Secondary Logo

Usefulness of Thrombelastography in Assessment of Trauma Patient Coagulation

Kaufmann, Christoph R. MD, MPH; Dwyer, Kevin M. MD; Crews, John D. BS; Dols, Sheila J. MT; Trask, Arthur L. MD

The Journal of Trauma: Injury, Infection, and Critical Care: April 1997 - Volume 42 - Issue 4 - p 716-722
Article
Free

Objective  Thrombelastography (TEG) is used to rapidly assess coagulation abnormalities in cardiac and transplant surgery. The purpose of this study was to investigate TEG in the initial assessment of trauma patient coagulation.

Methods  TEG was performed on 69 adult blunt trauma patients during their initial evaluation. Demographics, history of inherited coagulopathies, medications, TEG parameters, platelet count, prothrombin time/partial thromboplastin time, Revised Trauma Score (RTS), Injury Severity Score (ISS), use of blood products, and outcome were recorded.

Results  Mortality was 4.3%. Fifty-two patients demonstrated coagulation abnormalities by TEG; of these, 45 were hypercoagulable (mean ISS 13.1), and seven were hypocoagulable (mean ISS 28.6). Six of the seven hypocoagulable patients received blood transfusions within the first 24 hours. Mean ISS of the 17 patients with normal TEG parameters was 3.7. Logistic regression of ISS, Revised Trauma Score, prothrombin time/partial thromboplastin time, and TEG on use/nonuse of blood products within the first 24 hours demonstrates that only ISS (p < 0.001) and TEG (p < 0.05) are predictive of early transfusion.

Conclusions  The majority of blunt trauma patients in this series were hypercoagulable. TEG is a rapid, simple test that can broadly determine coagulation abnormalities. TEG is an early predictor of transfusion in blunt injury patients.

From the Departments of Surgery (C.R.K., K.M.D., A.L.T.) and Pathology (J.D.C., S.J.D.), Fairfax Regional Trauma Center, Falls Church, Virginia, and the Department of Surgery (C.R.K., K.M.D.), Uniformed Services University of the Health Sciences, Bethesda, Maryland.

Nonmonetary support for this study was provided by the Haemoscope Corporation, Skokie, Illinois.

Presented at the 55th Annual Meeting of the American Association for the Surgery of Trauma, September 27-30, 1995, Halifax, Nova Scotia, Canada.

Address for reprints: Christoph R. Kaufmann, MD, MPH, Trauma Service, Fairfax Regional Trauma Center, 3300 Gallows Road, Falls Church, VA 22046.

Key Words: Thrombelastography, Trauma, Coagulation, Transfusion.

Hemostasis, both physiologic and surgical, is critical to the successful resuscitation of the severely injured patient. Exsanguination is second only to central nervous system injuries as a cause of trauma death. [1,2] Appropriately, the control of hemorrhage has been identified as a priority in modern trauma patient care, second in importance only to adequate ventilation. [3] Today's inclusive trauma care systems, predicated on ready availability of specialized operative care 24 hours a day, provide surgical hemostasis. [4] More important is each patient's ability to arrest hemorrhage through physiologic hemostatic mechanisms. Thus, appropriate care of the injured patient not only includes the ability of the clinician to control hemorrhage surgically, but also to assess and correct existing or acquired abnormalities.

Many laboratory tests exist to assess hemostasis; each is specific to some portion of the hemostatic mechanism and none can stand alone. Furthermore, use of multiple tests is expensive. An ideal test for hemostasis would be rapid, inexpensive, and broad in its ability to measure coagulation. Thrombelastography (TEG), a simple test developed in 1948, is primarily used today to rapidly assess the viscoelastic properties of whole blood during cardiac and transplant operations. [5-9] TEG documents the interaction of platelets with the protein coagulation cascade from initial platelet-fibrin interaction, through platelet aggregation, clot strengthening, and fibrin cross-linking, to eventual clot lysis. Within 20 to 30 minutes, a TEG tracing can provide information on clotting factor activity, platelet function, and any clinically significant fibrinolytic process. [10] This study was performed to investigate the usefulness of TEG in assessing coagulation in injured patients.

