Anesth Analg 2007;104:59-64
© 2007 International Anesthesia Research Society
doi: 10.1213/01.ane.0000248223.05152.a1
CARDIOVASCULAR ANESTHESIA
Qualitative Thrombelastographic Detection of Tissue Factor in Human Plasma
Vance G. Nielsen, MD*,
Paul Audu, MD
,
Lana Cankovic, MD*,
Ralph T. Lyerly, III, MD*,
Brad L. Steenwyk, MD*,
Valerie Armstead, MD, and
Garry Powell, BS
From the Department of *Anesthesiology, The University of Alabama at Birmingham, Birmingham, Alabama and Department of Anesthesiology, Thomas Jefferson University, Philadelphia, Pennsylvania.
 |
Abstract
|
|---|
BACKGROUND: Tissue factor (TF) is the principal in vivo initiator of coagulation, with normal circulating TF concentrations reported to be approximately 23–158 pg/mL. However, patients with atherosclerosis or cancer have been reported to have TF concentrations ranging between 800 and 9000 pg/mL. Of interest, thrombelastographic (TEG®)-based measures of clot initiation and propagation have demonstrated hypercoagulability in such patients at risk for thromboembolic events. Thus, our goal in the present investigation was to establish a concentration-response relationship of the effect of TF on TEG® variables, and determine specificity of TF-mediated events with a monoclonal TF antibody.
METHODS: Thrombelastography was performed on normal human plasma exposed to 0, 500, 1000, or 2000 pg/mL TF. Additional experiments with plasma exposed to 0 or 750 pg/mL TF in the presence or absence of a monoclonal TF antibody (1:360 dilution, 10 min incubation) were also performed. Clot initiation time (R) and the speed of clot propagation (MRTG, maximum rate of thrombus generation) were determined.
RESULTS: The addition of TF to normal plasma resulted in a significant, concentration-dependent decrease in R and increase MRTG values. The addition of TF antibody to samples with TF significantly increased R and decreased MRTG values compared to samples with TF addition.
CONCLUSIONS: In conclusion, changes in TEG® variables in conjunction with use of a TF antibody can detect pathological concentrations of TF in human plasma in vitro. Further investigation is warranted to determine if TEG®-based monitoring could assist in the detection and prevention of TF-initiated thromboembolic events.
|
Introduction
|
|---|
Tissue factor (TF) is the principal in vivo initiator of coagulation, with normal circulating TF concentrations reported to be approximately 23–158 pg/mL (1–3). However, patients with unstable angina have been reported to have TF concentrations over 800 pg/mL (4), and patients with cancer have been noted to have maximum TF concentrations ranging from approximately 900 (5–8) to 9000 pg/mL (9). Prior to the present investigation, no clotting-based methodology has been described that is capable of specifically implicating increased plasma TF concentrations as the basis for in vitro or in vivo hypercoagulability. Of interest, thrombelastographic-based parameters have demonstrated hypercoagulability in patients with various cancers at risk for elevated circulating TF concentrations (10–12). Nonetheless, a concentration-response relationship of the effect of TF on TEG® has not been established, making a correlation of changes in TEG® parameters to circulating TF concentrations difficult to discern. Further, modification of samples to determine TF-mediated effects has not yet been described. The identification of TF-mediated hypercoagulability would have potential clinical utility, perhaps serving as a prognostic indicator of impending thrombotic events, or as an indicator of disease progression (8). Other clinical scenarios may result in a hypercoagulable state as determined by TEG®, such as antithrombin deficiency (13). Thus, the first goal of this study was to characterize the effect of clinically encountered concentrations of TF on coagulation kinetics via TEG®. The second goal was to use a commercially available TF antibody to neutralize TF-mediated clot initiation to ascertain TF-specific effects on coagulation kinetics.
|
METHODS
|
|---|
Rationale for Plasma-Based TEG® Assays
Plasma, unlike whole blood from volunteers, is devoid of individual hemostatic variation mediated by platelets. As the precise activities of individual procoagulants and anticoagulants are known, normal, pooled plasma is used as a standard for most hematological analyses performed in clinical laboratories. These products are commercially available, are noncellular, and cannot be linked to individual donors, institutional ethical approval is not required as per the guidelines of the National Institutes of Health. Lastly, unlike whole blood, patient plasma samples (usually 1–2 mL obtained per citrated whole blood sample) are stable for hours, permitting repeat assay if needed and once frozen, can be transported to other laboratories/institutions for further biochemical or TEG®-based analyses.
