the kinetics of clot formation, and the stability or lysis of the formed clot. The initiation of coagulation, measured as clotting time (CT, s), is a function of the concentration of coagulation factors/inhibitors and shows the time to initial thrombin and fibrin formation. Clot propagation is a function of the concentration of coagulation factors/inhibitors and also the interaction of fibrinogen with platelets. It is measured as CFT (s) and is defined as the time needed from the end of the clotting time until a firm clot of 20 mm can be achieved. At this point, the color of the graph changes from purple to blue. The strength of the final clot results from firm aggregation of the platelets and formation of a stable fibrin network and is measured as the maximum width of the trace or MCF (mm). Fibinolysis can be diagnosed from abnormally large reduction of clot firmness after it has reached its maximum, i.e., increased ML (%). factor (extrinsic activation, [EXTEM]) or ellagic acid (intrinsic activation [INTEM]), the coagulation process begins in the test cell. Measurement of coagulation in the ROTEM device is performed by the insertion into the blood sample of a vertically immersed plastic pin that slowly rotates backward and forward through an angle of The rotation range of the pin is reduced after the generation of the first fibrin filaments between the pin and the wall of the test cell. The increasing adhesive forces exerted on the movement of the pin are converted and transferred to a graphical display, which plots the changes in the viscoelastic properties of the clot over time. 10 ROTEM provides complete information about the kinetics of coagulation initiation, kinetics of clot growth, clot strength, and breakdown (Fig. 1). ROTEM is a four channel system that is able to run four tests simultaneously. A set of standard reagents is used to discriminate between several potential causes of bleeding. Two basic tests using INTEM and EXTEM provide information about the general coagulation status (impaired, normal, and hypercoagulable). In an additional FIBTEM assay, platelet activity is inhibited by cytochalasin D. 13 This allows the separate evaluation of the plasmatic component of the clot, as the maximum clot firmness (MCF) in FIBTEM is independent of platelet activity. The platelet component of a 126 clot can be determined from the formula MCF EX MCF FIB (mm). HF is depicted in the thrombelastography trace as a marked reduction in the width of the trace or clot firmness; this reflects various degrees of clot breakdown after clot formation and may be expressed numerically as the ML (in %). The aprotinin thrombelastometry (APTEM) test is used for confirmation of HF. In this test, coagulation is activated by tissue factor as in the EXTEM assay, but the excess addition of aprotinin to the blood sample inhibits fibrinolysis. If the sample is hyperfibrinolytic, the clot lysis seen in INTEM and EXTEM testing (ML 15%/h) is not present in the APTEM test (14; Fig. 2). Reference ranges for the different tests were previously determined in a multicenter investigation. 13 The standard of care for ER management in our institution was in congruence with the study protocol: blood for both ROTEM point of care and routine laboratory testing was drawn immediately after placement of a central venous line on admission to the ER. Blood samples for ROTEM analysis were collected in a standard coagulation tube containing a mol/l citrate solution, resulting in a blood to citrate ratio of 9:1. ROTEM tests were performed according to the manufacturer s recommendations. Analysis was started within 5 minutes after blood sampling. For the assessment of patient s coagulation status, preliminary test results were available after a 5-minute waiting period, and the majority of the full test information after 10 to 20 minutes. When HF was detected, aprotinin therapy was initiated in a single bolus of KIU followed by continuous drip infusion of KIU/h. In parallel, laboratory analyses were performed as follows: fibrinogen levels were determined by the Clauss method using a fully automated blood coagulation analyzer (Sysmex Corporation, Kobe, Japan; normal range, mg/dl). Hemoglobin, hematocrit, and platelet count (PC) from ethylenediamine tetra-acetic acid blood samples were analyzed with an SF 3000 analyzer (Sysmex Corporation, Kobe, Japan; normal range: hemoglobin, g/dl; hematocrit, ; and PC, /nL). Age, sex, trauma type, injury pattern, ISS, coagulation results, blood pressure, and Glasgow Coma Scale score at admission were noted for all subjects. Based on the time Figure 2. Four channel ROTEM analysis. HF is seen in INTEM and EXTEM assays. Additionally, there is no measurable fibrin generation in the FIBTEM analysis. The APTEM test shows a stable clot when the antifibriolytic aprotinin was added to EXTEM Lippincott Williams & Wilkins

3 The Journal of TRAUMA Injury, Infection, and Critical Care Volume 67, Number 1, July 2009 Hyperfibrinolysis After Major Trauma course of fibrinolysis, hyperfibrinolytic subjects were assigned to three groups 14 : fulminant HF with breakdown of the clot within 30 minutes (group A), intermediate HF with breakdown between 30 and 60 minutes (group B), and late HF with breakdown of the clot after 60 minutes (group C) (Fig. 3, A to C). The clinical outcome of patients with HF was measured by mortality and survival time. Predicted mortality for each patient was estimated using trauma and injurity severity score (TRISS) methodology modified for intubated patients Groupwise and for the whole population, observed mortality was compared with the predicted TRISS mortality. Survival time was analyzed with respect to the HF pattern. Data analysis was approved by the local ethics committee. Statistical Analysis Statistical analysis was performed using Sigma Stat statistical software (Jandel, San Rafael, CA). The z-test was used to compare overall and groupwise mortality in the various groups, and Fisher s Exact Test was used to investigate the distribution of survivors and nonsurvivors. Kaplan- Meier survival curves were created for the three HF groups, and survival times were compared using the log-rank test. Student s t test or the Mann-Whitney Rank Sum Test was used to detect the differences between survivors and nonsurvivors, depending on the distribution of the data. Results were expressed as the mean SD or as the median (25th/75th percentiles) The level of significance was set to p Figure 3. (A) Fulminant HF shows an immediate breakdown of the clot within 30 minutes. (B) Intermediate HF with a breakdown of the clot between 30 and 60 minutes. (C) Late HF with a complete clot lysis after more than 60 minutes. RESULTS From January 2003 to December 2007, we found thirtythree patients with HF after severe blunt trauma (Table 1). Demographics of groups including group size, age, and sex were comparable (Table 2). The mean ISS of all study patients was 47 14, and there was a trend toward lower ISS in group C when compared with groups A and B (p 0.15) (Table 2). Patients with HF had more severe injuries. Only three patients (9%) had an ISS of 25 or less; these included a severe brain injury, an arterial injury leading to fulminant hemorrhage, and a pelvic fracture. More patients presented with moderate ISS from 26 to 49 (39%, p 0.010, z-test) or with high ISS above 50 (52%, p 0.001, z-test) than with low ISS below 25 (Fig. 4). HF Subgroups: Clot Breakdown Dynamics In 11 cases, HF was categorized as fulminant; total clot lysis occurred within 30 minutes in the INTEM and EXTEM tests, or there was no clot formation at all with recovered clot formation and stability in the APTEM test (group A). Figure 3A shows an example of such a case. Eleven patients showed intermediate HF characterized by clot breakdown occurring between 30 and 60 minutes (Fig. 3B). Eleven patients showed late breakdown of the clot, occurring after 60 minutes (Fig. 3C). Mortality of HF The overall mortality of hyperfibrinolytic subjects was 88%. In subgroup A with fulminant HF, 55% of patients