Back to Top | Article Outline

PRINCIPLES OF THROMBELASTOGRAPHY

The thrombelastograph consists of two mechanical parts separated by the blood specimen: a plastic cup or cuvette into which 0.36 mL of blood specimen is pipetted, and a plastic pin attached to a torsion wire and suspended within the specimen. A layer of oil is added to prevent drying of the specimen. The cup oscillates slowly through an arc of 4 degrees 45 minutes at 37degreesC (Figure 1). Initially, movement of the cuvette does not affect the pin, but as clot forms, the pin becomes coupled to the motion of the cuvette. In turn, the torsion wire generates a signal that is amplified and recorded as the TEG tracing. Previously, the tracing was recorded on heat-sensitive paper moving at a rate of 2 mm per minute; today the recording is made by computer with automatic calculation of TEG indices. [10]

Figure 1

Figure 1

Several objective measurements can be determined from a TEG tracing (Figure 2): reaction time, r, is the interval between the start of the recording until the first sign of clot formation. The r demonstrates initial fibrin formation (normal range, 5-7 minutes). Clot formation time, K, is the interval measured from r time to a fixed level of clot firmness, the point that the amplitude of the tracing reaches 20 mm (normal range, 1.5-3 minutes). (Coagulation time, r + K, is the total time required for the amplitude to reach 20 mm and provides information about platelets and fibrinogen as well as intrinsic factors.) Alpha angle, alpha degrees or A, is the slope of the TEG tracing and demonstrates the rate of clot formation (normal range, 54-67 degrees). Maximum amplitude, MA, is the greatest amplitude of the TEG tracing and is a reflection of the absolute strength (maximum shear modulus) of the clot. Clot strength is composed of two components, the modest strength of the fibrin clot and the larger contribution of platelets to clot strength. [5,10] The minute times listed above for r and K can be converted to millimeters of TEG chart tracing by multiplying by 2.

Figure 2

Figure 2

Whole blood clot lysis index is the amplitude of the tracing 60 minutes after MA is achieved as a percentage of MA and denotes the loss of clot integrity or lysis. Another method of evaluating fibrinolytic activity with TEG is the fibrinolytic index in which the computerized TEG determines the actual area under the curve and reports the percentage change for 30 and 60 minutes. [10]

Although TEG analysis can be performed on native whole blood, all normal values above are for Celite-activated whole blood tests. To shorten coagulation time and speed the analysis of coagulation, 0.030 mL of the reagent, 1% Celite, is added to 0.330 mL of native whole blood. Celite acts as a contact surface (analogous to glass activation), which activates factor XII and platelets. A 1% Celite suspension provides a fast coagulation profile in the presence of heparin levels up to 2 U/mL of plasma. During full heparinization, a TEG profile can be run if the blood is first placed in an activated coagulation time tube before being placed on the TEG instrument; the r time is then equivalent to the activated coagulation time. Other modifications of TEG blood samples include addition of epsilon-amino-caproic acid to confirm the presence of a reversible fibrinolytic process or protamine to determine the presence of heparin in a blood sample. [10]

The above discussion presumes quantitative use of the TEG. It is important to note that TEG patterns can be easily interpreted without measurements to determine hyper-, hypo-, normal, and fibrinolysis (Figure 3). Quantitation of TEG results, however, will allow therapies to be judged in their ability to correct pathological states.

Figure 3

Figure 3

Back to Top | Article Outline

METHODS

After Institutional Review Board approval, the coagulation of blood samples from blunt trauma patients, who met trauma code criteria and were older than 14 years of age, was prospectively examined with the use of TEG at Fairfax Regional Trauma Center between August of 1994 and January of 1995. Initial TEGs were performed on the first blood sample drawn in the trauma suite and provide the data for this study. Clinicians providing care to these patients were blinded as to the results of the TEG so as not to allow the TEG to affect decisions regarding transfusion.

