Introduction. The technique of thrombelastography, or TEG®, offers clinicians a powerful tool for the evaluation of possible coagulation disorders in a wide variety of clinical settings. This exercise will cover the principles of thrombelastography and review application of the technique to real-world clinical problems.The more commonly used tests of coagulation, particularly the prothrombin time and the activated partial thromboplastin time, measure extremely limited portions of this system and do not provide a realistic picture of coagulation status. TEG, on the other hand, provides the most nearly complete evaluation of coagulation status and allows excellent prediction of the probability of pathological thrombosis, hemorrhage, or clot lysis in difficult clinical situations.

Brief review of coagulation

    The blood coagulation and fibrinolytic system is a remarkably complex balance of up- and down-regulating proteins with constant feedback among a dizzying array of factors. The common graphical representation of a dozen or so proteins and a few ions here and there severely understates the complexity of the system.

    The purpose of the coagulation system is to control local bleeding at sites of vascular damage. The clot is composed of fibrin and platelets; virtually every component in the complex system of fibrin formation and platelet recruitment has positive and negative feedback roles in the formation and dissolution of a clot. 

    Vascular injury exposes proteins normally shielded from the intravascular environment. These proteins activate the so-called coagulation cascade, which is a series of proteins activating and inhibiting one another with the eventual cleavage of fibrinogen to form fibrin, with concurrent recruitment and activation of platelets; the latter adhere to the vascular wall and are incorporated into the clot, thus anchoring the clot to the wall and strengthening it against blood flow. The tensile strength of the clot–its ability to withstand shearing forces of blood flow–and to adhere to the vessel wall determine the effectiveness of the clot in stopping blood leakage at the site of damage.  The majority of the clot strength is a function of the platelets. 

    Clot formation will be inhibited by deficiency or inhibition of one or more coagulation  factors or of platelets. We can, artificially but conveniently, consider the process of clot formation to be composed of two arms, the enzymatic side and the platelet side. The various therapeutic regimens affect one or the other of these arms; of course, inasmuch as the coagulation factors are involved in platelet recruitment and activation, and platelets affect coagulation factors, the separation of the processes is not realistic, but we can still view them semi-independently, for purposes of evaluation of deficiencies.

    The positive feedback mechanisms produce an initially small, then rapidly progressing, degree of thrombin formation, with thrombin then cleaving fibrinogen to fibrin. The formation of fibrin is largely the result of purely enzymatic processes not involving platelets. Thus the enzymatic inhibitors (e.g., warfarin, heparin) will most significantly affect this portion of the process; their effect will extend to the platelets indirectly, by decreasing platelet recruitment rather than direct inhibition of platelet function.

    Platelet activation is, in itself, a complex process.  In the ADP activation pathway, stimulation by adenosine diphosphate (ADP) leads to a conformational change on the glycoprotein (GP) IIb/IIIa receptor, which allows binding to fibrinogen (and other platelets). Clopidogrel (Plavix®) and similar drugs inhibit the ADP-induced change, reducing platelet activity.  Such drugs typically have long half-lives, often leading to delays in surgery for patients taking them. To complicate matters, clopidogrel is a prodrug that must undergo an activation step in the liver via the cytochrome p450 pathway in hepatocytes, a step that is also required for activation of atorvastatin (Lipitor®); the presence of atorvastatin will often prevent activation of clopidogrel. Thus the platelet inhibitor may be ineffective in patients taking the cholesterol-lowering agent, a fact unrecognized by many physicians and that may not be apparent until the patient taking both drugs suffers a pathological thrombosis. In addition, some patients have a hereditary inability to activate clopidogrel, making the drug largely or entirely ineffective in these patients; genetic testing is available to identify these patients. 

    Platelets adhere to the subendothelial collagen partly via the GP-Ib receptor, mediated in part by von Willebrand factor; this receptor is inhibited by aspirin (which has other antithrombotic effects as well). Thus aspirin limits platelet adherence and, to a lesser extent, aggregation.