died in the ER, 36% during immediate emergency surgery in the operating room (OR), and 9% in the intensive care unit (ICU) on postoperative day 1. No patients in this group survived (mortality of 100%). In subgroup B, 27% of patients died in the ER, 18% in the OR or in the ICU within a few hours after arrival, and 45% in the ICU between days 1 and 13. Only 9% of the patients survived long term (mortality of 91%). In subgroup C, 18% of patients died in the ER, 9% in the OR or shortly after admission in the ICU, 45% in the ICU between days 1 and 10, and 27% survived until hospital discharge (mortality of 73%). Subgroup analysis of survival curves (Kaplan-Meier, Fig. 5) revealed the most favorable prognosis for late HF, and the shortest survival period (less than 24 hours) for fulminant HF (p , log-rank test). The differences in mortality among the three groups, however, (100%, 91%, and 73% for groups A, B, and C, respectively) did not reach significance, most likely because of the small sample size (p for group A vs. C and p for group B vs. C, z-test). The observed mortality of patients with HF was greater than that predicted by TRISS (88% vs. 70%, p 0.039, z-test). Differences between the real and predicted mortality among subgroups were not statistically significant, again most likely because of the small sample size (Table 2). Characterization of Survivors and Nonsurvivors Mortality of HF combined with hemorrhagic shock was significantly higher than that without hemorrhagic shock (96.3% vs. 50%, p 0.014, Fisher Exact test). The mean hemoglobin level at admission was 7.8 g/dl 2.9 g/dl, hematocrit was 23.8% 8.6%, PC was 132 /nl 62 /nl, and mean fibrinogen level was Lippincott Williams & Wilkins 127

4 Schöchl et al. The Journal of TRAUMA Injury, Infection, and Critical Care Volume 67, Number 1, July 2009 TABLE 1. Patients Data for Hyperfibrinolysis Cases No. Age Sex Type of Trauma ISS Patterns of Injury HF Group Outcome 1 29 M MVA 54 TBI, thoracic trauma B Died in ER 2 32 M Fall (climbing) 34 Thoracic trauma, pelvic fracture, femur fracture A Died in ER 3 26 M MVA 54 Thoracic trauma, pelvic fracture, tibia fracture A Died in OR 4 20 M MVA 64 TBI, thoracic trauma, abdominal trauma (rupture of liver and A Died in ER spleen), severe shock 5 28 F MVA 57 Subluxation of atlanto-occipital joint, thoracic trauma, B Died in ICU (day 0) abdominal trauma (ruptured spleen) 6 27 F Fall (suicide) 75 Thoracic trauma, abdominal trauma (rupture of liver and B Died in ICU (day 0) spleen), pelvic fracture 7 22 M MVA 64 TBI, dissection of the descending aorta, abdominal trauma C Died in ICU (day 5) (rupture of the spleen) 8 72 F MVA 60 Thoracic trauma, open pelvic fracture, open bilateral femur, A Died in ER and tibia fractures 9 40 M Fall 34 Thoracic trauma, abdominal trauma A Died in ER M MVA 50 Thoracic trauma, abdominal trauma (rupture of liver and B Died in ER spleen) M Fall 34 TBI, tibia fracture C Died in ICU (day 2) F MVA 64 TBI, amputation of the lower limb C Died in ER F MVA 50 TBI, thoracic trauma, pelvic fracture, tibia fracture B Died in ICU (day 1) F MVA 54 Thoracic trauma, abdominal trauma (rupture of liver and spleen) A Died in OR M Fall (parachute) 38 Thoracic trauma, pelvic fracture, humerus fracture C Survived M Burying 54 Pelvic fracture, rupture of small bowel, femoral fracture, B Survived dissection of the popliteal artery, amputation of lower leg, massive trauma of soft tissue M Fall (climbing) 45 Thoracic trauma, abdominal trauma, (rupture of the spleen), C Died in ICU (day 10) pelvic fracture M MVA 38 TBI, thoracic trauma, pelvic fracture, tibia fracture B Died in ICU (day 2) M MVA 41 TBI, thoracic trauma, abdominal trauma (rupture of the liver) A Died in ER M MVA 34 Thoracic trauma, vertebral fracture A Died in ICU (day 1) M MVA 50 Abdominal trauma, pelvic fracture, femoral fracture, tibia B Died in ICU (day 5) fracture F MVA 41 Thoracic trauma, abdominal trauma B Died in ICU (day 13) M MVA 45 TBI, abdominal trauma, thoracic trauma, pelvic fracture C Survived F MVA 20 