Other coagulation parameters determined in the trauma suite included routine initial platelet count, prothrombin time (PT), and partial thromboplastin time (PTT). History of medication use as well as inherited coagulopathies was also obtained. Demographics, mechanism of injury, initial Revised Trauma Score (RTS), Injury Severity Score (ISS), and survival were determined through retrospective chart review and recorded. Type and number of blood products (packed red blood cells transfused with or without fresh frozen plasma, platelets, or cryoprecipitate) used with time of administration was also tabulated for each patient. Blood use during the first 24 hours included any products transfused from the time of presentation to the emergency department until 24 hours later.

A Computerized Thrombelastograph Coagulation Analyzer (CTEG model 3000, Haemoscope Corporation, Skokie, Ill) was used for each analysis in conjunction with an IBM-compatible 486 notebook computer for data recording and quantitative output. Measurements of r, K, alpha-angle, and MA were measured and recorded for each patient using Celite-activated whole blood at 37degreesC. Statistical evaluation of the resulting data was performed using SPSS for Windows 6.0 (SPSS Corporation, Chicago, Ill) on a 486 computer. Logistic regression was used to examine the ability of ISS, RTS, PT/PTT, and TEG to predict use or nonuse of blood products in this group of patients.

A hypercoagulable state was defined as two or more of the following: short r and/or K time, increased alpha-angle, and increased MA. Hypocoagulable was defined as two or more of the following: long r and/or K time, decreased alpha-angle, and decreased MA. Normal was defined as all indices being within the normal ranges. In the occasional case of TEGs that did not fall into one of the three categories above (i.e., only a single abnormality or mixed hyper- and hypoindices in the same TEG), the predominant abnormality was used to categorize the TEG into the most appropriate category.

Back to Top | Article Outline

RESULTS

Of the initial 88 trauma code patients on whom TEG was performed, 19 were eliminated from the study because of age, incomplete data, technical difficulties, or penetrating injuries. The final study group consisted of 69 adult patients with blunt injuries: 41 male patients and 28 female patients with a mean age of 40 years (range, 16-82 years). The mean ISS was 12.3 (median, 5; range, 1-75). Mean RTS was 11.3 (median, 12; range, 3-12), and overall mortality in this group was 3 of 69 (4.3%).

Fifty-two patients demonstrated coagulation abnormalities by TEG; 45 were hypercoagulable (mean ISS, 13.1; range, 1-54), whereas seven were hypocoagulable (mean ISS, 28.6; range, 9-75). Six of these hypocoagulable patients received blood products within the first 24 hours and the remaining patient stabilized at a hematocrit of 19 without transfusion. All three deaths in this study occurred in patients who were transfused within the first 24 hours. Only two of the 45 hypercoagulable patients received blood products, and none of the 17 patients with completely normal TEG parameters (mean ISS, 3.7; range, 1-13) required transfusion within the first 24 hours (Table 1).

Table 1

Table 1

All PT/PTTs were normal in this group, except for one patient with a PT of 14.9 (normal, 11.0-14.0) and PTT of 43 (normal, 25-36). This patient was demonstrated to be hypocoagulable by all TEG indices, received blood products within the first 24 hours, and died secondary to head trauma. All platelet counts were normal except in three patients, with two having elevated platelet counts (566,000 and 502,000); each of these patients was hypercoagulable by all TEG indices, neither required blood transfusion during the hospitalization. Only one patient had an abnormally low platelet count (99,000). This patient's TEG revealed a low MA; she was transfused for a kidney hematoma beyond the first 24 hours.

No patients with inherited coagulopathies were found within this study group. Although three patients were identified that had recently taken aspirin, no potential hypocoagulability was identified by any test in this study (as would be expected). Indeed, two of these patients were hypercoagulable by TEG. There were no patients on warfarin.