    Platelets may be hyper- or hypofunctional in the absence of medications, with marked platelet hyperactivity placing patients at risk for thromboembolic disease. Native platelet abnormality is a cause of concern for physicians for two reasons: (1) their patients may be afflicted with hyper- or hypocoagulability (and the attendant risk of thrombosis or hemorrhage, respectively); and (2) blood donation centers perform no tests of donor platelet function–there is no guarantee that donated platelets will be adequately functional.

    The properly functioning coagulation system also accounts for clot lysis (primarily through plasmin), and a defective or hyperfunctioning fibrinolytic system can produce pathology. Fibrin degradation products can alter vascular permeability. Antifibrinolytic therapy typically involves tranexamic acid, which has largely replaced aminocaproic acid for this purpose.

    Massive derangement in the coagulation system may result from hypersensitivity to various inhibitory drugs but can also be seen in massive trauma, tissue necrosis, or infection, particularly sepsis. Disseminated intravascular coagulopathy (often incorrectly termed disseminated intravascular coagulation), or DIC, results from an unrestrained, system-wide activation of the coagulation system and is characterized initially by hypercoagulability with rapid clot lysis (stage I) with progression to inability to clot (with subsequent generalized oozing or frank hemorrhage) due to exhaustion of coagulation factors (stage II).  It is a mistake to "fuel the fire" of stage I DIC by administering coagulation factors in the form of plasma or cryoprecipitate. Rather, the treatment of stage I DIC (which is hypercoagulability with rapid secondary fibrinolysis) is anticoagulation (heparin) and administration of an appropriate antibiotic; the anticoagulant will interrupt the runaway feedback mechanisms to prevent progression to stage II DIC. Characterized by rapid depletion of coagulation factors and platelets with severe, widespread hemorrhage, this condition is often fatal and can treated only by massive infusion of plasma, red blood cells, and cryoprecipitate.  The usefulness of recombinant factor VII is controversial.

Coagulation “pathways.” The blood coagulation system may be activated by two primary methods; the enzymatic reaction cascades that are activated eventually converge in a “final common pathway” that produces a functional clot. The tissue factor pathway involves exposure to tissue factor following vascular injury. Factor VII binds to tissue factor and initiates a complex series of reactions that produce a complex that activates Factor X; Xa then cleaves prothrombin to for thrombin, which produces fibrin from fibrinogen. The contact activation pathway is somewhat less involved, beginning with formation of a complex by several factors (including prekallikrein and Hageman factor) with collagen, eventually leading to the formation of the complex that activates Factor X. The formation of activated Factor X (i.e., Factor Xa) begins the so-called common pathway, which produces thrombin from prothrombin (among other effects). Thrombin has a number for actions, including cleaving fibrinogen to form fibrin. It is important to remember, however, that the eventual formation of fibrin is insufficient for effective hemostasis in most cases; the recruitment and activation of platelets provide the majority of the strength of the clot.

Antithrombotic therapy

    Prevention of pathologic thrombosis typically takes on of two approaches, targeting either the enzymatic elements or the platelets. In long use among enzymatic inhibitors (anticoagulants) are warfarin and heparin. Warfarin is a vitamin K antagonist; in its presence, vitamin K (technically K1) cannot participate in certain structural changes to factors II, VII, IX, and X, resulting in reduced ability to form a clot. Heparin (unfractionated and its low molecular-weight counterparts) is a glycosaminoglycan that potentiates antithrombin III; the resulting decreased activity of thrombin inhibits clot formation. Other classes of anticoagulants include the factor Xa inhibitors, such as fondiparinux, and direct thrombin inhibitors (dabigatran). 

    Antiplatelet therapy most commonly includes inhibitors of ADP (e.g., clopidogrel), GPIIb/IIIA (abciximab, eptifibatide, and tirofiban), and thromboxane A2 (dipyridamole and aspirin).