Pelvic fracture, tibia fracture C Survived M MVA 45 TBI, thoracic trauma B Died in ICU (day 1) F MVA 25 TBI C Died in ICU (day 1) M Fall 57 TBI, thoracic trauma, pelvic injury B Died in ER M MVA 34 TBI, vertebral fracture, resuscitation at the scene, severe C Died in ICU (day 1) shock M Fall (suicide) 34 Pelvic fracture, vertebral fracture, severe shock C Died in ER F MVA 64 Thoracic trauma, abdominal trauma (rupture of the spleen), C Died in ICU (day 0) pelvic fracture F MVA 64 TBI, thoracic trauma, abdominal trauma (rupture of liver and A Died in OR spleen), pelvic fracture M Train accident 59 TBI, thoracic trauma, femoral fracture A Died in OR M Suicide 25 Exsanguination A Died in ER MVA, motor vehicle accident; TBI, traumatic brain injury; Groups A, clot breakdown 30 min; B, min; C, 60 min. mg/dl 40 mg/dl. Hemoglobin, hematocrit, PC, and fibrinogen levels were higher in survivors than in nonsurvivors, but the difference was only statistical significant for platelets (p 0.034) (Table 3). Clot formation time (CFT) was longer (p 0.042), and platelet contribution to maximum clot firmness (MCF EX MCF FIB ) was lower (p 0.026) in nonsurvivors. There were no other significant independent predictors of survival. However, young age [24.5 (23.5/36.5) years vs (28.8/76.5) years, p 0.078] and less severe injury (mean ISS vs , 0.166) tended to predict survival. Laboratory Parameters of Different HF Pattern There were no differences in hemoglobin, hematocrit, and PC among the three HF subgroups (data not shown). However, median fibrinogen concentration was lower in groups 2009 Lippincott Williams & Wilkins

5 The Journal of TRAUMA Injury, Infection, and Critical Care Volume 67, Number 1, July 2009 Hyperfibrinolysis After Major Trauma TABLE 2. Basic Characteristics, ISS Score, Observed and Predicted (TRISS) Mortality Overall and in the Hyperfibrinolysis Groups Fulminant (Group A), Intermediate (Group B), and Late (Group C) Overall Fulminant Intermediate Late Number of patients Age (median/range) 45/ / / /22 88 Male (%) ISS (Mean SD) Mean observed mortality 88* (%) Mean predicted mortality (TRISS, %) *p 0.039; p 0.065; p 0.499; p compared with predicted mortality. Number of patients ISS Figure 4. Number of HF cases according to ISS in the study population (n 33). A [50 (45/77) mg/dl] and B [49 (44/87) mg/dl] when compared with group C [104 (85/131) mg/dl] (p for both). DISCUSSION The main finding of this prospective cohort study was an association between the pattern of HF diagnosed by ROTEM and the outcome of severe trauma in patients admitted to our hospital during a 5-year period. Additional independent predictors of death included the combination of HF with hemorrhagic shock, low PC, and prolonged CFT in ROTEM. ROTEM-based point of care testing in the ER was able to identify prognostic factors such as prolonged CFT and low platelet contribution to clot firmness (MCF EX MCF FIB ), earlier than standard laboratory based monitoring. Our results have considerable implications for the current practice of severe trauma care. Although HF occurs with an unknown incidence, we have associated the outcome of severe trauma to various forms of HF and the coexistence of hemorrhagic shock. Although the prognostic value of low perfusion in shock and coagulopathy, as determined by low PC is not surprising for severe trauma, the presence and pattern of HF has not been widely recognized in that context. Thrombelastography-based diagnosis and treatment of HF has been thoroughly discussed in liver transplantation, cardiac surgery, and other types of surgery, but not in trauma thus far. The relevance of HF to trauma remains unknown, although coagulopathy occurs in a fourth of trauma patients. 