To examine the ability of various measurements to predict blood use within the first 24 hours, TEG, an anatomic measure of injury severity (ISS), a measure of physiologic derangement (RTS), and standard measures of blood clotting ability (PT/PTT) were included in a logistic model. Logistic regression of ISS, RTS, PT/PTT (normal or hypocoagulable), and TEG (hyper-, normal, or hypocoagulable) on use/nonuse of blood products within the first 24 hours demonstrated that only ISS (p < 0.001) and TEG (p < 0.05) are predictive of transfusion within the first 24 hours.

Back to Top | Article Outline

DISCUSSION

Care providers consider trauma hemorrhage control to be primarily concerned with stopping blood loss, either through application of direct pressure to blood vessels or via operative interventions. Another important factor to be considered in control of hemorrhage is identification of coagulopathies, including inherited, pharmacologic, and acquired. Experience with even a few trauma patients having deficiencies of coagulation quickly focuses the clinician's attention to this critical portion of homeostasis. Restoration and maintenance of normal coagulation is clearly a critical step in resuscitation of the severely injured patient.

Early identification of coagulopathies gives the clinician more opportunity to make effective interventions. As we begin to better understand some physiologic processes, we are able to intervene at earlier points in time in physiologic derangement. It thus becomes more important to identify problems before they become clinically apparent to more effectively intervene on behalf of the trauma patient.

Platelet plug formation initially stops bleeding from injured vessels of moderate size. Injury results in exposure of subendothelial collagen to which platelets adhere. This adherence requires von Willebrand factor. Immediately, the platelets change shape, expanding their surface area, and the platelet release reaction occurs, which results in adherence of circulating platelets to those adherent to the extravascular collagen until a platelet mass is formed that stops the bleeding. [11,12] This entire process is measured by the bleeding time and TEG r + K time.

Many agents cause platelet aggregation in vitro. These agents can induce the release of arachidonic acid (from the platelet membranes), which is eventually converted to thromboxane A2, itself a potent platelet aggregator and powerful vasoconstrictor. Platelet aggregation releases the principal circulating inhibitor of plasminogen activation, thus well-placed to enhance local persistence of fibrin. [11]

Fibrin is essential to stabilize the initial fragile platelet plug to prevent secondary bleeding. Fibrin formation is a result of the conversion of its soluble predecessor, fibrinogen, via the action of thrombin. Thrombin, a protease, is generated by the coagulation cascade after tissue injury. Thrombin can be generated by either the intrinsic pathway (screened by the PTT) or the extrinsic pathway (screened by the PT). This classic division of the coagulation cascades into intrinsic and extrinsic is not relevant to humans because no coagulopathy or disease state is associated with deficiencies in several of the proteins of the intrinsic system. [12] Fibrin thus serves to reinforce the platelet plug and also forms in the wound, allowing both cessation of bleeding and wound healing. The TEG MA measures the strength of the clot that is able to be generated by platelets with the help of fibrin.

Fibrinolytic processes that begin as soon as fibrin formation occurs, to modify the amount of fibrin deposited as well as how long it will persist, depend on a series of proenzyme to enzyme conversions to produce the protease plasmin from circulating plasminogen. Plasmin can hydrolyze fibrinogen, fibrin, factor V, and factor VIII. [12] The potential for active fibrinolysis depends on the amount of circulating plasminogen activators, whose levels rise in response to many physiologic and some pathologic stimuli. Nevertheless, high levels of plasminogen activator do not mean fibrinolysis is actually occurring (just as high levels of clotting factors do not imply fibrin formation). [11] TEG identifies loss of clot integrity or lysis through the whole blood clot lysis index or the fibrinolytic index.

All four aspects of the hemostatic system mechanism come into play when an individual is injured. Vasoconstriction and platelet plug formation result in primary hemostasis, with fibrin deposition acting to stabilize the plug and seal the wound. The fibrinolytic system then acts to remove the fibrin after adequate healing has occurred. It is important to understand that all of these mechanisms act only locally and in an integrated manner. TEG measures coagulation from primary hemostasis through fibrinolytic activity and allows the clinician the opportunity to assess the balance between clot formation and clot lysis.