Note: It is important to recognize that a number of pharmaceutical and non-pharmaceutical preparations can have significant anticoagulant effects or show interference with medications. The SSRI and SNRI agents (typically used to treat depression or fibromyalgia) may cause platelet dysfunction. Many non-pharmaceutical agents, including vitamin E, fish oils high in omega-3 fatty acids, anise, dong quai, and ginseng) may have anticoagulant effects or may interfere with other medications. Thus it is important to get the full medication and OTC preparation history for evaluation of coagulation status.

Principle of TEG®

    The coagulation system is a function of whole blood, and whole blood is the sample required by TEG. The primary assay–known as BasicTeg®is used for general assessment of coagulation in patients who are not on antiplatelet therapy. 360 microliter of a citrated sample are placed into the sample cup with calcium chloride (to overcome the citrate), kaolin (to stimulate surface activation of platelets), and phospholipid (to activate the intrinsic coagulation pathway). These reagents will produce maximum thrombin from the sample, thereby activating platelets to the greatest degree possible (even overcoming extrinsic platelet inhibitors). The cup is rotated through 4° 45‘ of arc, and then rotated in the opposite direction, for the duration of the test. The test cup is raised so that a pin, which attached to a torsion wire and transducer, is placed into the sample. As the clot begins to form, the clot drags on the pin, which creates a current. This current is translated into a split, symmetrical reaction curve, the magnitude of which is directly related to the strength of the clot. As the test proceeds, the fibrinolytic phase begins to break down the clot, reducing the size of the curve. Figure 1 reproduces a typical, normal TEG tracing.

Fig. 1. BasicTEG. All parameters are within the normal ranges. There is no significant risk of hemorrhage or thrombosis.

Fibrin clot formation. Certain portions of the reaction tracing are associated with particular phenomena in the formation of the clot. The time in minutes from the start of the test to the split of the tracing (or "curve") is the SPLIT POINT or SP time; this represents the system “gearing up“ to begin the conversion of fibrinogen to the first detectable fibrin. The continued production of thrombin and the conversion of fibrinogen to fibrin is represented by the REACTION or R time, which is the time from test start to curve amplitude = 2 mm. R will be increased by anticoagulants and factor deficiency, although TEG® cannot distinguish among factor deficiencies.  A short R time indicates risk of fibrin clot formation and requires anticoagulant therapy.   The difference between R and SP, known as "DELTA" (R - SP = Δ), represents the “burst” of thrombin–i.e., the rapidly increasing completion of the prothrombin-thrombin conversion. Δ is a good indicator of enzymatic hyper- or hypocoagulability; a low Δ indicates hypercoagulability, a high delta indicates the opposite condition.

    The test can also be performed with addition of heparinase. As a rule (depending on the clinical situation), the so-called TEG-H® should be performed in addition to, rather than instead of, the BasicTEG. Removing the effect of heparin and heparin-like compounds will allow evaluation of the effectiveness of heparin therapy and will also reveal the presence of residual or unsuspected heparin or similar compounds, to direct therapy with protamine or other heparin-reversal agents. A rule of thumb for this assay is that a difference of two minutes or more between the delta values from the BasicTEG and the BasicTEG-H indicates the effect of heparin.

    The next values are determined by the time interval between curve amplitude of 2 mm and 20 mm  (the latter value chosen based on ultrastructural studies of platelet recruitment into the clot). This time interval is known as the clot KINETICS or K and reflects the rate of the bonding between fibrin and platelets. A related measure of fibrin-platelet interaction, and therefore of functional fibrinogen, is the angle at which the curve rises from the perpendicular beginning at the split point and ending at the point corresponding to the K value point, which is (logically enough) known as the ANGLE.  A steep angle (corresponding with a low K) indicates rapid association of  fibrin and platelets; a shallow angle (and a high K value) represents a less vigorous response. An angle of less than 45° generally indicates the need for cryoprecipitate, to replace fibrinogen and Factor VIII.

    This completes the assessment of the enzymatic side of the coagulation system. Some general rules for interpretation and therapy are as follows:

  • One unit of plasma will lower the R time approximately 2.5 minutes; for very long R times, up to four units of plasma may be administered.
  • An angle of less than 45º requires cryoprecipitate 0.06 units/kg; however plasma and even platelets can also be used to improve the angle and may be preferred, due to lower exposure to donor antigens (cryoprecipitate is a pooled component).
  • Treat the R time first! Correcting the R time may correct many other abnormalities in the TEG results.