TEG® Analyses
Plasma sample mixtures were placed in a disposable cup in a computer-controlled TEG® hemostasis system (Model 5000, Hemoscope, Niles, IL), with addition of CaCl2 as the last step to initiate clotting. Thrombelastographic data were collected until maximum elastic modulus (MG) was reached or 60 min had elapsed. MG (dynes/cm2) was determined from maximum amplitude (MA), expressed by the following equation: MG = (5000 x MA)/(100 – MA). In addition to the determination of reaction time (R, a measure of clot initiation denoted by an amplitude of 2 mm/s), several other recently described (13) TEG® variables were determined as subsequently described. Time to maximum thrombus generation (TMG) is defined as the time interval (s) observed before maximum velocity of clot growth. Maximum rate of thrombus generation (MRTG) is the maximum velocity of clot growth observed (dynes · cm–2 · s–1). Total thrombus generation (TTG) is denoted as the total area under the velocity curve during clot growth (dynes/cm2), representing the amount of clot strength generated during clot growth. Version 4.2 TEG® software was used to analyze all experiments.
FXII-Deficient Plasma Experiments
To define the smallest concentration of TF required to affect TEG®-based parameters, the effects of contact activation exerted by TEG® cups were eliminated. Contact activation was maximally decreased by using citrated, FXII-deficient plasma (George King Bio-Medical, Overland Park, KS). Plasma was exposed to 0, 100, 500, 1000, or 2000 pg/mL lipidated recombinant human TF (a generous gift of S. Vidan of American Diagnostica, Stamford, CT). The sample composition placed in the reaction cup was 310 µL plasma, 10 µL TF serially diluted in 0.9% NaCl from stock solutions to obtain the aforementioned concentrations, 20 µL 0.9% NaCl and 20 µL of 200 mM CaCl2.
Normal Plasma Experiments
The second series of experiments were performed in citrated, normal plasma (George King Bio-Medical). Plasma was exposed to 0, 100, 500, 1000, or 2000 pg/mL TF. The sample composition placed in the reaction cup was 310 µL plasma, 10 µL TF in 0.9% NaCl, 20 µL 0.9% NaCl, and 20 µL of 200 mM CaCl2.
Inhibition of Contact Activation in Normal Plasma
In a third series of experiments, corn trypsin inhibitor (CTI, Enzyme Research Laboratories, South Bend, IN) was placed into six normal plasma samples (final concentration 130 µg/mL) in order to inhibit activated FXII as has been previously described (14,15). The sample composition placed in the reaction cup was 310 µL plasma, 10 µL CTI in 0.9% NaCl, 20 µL 0.9% NaCl, and 20 µL of 200 mM CaCl2. The purpose of these experiments was to determine if the addition of CTI could potentially improve TEG®-based detection of changes in TF concentration by diminishing contact activation.
Inhibition of TF-Mediated Effects on Coagulation
In a fourth series of experiments, normal plasma was exposed to 0 or 750 pg/mL TF in the presence or absence of a 1:360 dilution of a monoclonal antibody against human TF (Product number 4509, American Diagnostica). This antibody is epitope specific for amino acids 1–25 within the extracellular domain. The antibody dilution was chosen on the basis of pilot study observations that 1:50, 1:100, and 1:200 dilutions (n = 1–2 each) depressed contact activation in normal plasma, but a 1:360 dilution did not affect contact activation (n = 6). The plasma mixtures were as described in the first two series of experiments, and the samples were incubated in TEG® reaction cups at 37°C for 10 min before the addition of CaCl2 to facilitate TF-antibody binding.
Statistical Analyses
All experiments had n = 6 per condition, as this number of experiments typically required to obtain a ß 
0.8 with an 
< 0.05 for most thrombelastographic variables as demonstrated in a previous in vitro investigation (13). Analyses of the effects of TF concentration and TF-antibody exposure were conducted with one-way analysis of variance (ANOVA) with the Holm–Sidak post hoc test for multiple comparisons (SigmaStat 3.0, SPSS, Chicago, IL). Graphical representation of the data was generated with commercially available software (Origin 7.5, OriginLab, Northampton, MA). Values were expressed as mean ± sd. A P value of <0.05 was considered significant.