Trauma can cause changes in coagulation, both increases and decreases, through several mechanisms. [13-15] Damaged tissues and platelets release thromboplastic substances including thromboxane, epinephrine, serotonin, and adenosine, which contribute to local vasoconstriction. The ischemia and acidosis associated with trauma also enhance clotting. Injury is also associated with reduced levels of plasma prekallikrein, Hageman factor, and antithrombin III. Additional wellknown plasma abnormalities after trauma include elevation in fibrin split products and reductions in platelet counts. The most profound abnormalities in hemostasis after trauma are associated with head injury, particularly suppression of fibrinolysis and hypercoagulability. [15]

That the majority of patients in this study had detectable hypercoagulability is not surprising. One can hypothesize that patients with normal TEGs had either the least systemic response to their injuries, the least injury, or both-as supported by this group having the lowest mean ISS (3.7). The hypercoagulable group predictably had a higher mean ISS (13.1), demonstrating a response in coagulation consistent with increased magnitude of injury. Finally, the hypocoagulable group had the most severe injuries (mean ISS, 28.6) with the TEG providing evidence of hemostatic balance being overwhelmed by severity of injury resulting in a hypocoagulable state.

TEG is not able to identify aspirin effect on platelet function even though aspirin effect is known to cause bleeding problems during operation. As TEG does not measure platelet adhesion/release actions, and platelet aggregation by thrombin is relatively unaffected by aspirin, the lack of aspirin effect on TEG is not surprising. [5] Warfarin will demonstrate an effect on TEG with prolonged r and K, and decrease of both MA and alpha-angle. Heparin has the same effect. [10]

As this study was performed as an initial step in exploring the utility of thrombelastography in trauma settings, many limitations are apparent. Sixty-nine adult patients with blunt injuries including only three deaths may not adequately represent the overall trauma population. The study population is not a consecutive group of trauma patients, but rather a sample of convenience consisting primarily of daytime trauma patients. If resources had allowed, it also would have been preferable to assess coagulation tests other than platelet counts as part of a complete blood count and PT/PTT. It may be that other coagulation tests such as bleeding time will also be proven useful in the trauma bay. Template bleeding time is known, however, to be subject to inaccuracies unless performed by experienced personnel and is difficult under certain circumstances such as during operation. [16] TEG, based on our experience, is easy to perform with a sample of whole blood and can be accurately performed with less than an hour of training. Finally, the effect of hypothermia was not assessed in this study as all TEG specimens were run at 37degreesC. A variable temperature TEG unit is available, which allows a specimen to be run at patient temperature to assess temperature effect or at a corrected temperature of 37degreesC.

An approximate routine charge for TEG is $26; a PT/PTT costs $25, and a complete blood count without differential costs $25 (a hemoglobin and hematocrit test is $21) at our institution. The test to determine fibrinogen level $39. As TEG provides information regarding platelet function, clotting factors, and adequacy of fibrinogen (as well as assessing for early fibrinolysis), TEG used alone without PT/PTT, platelet count, and occasional fibrinogen levels could save between $3 and $42 dollars per patient, while providing information on coagulation that is both rapid and comprehensive.

As time is critical in trauma patient resuscitation, rapid results from laboratory studies are necessary. Because TEG can be run in as little as 20 minutes (excluding the fibrinolytic indices), sequential tests can be performed to rapidly assess effectiveness of therapeutic interventions. This study used Celite-activated TEG to make the results available more quickly, as TEG utility in time-sensitive trauma patients was being evaluated.

Although ISS was more predictive of transfusion within the first 24 hours, this anatomic measure is not able to be calculated until the time of patient discharge or autopsy at which time all injuries have been identified; therefore, in this study, only TEG was useful as an early predictor of blood transfusion within the first 24 hours. Although TEG coagulation analysis did not alter the clinical course of the patients in this study (by design), an abnormal TEG result not only provides evidence regarding magnitude of injury/derangement of homeostasis, but can be used to make informed decisions regarding specific treatment of hypocoagulable states. Rapid and repeated use of this relatively inexpensive test allows judicious use of blood products in trauma patients, avoiding the "shotgun" approach to coagulopathy often used today.