Platelet contribution to the clot. As the clot develops and increases in tensile strength–that is, as platelet activation continues–the tracing increases to its MAXIMUM AMPLITUDE, or MA. Weak enzymatic activity or poor platelet function (including low platelet count) will result in a low MA, and platelet hyperactivity will produce a high MA. A value calculated from platelet and fibrin performance is the CLOT STRENGTH or CLOT STABILITY, which has the units of Kdynes/cm2 and is represented by the letter G. The G is perhaps the single most important value in TEG, representing the overall function or effectiveness of the clot. G values below 5 are associated with increased risk of hemorrhage, while values above 10 are associated increased risk of thrombosis. In the absence of abnormal enzyme-associated values, treatment of a high G is best accomplished with platelet inhibitors.

Clinical note: The use of platelet inhibitors is increasingly common, with drugs like clopidogrel being perhaps most effective. The effect on the G value varies, with some patients exhibiting extreme drug resistance (virtually no change in G) and others exhibiting extreme hypersensitivity (virtually complete inhibition of platelet function on the lowest dose). Clearly, the therapeutic goal lies between these extremes.

     Aspirin is also commonly prescribed, but its effect lies chiefly in the inhibition of platelet adherence rather than platelet aggregation; thus, it has relatively little effect on the G value at therapeutic doses.

Some general rules for interpretation and treatment:

  • The MA value is reflected in the G (CLOT STRENGTH) value.
  • A low MA can be improved by adding platelets; one apheresis unit of platelets should increase the MA by 9 mm. (Note also that the platelet unit also contains significant plasma and that the platelets also will release fibrinogen; these components will improve R time.)
  • A high MA or G usually requires platelet inhibition.

Clot lysis. The tracing will typically remain at the maximum amplitude for a short time, until clot lysis begins. In normal cases, lysis will continue for up to fifteen minutes or so. While the assay is running, the software algorithm estimates the percentage of lysis, with this ESTIMATED PERCENT LYSIS (EPL) value changing (at least slightly) throughout the duration of the assay. After thirty minutes, the EPL becomes the LY30, or PERCENT LYSIS AT 30 MINUTES.  The EPL/LY30 can signal excessive fibrinolysis. When the percentage is between 7.5 and 15, this may indicate stage I DIC when coupled with a very high G–these values reflect the hyperfibrinolytic and hypercoagulable nature of early DIC. Likewise, EPL above 20% suggests primary fibrinolysis and may be seen in a patient with known or diagnosed deep venous thrombosis, for example. Very high EPL values suggest hyperfibrinolysis and may indicate the need for antifibrinolytic therapy. 

Determining effect of antiplatelet agents: PlateletMapping®.  Routinely available coagulation tests do not provide evaluation of the effectiveness of antiplatelet agents. One adjunct of TEG® testing, however, can demonstrate the degree of platelet inhibition produced via the ADP and AA pathways. The PlateletMapping® assay involves four separate analyses:

  • baseline cup, which activates platelets maximally (as in the BasicTEG)
  • activator cup, in which heparin suppresses native thrombin activity, and an activator mixture cleaves fibrinogen to fibrin (in place of thrombin) and converts FXIII to FXIIIa–by effectively bypassing/inactivating the platelets, we can examine the contribution of fibrin to the clot;
  • ADP cup, which adds the activator to form the fibrin clot and ADP to activate non-inhibited platelets via the GP IIb/IIIa receptor;
  • AA cup, which adds the activator (for fibrin clot) and arachidonic acid to activate the thromboxane A2 pathway that leads to platelet aggregation. Via the inhibitory effect on COX1, aspirin also affects the stimulation of the GPIb receptor and thus inhibits platelet adherence to the blood vessel wall. Thus the AA cup assay is a reflection of the effect of aspirin and is used primarily as a surrogate measurement of platelet adherence. The AA percentage is not part of the calculation of clot strength (G).