|
RESULTS
|
|---|
FXII-Deficient Plasma Experiments
Five of six experiments with the FXII-deficient plasma samples not exposed to exogenous TF displayed no sign of coagulation by 60 min, and one sample had an R value of 2810 s. Thus, 2810 s was considered the lowest limit of detection for 0 pg/mL added TF, and no other sample in the remaining data sets had R values this large. Table 1 contains the coagulation kinetic data comparing 100–2000 pg/mL additions of TF to FXII-deficient plasma. When compared with 100 pg/mL, samples containing 500–2000 pg/mL had significantly faster clot initiation denoted by smaller R and TMG values. Further, samples exposed to 2000 pg/mL TF had significantly smaller R values than samples exposed to 500 pg/mL TF. With regard to the velocity of clot propagation, plasma exposed to 2000 pg/mL TF had significantly greater TMG values when compared with all other conditions, whereas samples exposed to 1000 pg/mL had MRTG values only significantly greater than samples with 100 pg/mL TF, and not different from samples with 500 pg/mL TF. Lastly, clot strength (MG, TTG) was significantly decreased by addition of 500 and 1000 pg/mL TF compared to samples with 100 pg/mL TF. Representative clot growth velocity curves of the effects of TF addition to FXII-deficient plasma are displayed in Figure 1.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 1. Representative clot growth velocity curves of progressive tissue factor (TF) addition to FXII-deficient plasma. The y-axis denotes change in the velocity of clot propagation, with the peak of the curve corresponding to time to maximum thrombus generation. TF concentrations are as follows: black = 2000 pg/mL, diagonal lines = 1000 pg/mL, dark gray = 500 pg/mL, and cross-hatched = 100 pg/mL.
|
|
Normal Plasma Experiments
The effects of adding exogenous TF to normal plasma on TEG® parameter values are listed in Table 2. When comparing the samples with 0 pg/mL TF, samples with TF added had a concentration-dependent acceleration of clot initiation, denoted by decreased R and TMG values, with all concentrations of TF significantly different from each other. Similarly, the addition of TF to normal plasma increased the velocity of clot propagation, manifesting as an increase in MRTG. Lastly, the addition of TF resulted in a significant decrease in clot strength at all tested concentrations compared to samples with 0 pg/mL addition. Representative clot growth velocity curves of the effects of TF addition to normal plasma are displayed in Figure 2.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 2. Representative clot growth velocity curves of progressive tissue factor (TF) addition to normal plasma. The y-axis denotes change in the velocity of clot propagation, with the peak of the curve corresponding to time to maximum thrombus generation. TF concentrations are as follows: black = 2000 pg/mL, diagonal lines = 1000 pg/mL, dark gray = 500 pg/mL, and cross-hatched = 0 pg/mL.
|
|
Inhibition of Contact Activation in Normal Plasma
Exposure of plasma to 130 µg/mL CTI resulted in R values of 2215 ± 470 s (range 1800–3065 s), TMG values of 2526 ± 542 s, MRTG values of 1.0 ± 0.3 dynes · cm-2 · s–1, TTG values of 125 ± 43 dynes/cm2, and MG values of 1209 ± 377 dynes/cm2. Given the range of R values observed in FXII-deficient plasma exposed to 100 pg/mL TF (1160–2540 s), and the range of R values of normal plasma not exposed to TF (535–850 s), it seemed unlikely that the addition of CTI would remarkably improve the ability to discriminate TF concentrations in the 100–500 pg/mL range, and therefore CTI was not used in experimentation involving TF antibodies.
Inhibition of TF-Mediated Effects on Coagulation
Data evaluating the effects of TF antibodies on coagulation kinetics in the presence or absence of 750 pg/mL TF are displayed in Table 3 and Figure 3. The addition of TF antibodies to normal plasma at a 1:360 dilution did not affect coagulation kinetics when compared with normal plasma without TF-antibody addition. The addition of 750 pg/mL TF to normal plasma significantly accelerated clot initiation, enhanced the velocity of clot propagation and decreased clot strength compared to normal plasma with or without TF antibodies presence. The combination of 750 pg/mL TF and TF-antibody addition resulted in significantly delayed clot initiation and depressed the velocity of clot propagation when compared with all other conditions; however, the TF-associated decrease in clot strength persisted. Representative clot growth velocity curves of the effects of TF addition in the presence or absence of TF antibodies in normal plasma are displayed in Figure 3.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 3. Representative clot growth velocity curves of plasma exposed to tissue factor (TF) and TF-antibody additions. The y-axis denotes change in the velocity of clot propagation, with the peak of the curve corresponding to time to maximum thrombus generation. A. Normal plasma. B. Normal plasma with 1:360 TF-antibody addition. C. Normal plasma with 750 pg/mL TF addition. D. Normal plasma with both TF antibody and TF additions.