In summary, it has been less than 50 years since today's model of coagulation was completed. Because of this work over centuries, we are able to perform ever more complex operative procedures, comfortable in our ability to prevent hemorrhage and our ability to achieve hemostasis when bleeding occurs. These skills have also permitted us to successfully resuscitate injured patients with ever increasing degrees of injury because of our ability to control bleeding rapidly and preserve homeostasis. This first study to specifically assess the usefulness of TEG in trauma patients confirms that TEG is a rapid, simple test that can broadly determine coagulation abnormalities and is an early predictor of transfusion in patients with blunt injury. The majority of patients in this study were demonstrated to be hypercoagulable by TEG, although coagulation function varied with severity of injury. Further research regarding use of thrombelastography in trauma patient care is indicated.

Back to Top | Article Outline

PAPER DISCUSSION

Dr. Robert C. Lim (San Francisco, California): Thromboelastogram has been described very nicely by Dr. Kaufmann in his article and in his presentation this afternoon. It is not a new test. However, it is a test that has been used primarily in research in coagulation. It is a quick method to measure global changes in coagulation and to sort out its different components.

The profile in the graph elucidate changes from the initial phase of fiber and strand formation to the point of clot solidification and clot retraction and lysis.

Presently, the thromboelastogram is used in monitoring patients during cardiopulmonary bypass and in liver transplantation. The different components, as Dr. Kaufmann has elucidated, as shown on this elastogram, corresponds to the different steps in hemostasis. The r time tells us the clotting time, and is reflected by the distance between the taking of blood to the onset of clot formation. The angle of the clot formation reflects how quickly the clot solidifies.

The profile of this angle denotes whether there is normal coagulation, hypercoagulation, or prolonged coagulation. It is reflected by fibrinogen concentration.

Once the plug is formed, the size of the MA distance is a reflection of the clotting proteins and the platelets in forming this clot. If there is a decrease in concentration or platelet number, then the MA would be smaller. The distance between MA to that point at 60 minutes reflect the quality of the clot. A smaller, A60, is noted in early clot degeneration and lysis. And on this slide, the T part here denotes thrombosis, and at this point fibrinolysis begins.

Getting back to Dr. Kaufmann's study, I would agree with the authors that further research regarding the use of the thromboelastogram in trauma patients is needed. His paper demonstrates only the initial evaluation or assessment of the trauma patient's coagulation status. They have demonstrated very well and nicely how those who were hypercoagulable with a mean ISS score of 13.1 versus those who are hypocoagulable with a mean ISS score of 28.6; however, they did not compare this with standard coagulation tests.

However, other authors did show this, and it correlates rather well as compared to the data that Dr. Kaufmann just presented to you.

Other investigators have shown that the thromboelastogram correlates with standard tests, such as PTT, PT, platelet count, etcetera. As the authors point out, the results from PT and PTT take some time to obtain, whereas this thromboelastogram is obtained right away. What this paper does not address is its use in management on an ongoing basis.

Obviously, those trauma patients who are hypocoagulable would need both correction of their coagulopathy and control of blood loss from their injuries. It is in this setting where the thromboelastogram would be helpful to the trauma surgeon.

As the authors demonstrated, the thromboelastogram gives us a very nice global picture of the hemostatic status of the injured patient on admission. It also serves as a baseline. During resuscitation and management, if coagulopathy develops such as in massive blood transfusion, subsequent tracings will guide us in its therapy.

The other area which would be very useful and not discussed at this time is the early identification of fibrinolysis. The euglobulin clot lysis time is not a routine test and is difficult to obtain in most hospitals. In the thromboelastogram tracing, the tail and the height would demonstrate the presence of early clot lysis and its severity. This would enable early identification and initiate immediate treatment with EACA or other fibrinolytic drugs.