    By comparing the relative clot strengths of the baseline, activator, and ADP assays, it is possible to determine the degree of platelet inhibition by the particular agent. Examine figure 2; the ADP curve has an amplitude that is 57.9% of the difference between the activator (fibrin only) and baseline (maximum potential amplitude). This percentage is the percent aggregation for ADP.  Multiplying this by the baseline G gives the NET G, or net clot strength (net G = G x % ADP aggregation). Thus, the platelet inhibitor has left the patient with 57.9 percent of his native clot strength, or 10.4 x 0.579 = 6.02. The goal is to maintain the G (or net G) between 5 and 9, to avoid hemorrhage and thrombosis.

Fig. 2. PlateletMapping ADP tracing. White curve is the baseline. Green curve is the activator tracing (fibrin clot only).  Red curve is the ADP-activated platelets, showing the inhibition of the platelet inhibitor that the patient was taking.

    A similar calculation is performed with the AA cup, although the result is then subtracted from 100 to give the percent inhibition (of platelet adherence) produced by aspirin. This is the result that the clinicians are interested in for patients on aspirin therapy–the percent inhibition. A typical therapeutic goal is between thirty and seventy percent inhibition. Aspirin does not generally have a significant effect on the reported clot strength (recall that the net clot strength is a function of baseline clot strength, MA and the ADP aggregation, and the last value is independent of the AA activation pathway). 

Fig. 3: The AA inhibition is approximately 57%, a therapeutic level of inhibition.

Clinical note: On occasion, infusion of platelets will have little effect on the MA and G, because the donated platelets are partly inhibited, due to a natural defect or due to certain ingested substances. Donor centers do not routinely test donor platelet function. 

Interpretation of TEG results. In most cases it is important to have access to pertinent clinical information, including whether the patient is bleeding or has thromboses, what medications are in use, and whether the patient has a history of bleeding or thrombosis. 

    Abnormal values may be single or multiple in any assay. Single values are relatively straightforward to understand. Enzymatic values (R, delta) lower than normal indicate enzymatic hypercoagulability and suggest the need for anticoagulant therapy; when these values are high, they suggest the presence of an inhibitor or factor deficiency. A low MA or G value suggests poor platelet function requiring either DDAVP for mild dysfunction or platelets for more extensive dysfunction. A high MA or G value calls for platelet inhibition. 

    Multiple abnormal values are somewhat more complex but generally fall into one of a few categories. Factor deficiency will include a low G and a high R, with or without a high Δ. Low to normal G with normal R and an EPL greater than 20% indicates primary fibrinolysis. A high G with R less than 4 and EPL between 7.5% and 20% indicates stage I DIC. 

    A bleeding patient with normal TEG values suggests surgical bleeding. DDAVP may help some cases by increasing platelet adherence, but surgical exploration is likely to be required. Finally, a low to normal G with an R above 8 may indicate hemodilution; DDAVP may be useful in this circumstance as well.

    A general rule that is worth remembering is TREAT THE R VALUE FIRST. In many cases, correction of the R value–whether by infusion of plasma, protamine, or anticoagulant–will correct other abnormal values as well. 

Table 1: Summary of abnormal values with interpretation and suggested therapeutic measures. NOTE: THese are general guidelines; ALWAYS consider the clinical situation and direct therapy accordingly.

RapidTEG®. One additional modality is used predominantly in cardiac surgery, for on-pump hemostasis evaluation. The RapidTEG uses multiple activators to “jump-start” the coagulation cascade, providing the equivalent of an ACT assay with the addition of K, angle, MA, G, and LY30. Thus there is no opportunity to assess the SP, R, or delta status; the SP and R values are replaced with the “TEG ACT” value. Also, the sample must be drawn in a plain syringe and placed on the analyzer within four minutes, so it can be a logistical problem. It is slightly faster to obtain a result (perhaps ten or so minutes faster), but there is a bit less information forthcoming. A detailed discussion of RapidTEG is beyond the scope of this offering. As a rule, RapidTEG is reserved for intra-operative use, particularly in cardiac surgery.