3 The consecutive documentation of HF cases in this study cannot give a reliable estimate of prevalence, because data from trauma patients without HF were not collected. The consecutive enrolment of HF cases favors a more homogenous group of patients with mostly severe trauma and higher ISS. Because the exact ISS is taken in the ER during the secondary survey, the exact incidence of HF in severe trauma patients at our center (approximately 80 full trauma alarms per year) cannot directly extracted from the admissions database. In a great number of cases, trauma cases are down-graded after the ER survey. Thus, we assume that the incidence of HF in severe trauma is approximately 8.25% (33 patients during 5 years, 80 severe trauma cases per year). A prospective study is now underway. Levrat et al. 23 recently observed five HF cases in a population of 89 consecutive trauma patients in a prospective trauma cohort at a university hospital. The incidence of HF was found to be 6%. However, all of these patients showed very severe HF with a mortality of 100%, an ISS of 75 (maximum) and no measurable fibrinogen concentration. We suggest that HF in these patients might have been a marker of a nonsurvivable injury and that 6% was actually the incidence of late stage coagulopathy or nearly dead ER admission. This study suggests that HF solely occurs in severe trauma (ISS 25), as the mean ISS in our collective was 47 (Table 2), and ISS values of less than 20 were not observed (Fig. 4). The absolute number of patients with HF increased concomitantly with the ISS (Fig. 4). However, in general trauma cohorts, higher ISS scores are not seen as frequently as moderate or low level injuries. 24 The association of higher ISS and HF is in accord with the results of Kaufmann et al., 25 who used thrombelastography for coagulation monitoring of trauma patients in the ER and did not observe HF. The mean ISS of the patients in that study was low (mean ISS 12.3). 25 These findings suggest a disproportionately higher incidence of HF for injuries with increased severity. An important result of our study, however, is that the severity of injury influences the pattern of HF. In our small study population, there was a trend toward fulminant and intermediate HF in patients with more severe injuries in groups A and B (mean ISS and 52 10, respectively). Late HF and the best prognosis were associated with group C, which presented the lowest mean ISS (mean ISS 42 16). Facing an unknown but sufficient (considering our estimation of around 8%) incidence of HF in severe trauma, the importance of early diagnosis becomes evident for daily practice, as effective treatment might have impact on outcome. In this context, the question arises as to whether routine laboratory panels can be considered standard of care, since identification of HF now seems to be the domain of thrombelastographic methods. 10,11 Gando et al. 26 failed to demonstrate D- dimers (fibrin formation and degradation), plasmin-antiplasmin complexes (plasmin generation), fibrinopeptide B (plasmin activation), or plasminogen activator inhibitor 1 (plasmin inhibition) as indicators for increased fibrinolysis 2009 Lippincott Williams & Wilkins 129

6 Schöchl et al. The Journal of TRAUMA Injury, Infection, and Critical Care Volume 67, Number 1, July fulminant HF intermediate HF late HF number of survived patients admission day 0 day 1 day 2 day 3 day 4 day 5 day 6 day 7 day 8 time day 9 day 10 day 11 day 12 day 13 day 14 day 15 longterm Figure 5. Number of surviving patients at different times between admission to the ER and the long term, grouped by HF group. p (log-rank test), Group A, fulminant HF with breakdown of the clot within 30 minutes; group B, intermediate HF with breakdown of the clot in 30 to 60 minutes; group C, late HF with breakdown of the clot 60 minutes. TABLE 3. Laboratory and ROTEM Results of Patients With Hyperfibrinolysis Survivors Nonsurvivors p Hb (g/dl) Hematocrit (%) Platelets (/nl) Fibrinogen (mg/dl) CFT INTEM (s) 82 (14/190) 359 (140/632) MCF EX MCF FIB (mm) 46 (40/53) 34 (20/40) Data are expressed as mean values SD or median (25th/75th percentile) according to the distribution of the data. after trauma. Enderson et al. 27 used plasminogen activators and D-dimers but found hypercoagulability and suppression of fibrinolysis in trauma. Sorensen 28 found only weak correlations between fibrin monomers and fibrin degradation products to ISS and transfusion requirements. The euglobulin lysis time has also been used for estimating the functional fibrinolytic capacity of plasma. 