In summary, I think the thromboelastogram is a very useful test, but I believe its usefulness is in ongoing management of the patient as Dr. Kaufmann mentioned in the operating room or in the ICU setting. I agree with the authors that further studies are needed.

In closing, I would like to ask two questions. The authors studied 69 patients identifying 45 who were hypercoagulable and seven hypocoagulable. Two out of the 45 hypercoagulable patients and six out of the seven hypocoagulable patients required blood transfusions. It would be interesting to know their injuries and how much blood was given in each of these patients. Would the authors share this with us?

My other question is, having identified patients who were hypercoagulable, should they be treated, such as with prophylactic low-molecular weight heparin?

I wish to thank the Association for the honor in discussing this paper and thank the authors for providing me a copy of their manuscript in a very timely fashion. Thank you.

Dr. Erwin F. Hirsch (Boston, Massachusetts): I share many of Dr. Lim's comments on this paper. About 25 years ago, Dr. Attar at MIEMS started using a thromboelastogram to evaluate trauma patients and some of the concerns and some of its shortcomings that were discussed this morning were present at that time.

I have a comment and a question to the authors. In their abstract, they indicate that a normal hypercoagulable thromboelastogram could predict the need for blood transfusions. In a time in which most people read abstracts and conclusions and don't read the entire paper, this can be an invitation to disaster in the Emergency Department where underresuscitation is the most common error in the care of the injured.

I wonder if the authors would consider changing the number 2 conclusion and perhaps make it somehow more a general to avoid a machine like this appearing in the Emergency Department to determine the use of red blood cell transfusion. Thank you for the floor.

Dr. Dale W. Oller (Raleigh, North Carolina): I, too, enjoyed your discussion. This is not a new technique.

I wonder if it would be possible to hypothesize some areas of current management where you would anticipate that use of this older technique might be brought to the fore?

For instance, if my memory serves me correctly, head injury causes a patient to be hypercoagulable. In the multiple-trauma patient, this could be problematic because of the loss of blood within cavities. Also, how about the management of patients that have nonoperative spleens and liver?

Dr. John T. Owings (Sacramento, California): I also, as the other discussants, enjoyed the paper, and I had a comment and then a question relative to that. TEG, as with ACT, is critically dependent upon fibrinogen levels; that is, if there's no fibrinogen, there can be no fibrin stranding to decelerate or decrease the amount of movement of the cup.

In that case, a hypercoagulable state leads initially to increased fibrin formation and subsequent depletion of fibrinogen followed by inadequate fibrinogen to form fibrin strands. My question is, since this test is critically dependent upon fibrinogen levels, were fibrinogen levels measured?

And then also, with regard to that, isn't it possible that the hypercoagulable state that you see may just be the prelude to the hypocoagulable state after the fibrinogen's been depleted? Thank you.

Dr. Christoph R. Kaufmann (closing): I'd like to thank Dr. Lim for his kind review and kind comments regarding the paper.

As regards management on an ongoing basis for hypocoagulable patients, there are data from the liver transplant community that there is substantial ability to decrease component usage with the onset of TEG utilization in the operating room. In cardiac surgery, they've shown a several hundred percent decrease in takebacks to the OR at the University of Washington in some data that was presented in 1991 when using the TEG as opposed to previously not having used the TEG, although that was based on historic data.

Regarding the amount of blood transfusion, approximately 3 units of pack cells were administered to the patients that did receive blood in this study on average, and only about a quarter to a third of these received anything other than packed cells, as in platelets or FFP or cryo.

Treatment of hypercoagulable patients, Dr. Lim mentions a good point. If indeed we have these hypercoagulable trauma patients, what should we do with these? Again, there was a study presented by the Canadian Trauma Association on Wednesday that looked at low-molecular weight heparin and its increased efficacy over our typical nonmonitored heparin therapy that we're using today in most places in the United States.

If indeed we do believe that it's more efficacious, this may be a place where we could look at using the TEG as a discriminator as to how aggressive to be in our anticoagulation of trauma patients, given that they don't have head injury or some other contraindication to the use of low-molecular weight heparin or other modalities.