Practical use of TEG.  Inpatient and outpatient settings offer many instances for improvement of patient care, with concomitant reductions of expense, using TEG. A few of these are discussed briefly here.

Meaningful evaluation of the effectiveness of antithrombotic therapy. Given the vast numbers of patients taking medication to prevent blood clot formation, it is appropriate to determine whether the medications are effective. TEG will detect cases of ineffective platelet inhibitors (due to genetic resistance or interference by other agents, such as atorvastatin), as well as cases of hypersensitivity to those agents or anticoagulants. 

Reduced patient “wait time.” It is not uncommon for patients with hip fractures (for example) to have been taking any of a number of anticoagulants/antithrombotics (clopidogrel and aspirin are encountered particularly frequently). The usual practice is to hold such patients in the hospital for several days before surgery, to allow the medication(s) to wear off. TEG can reveal the patient’s likelihood of surgical bleeding; TEG performed in the emergency department can (and often does) show that the patient is ready for surgery immediately. Similarly, numerous procedures (drainages, etc., with radiologic guidance) are delayed because the patients’ INR values are high. The INR, of course, is of limited value; TEG will accurately predict the risk of bleeding with the procedure, often obviating delay of the procedure or infusion of plasma or platelets. 

Reduced blood component use. If a patient bleeds or oozes after a surgical procedure, TEG can indicate whether this is a coagulopathy or a surgical problem. A bleeding patient with a normal TEG will probably need re-exploration rather than plasma, cryoprecipitate, or platelets. Likewise, the TEG can suggest that administration of DDAVP may be sufficient to stop the bleeding. TEG can show residual or unexpected heparin/heparinoid effect (such as heparin rebound following surgery, or endogenous heparinoids following liver trauma), allowing rapid use of protamine to stop bleeding.

DIC. A markedly elevated clot strength (G) value with LY30 (or EPL) between 9 and 15 is strongly suggestive of stage I DIC. Early detection can allow treatment with heparin in time to slow the process and stop the runaway consumption of coagulation factors that would otherwise lead to factor depletion with subsequent hemorrhage.

Thus, TEG offers a wide range of benefits that will simplify and improve patient care, with the added benefit of sparing blood components (with decreased exposure to potential infection) decreasing hospital stays and costs, and reducing cases of hospitalization by preventing thrombosis in at-risk patients.

Anticoagulant and antiplatelet agents. Recent years have seen the addition of a number of specialized anticoagulants to formularies, including the factor Xa inhibitors (particularly the xaban drugs) and direct thrombin inhibitors (e.g., argatroban and dabigatran). In our brief experience with these drugs and TEG, we have seen that these drugs will typically produces marked increases in the SP and R times but may allow normal clot formation after this marked delay in the initiation of clot formation. Direct platelet inhibitors, too, are being added to formularies. Of course, one must include the drugs used for claudication (e.g., pentoxifylline), some of which act by less-than-certian mechanisms. The following lists are not complete but they may be helpful.

Another class of drug that must be considered includes those agents that alter serotonin reuptake (the SSRIs and SNRIs; sertraline is a classic example). While increasing serotonin concentration in the CNS, these drugs will deplete serotonin peripherally, leaving less serotonin for platelet activation; patients on these drugs should be monitored with PlateletMapping® before surgical procedures, to determine risk of hemorrhage. It should go without saying that such drugs should be used in combination with anticoagulants and/or antiplatelet agents only with the evaluation of coagulation status by TEG with PlateletMapping®. 

More recently, investigators have recognized that opioids may suppress platelet function, this TEG's role (Particularly in emergency medicine) may well increase.  In addition, there is some early evidence that cannabinoids may suppress the ADP-pathway platelet activation, which PlateletMapping® will reveal; in view of the increasing acceptance of marijuana use (socially and legally), clinicians will do well to be aware of the role of TEG in this group of patients.

TEG, TEG-H, BasicTEG, and PlateletMapping are registered trademarks of Haemonetics Corporation, Braintree, Massachusetts.