9 These indirect indicators of HF, however, are not direct measurements of clot dynamics. Furthermore, they have thus far not been associated with relevant outcome data. This study demonstrates that through direct measurement of clot formation and lysis, point of care use of ROTEM can quickly provide relevant information to supplement standard or expanded routine coagulation panels in the central laboratory. The shorter time required to obtained results (10 20 minutes as opposed to minutes) may indicate ROTEM as a favorable guide for efficacious antifibrinolytic therapy. Starting all four channels for parallel measurements takes 2 minutes for users without laboratory experience after performance of approximately 20 measurements (usually performed by the nursing staff). The first results are available after 5 to 10 minutes. The amplitude of platelet inhibited thrombelastometry (FIBTEM) after 5 or 10 minutes gives the first indication, whether fibrinogen substitution should be applied. After 20 to 30 minutes, in the majority of cases, the information needed for guidance of therapy with coagulation factors, platelets, and antifibrinolytics is available. Since our study included the administration of antifibrinolytics on HF diagnosis, a comparative prospective trial must be conducted to determine whether mortality is actually is higher in a control group without drug administration. 29,30 Since some of these drugs have been suggested to produce worse outcomes in cardiac surgery, 31 alternative therapies may also prove beneficial. Nonetheless, the first-line therapy for HF has to be the administration of an antifibrinolytic drug, as the alternative substitution of coagulation factors is not a causal therapy and will result in consumption and excessive costs. Finally, the severe outcome of patients with HF on arrival in the ER was striking. The higher overall observed mortality when compared with the predicted TRISS mortality illustrates an independent contribution of HF to the poor outcome of this severe trauma population. Notably, differences in the pattern of HF had prognostic value. In the group with fulminant HF (clot lysis within 30 minutes), no survival was observed even with early diagnosis and therapy. The lower median fibrinogen concentration in this group (50 mg/dl) when compared with group C (104 mg/dl) reflects the excessive consumption of fibrinogen. This finding suggests that fulminant HF (group A) may be a sign of irreversible pathophysiological processes and refractory to immediate therapy with an antifibrinolytic agent, even in combination with all other Lippincott Williams & Wilkins

7 The Journal of TRAUMA Injury, Infection, and Critical Care Volume 67, Number 1, July 2009 Hyperfibrinolysis After Major Trauma therapeutic actions. Either current treatments are inadequate for these patients, or fulminant HF is a marker of a nonsurvivable injury. In conclusion, HF occurs in major trauma and is a predictor of poor outcome. Monitoring with thrombelastometry allows early diagnosis of HF and differentiation of benign forms from severe refractory forms with 100% mortality. ACKNOWLEDGMENTS We thank Dr. W. Korte and Dr. A. Nimmo for revision and valued commentary on the manuscript and T. Bruckner for statistical analyses. REFERENCES 1. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths: a reassessment. J Trauma. 1995;38: Huber-Wagner S, Lefering R, Qvick M, et al; for the Working Group on Polytrauma of the German Trauma Society (DGU). Outcome in 757 severely injured patients with traumatic cardiorespiratory arrest. Resuscitation. 2007;75: Brohi K, Singh J, Heron M, Coats T. Acute traumatic coagulopathy. J Trauma. 