Dr. Hirsch, regarding cavitary hemorrhage and the need for or the nonneed for TEG in the emergency department, I agree with your concern regarding physicians basing their therapy on what a TEG analysis shows as opposed to looking more broadly at the patient. Again, since we have primarily a bluntly injured population and not a lot of penetrating patients, we elected to do our first study in the emergency department as opposed to in the operating theater, although that may be the best application of this technology.

Dr. Oller, emphasizing places of current management, particularly in nonoperative spleen and liver, yes, I think this would be a good place to do that. If we optimize the coagulation of our splenic injury patients, we may find that we are able, as time goes on, to manage even a larger proportion of these patients nonoperatively than we are managing today. I think time will tell.

There are TEG analyses. If you have two TEG machines and therefore four channels, there are two channels for each machine that are protocols using Amicar and heparinase and trying to find out why a patient has a coagulopathy all in one fell swoop to run four different samples with different correction of platelets and fibrinolysis in these patients, and that can be done.

Dr. Owings, were fibrinogen levels measured, no, they weren't at this point, and in conjunction with Dr. Lim's concern about the hypercoagulability of these patients, when we saw the results of this first study, we have now just completed 2 weeks ago a consecutive series of 85 patients that either required operative care or ICU care after the emergency department evaluation.

We are looking at fibrinogen levels this time because of our concern, similar to yours, and hope to be able to present these data in the future.

I'd like to thank the Association for the privilege of new membership.

Back to Top | Article Outline

REFERENCES

1. Baker CC, Oppenheimer L, Stephens B, et al. Epidemiology of trauma deaths. Am J Surg. 1980;140:144-150.
2. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths: a reassessment. J Trauma. 1995;38:185-193.
3. Committee on Trauma. Advanced Trauma Life Support Program for Physicians. Chicago: American College of Surgeons; 1993.
4. Committee on Trauma. Resources for Optimal Care of the Injured Patient. Chicago: American College of Surgeons; 1993.
5. Mallett SV, Cox DJA: Thrombelastography. Br J Anaesth. 1992;69:307.
6. McNicol PL, Liu G, Harley ID, et al. Patterns of coagulopathy during liver transplantation: experience with the first 75 cases using thrombelastography. Anaesth Intensive Care 1994;22:659.
7. Kang YG, Martin DJ, Marquez J, et al. Intraoperative changes in blood coagulation and thrombelastographic monitoring in liver transplantation. Anesth Analg. 1985;64:888.
8. Martin P, Horkay F, Rajah SM, et al. Monitoring of coagulation status using thrombelastography during pediatric open heart surgery. Int J Clin Monit Comput. 1991;8:183.
9. Spiess BD, Gillies BSA, Chandler W, et al. Changes in transfusion therapy and reexploration rate after institution of a blood management program in cardiac surgical patients. J Cardiothor Vasc Anesth 1995;9:168.
10. Thrombelastograph Coagulation Analyzer. Skokie, Ill: Haemoscope Corporation; 1991.
11. Bennett B, Towler HMA. Haemostatic response to trauma. Br Med Bull. 1985:41;274-280.
12. Murphy WG, Davies MJ, Eduardo A. The haemostatic response to surgery and trauma. Br J Anaesth. 1993:70;205-213.
13. Bergentz SE. On bleeding and clotting problems in posttraumatic states. Crit Care Med. 1976:4;41-45.
14. Dyke C, Sobel M. The management of coagulation problems in the surgical patient. Adv Surg. 1991:24;229-257.
15. Rutledge R, Sheldon GF. Bleeding and coagulation problems. In: Feliciano DV, Moore EE, Mattox KL, eds. Trauma. 3rd ed. Stamford, Appleton & Lange; 1996:1059-1080.
16. Rodgers RPC, Levin J. A critical reappraisal of the bleeding time. Semin Thromb Hemostasis. 1990;16:1.
© Williams & Wilkins 1997. All Rights Reserved.