2003;54: Schreiber MA. Coagulopathy in the trauma patient. Curr Opin Crit Care. 2005;11: Lampl L, Seifried E, Tisch M, Helm M, Maier B, Bock KH. [Hemostatic disorders following polytrauma the role of physiologic coagulation inhibitors during the preclinical phase]. Anasthesiol Intensivmed Notfallmed Schmerzther. 1992;27:31 36 (in German). 6. Wolberg AS, Meng ZH, Monroe DM 3rd, Hoffman M. A systematic evaluation of the effect of temperature on coagulation enzyme activity and platelet function. J Trauma. 2004;56: Meng ZH, Wolberg AS, Monroe DM 3rd, Hoffman M. The effect of temperature and ph on the activity of factor VIIa: implications for the efficacy of high-dose factor VIIa in hypothermic and acidotic patients. J Trauma. 2003;55: Brohi K, Cohen MJ, Davenport RA. Acute coagulopathy of trauma: mechanism, identification and effect. Curr Opin Crit Care. 2007;13: Kapsch DN, Metzler M, Harrington M, Mitchell FL, Silver D. Fibrinolytic response to trauma. Surgery. 1984;95: Luddington RJ. Thrombelastography/thromboelastometry. Clin Lab Haematol. 2005;27: Spiel AO, Mayr FB, Firbas C, Quehenberger P, Jilma B. Validation of rotation thrombelastography in a model of systemic activation of fibrinolysis and coagulation in humans. J Thromb Haemost. 2006;4: Hartert H. Blutgerinnungsstudien mit der Thrombelastographie, einem neuen Untersuchungsverfahren. Klin Wochenschrift. 1948;26: Lang T, Bauters A, Braun SL, et al. Multi-centre investigation on reference ranges for ROTEM thromboelastometry. Blood Coagul Fibrinolysis. 2005;16: Jámbor C, Görlinger K. Use of antifibrinolytics in massive transfusion. Anasth Intensivmed. 2007;48:S167 S Boyd CR, Tolson MA, Copes WS. Evaluating trauma care: the TRISS method. Trauma Score and the Injury Severity Score. J Trauma. 1987; 27: Offner PJ, Jurkovich GJ, Gurney J, Rivara FP. Revision of TRISS for intubated patients. J Trauma. 1992;32: Davis EG, MacKenzie EJ, Sacco WJ, et al. A new TRISS-like probability of survival model for intubated trauma patients. J Trauma. 2003;55: Kang Y. Thromboelastography in liver transplantation. Semin Thromb Hemost. 1995;21(Suppl 4): Vorweg M, Hartmann B, Knuttgen D, Jahn MC, Doehn M. Management of fulminant fibrinolysis during abdominal aortic surgery. J Cardiothorac Vasc Anesth. 2001;15: De Souza RL, Short T, Warman GR, Maclennan N, Young Y. Anaphylaxis with associated fibrinolysis, reversed with tranexamic acid and demonstrated by thrombelastography. Anaesth Intensive Care. 2004;32: Palmer JD, Francis DA, Roath OS, Francis JL, Iannotti F. Hyperfibrinolysis during intracranial surgery: effect of high dose aprotinin. J Neurol Neurosurg Psychiatry. 1995;58: Shore-Lesserson L, Manspeizer HE, DePerio M, Francis S, Vela-Cantos F, Ergin MA. Thromboelastography-guided transfusion algorithm reduces transfusions in complex cardiac surgery. Anesth Analg. 1999;88: Levrat A, Gros A, Rugeri L, et al. Evaluation of rotation thrombelastography for the diagnosis of hyperfibrinolysis in trauma patients. Br J Anaesth. 2008;100: Maegele M, Lefering R, Yucel N, et al; for the AG Polytrauma of the German Trauma Society (DGU). Early coagulopathy in multiple injury: an analysis from the German Trauma Registry on 8724 patients. Injury. 2007;38: Kaufmann CR, Dwyer KM, Crews JD, Dols SJ, Trask AL. Usefulness of thrombelastography in assessment of trauma patient coagulation. J Trauma. 1997;42: Gando S, Tedo I, Kubota M. Posttrauma coagulation and fibrinolysis. Crit Care Med. 1992;20: Enderson BL, Chen JP, Robinson R, Maull KI. Fibrinolysis in multisystem trauma patients. J Trauma. 1991;31: Sorensen JV. Levels of fibrinolytic activators and inhibitors in plasma after severe trauma. Blood Coagul Fibrinolysis. 1994;5: Coats T, Roberts I, Shakur H. Antifibrinolytic drugs for acute traumatic injury. Cochrane Database Syst Rev. 2004;CD Spahn DR, Cerny V, Coats TJ, et al; Task Force for Advanced Bleeding Care in Trauma. Management of bleeding following major trauma: a European guideline. Crit Care. 2007;11:R Fergusson DA, Hébert PC, Mazer CD, et al; BART Investigators. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery. N Engl J Med. 2008;358: Lippincott Williams & Wilkins 131