|
|
|
DISCUSSION
|
|---|
One of the primary conclusions that can be drawn from the present study is that, within the normal range of approximately 23–158 pg/mL (1–3), TF-initiated coagulation would not be detectable in normal plasma samples. This is likely secondary to the more dominant effects of contact activation initiated by contact of plasma with the plastic surfaces of the reaction cup on coagulation kinetics. Indeed, TF concentrations must be in the pathological range of >500 pg/mL to be detected as an increase in clot kinetics via TEG® in normal plasma. This high or higher concentrations of TF are associated with biochemical markers of intravascular coagulation (4–7,16) or thrombotic events (7,16). Thus, it is likely that if hypercoagulability, denoted by accelerated clot initiation and enhanced velocity of propagation, is diminished by addition of TF antibodies, future patient samples displaying these clot kinetics would likely have pathological TF concentrations. Concordantly, given the association of increased plasma TF concentration with ischemic or neoplastic states (3–9), perhaps this TEG®-based approach could be used to detect such diseases and to monitor the results of surgical intervention (e.g., tumor resection, vascular reconstruction). Thus, the present investigation presents a novel, TEG®-based assay that could potentially (via monoclonal antibody) detect pathological TF concentrations associated with clinical thrombophilia in human plasma.
Other investigations have attempted to functionally link changes in coagulation with TF activity using whole blood and plasma. Santucci et al. (17) used a Sonoclot®-based (Sienco, Wheat Ridge, CO) method to quantify the effects of TF on coagulation with a tissue factor clotting time (TiFaCT) assay. Unlike in the present study, lipopolysaccharide was added to whole blood with/without a TF antibody added. Differences in TiFaCT values were suggested to be secondary to TF expression on activated white cells and platelets (17). Blood was incubated for 10 min, 2, or 4 h before assay with "TF-mediated" changes noted after two or more hours (17). In another series of experiments, these authors added 0–80 pg/mL of recombinant human TF to whole blood and demonstrated a linear decrease in TiFaCT values (17). Further, patients with unstable angina were found to have a decrease in whole blood TiFaCT values compared to a normal population (17). Importantly, these investigators did not expose either volunteer samples with known additions of TF or patient samples with measured TF concentrations to the anti-TF antibody and determine TiFaCT values (17). Further, the range of TF tested in this investigation (0–80 pg/mL with corresponding TiFaCT values between approximately 350 and 190 s) is far less than that associated with clinical thrombosis (4–7,16). Another investigation of subclinical sepsis via endotoxin infusion in volunteers used the TiFaCT assay, demonstrating that addition of an anti-TF antibody slightly (approximately 10% or less), but significantly increased TiFaCT values over time (18). However, the plasma concentrations of TF were not determined (18). Thus, the TiFaCT assay, when combined with an anti-TF antibody, is likely very sensitive for detecting very small changes in circulating TF–TF concentrations far less than that associated with clinical thrombosis (4–7,16). This is not to say that the TiFaCT assay could perhaps be optimized to detect pathological TF concentrations. However, given that 80 pg/mL TF in whole blood resulted in a TiFaCT value of approximately 190 s, it is unlikely that this assay would be able to discriminate between 100 and 500 pg/mL TF concentrations. Thus, the TEG®-based assay presented by this investigation may be the most promising potential methodology to specifically detect changes in coagulation secondary to pathological concentrations of TF.
An unexpected finding of this investigation concerned a decrease in clot strength with addition of TF to both FXII-deficient and normal plasma (Tables 1 and 2). This seemed a paradoxical effect, given the increase in speed of clot initiation and propagation. One possible explanation may be that progressive increases in thrombin generation may result in a faster polymerization of fibrin and, perhaps, a somewhat disordered/excessive compartmentalization within the forming clot. Orderly crosslinking of fibrin polymers by FXIII is critical to protein-mediated clot strength, and if movement of activated Factor XIII within the fibrin matrix is prematurely diminished, crosslinking of fibrin polymers would be similarly attenuated, potentially accounting for the decrease in clot strength observed in our study (19).
Another unexpected finding of this investigation was that the addition of anti-TF antibodies to plasma with pathological TF concentrations decreased coagulation kinetics to an extent less than that observed in plasma devoid of antibody, or with antibody present without pathological TF (Table 3, Fig. 3). One would have expected a decrease in clot kinetics to "normal" values, not depression of clot kinetics below normal values. We speculate that the intrinsic properties of TF—TF antibody conjugates affect the complex interactions involved in clot formation (e.g., activated FXIII-fibrin interactions), as both coagulation kinetics and clot strength were reduced by TF–TF antibody conjugates. Given both of these phenomena, the primary TEG®-based variables that demonstrated the greatest specificity for increased TF concentrations were those that denote changes in clot initiation time and velocity of clot propagation rates.
There are potential limitations to this TEG®-based assay of TF activity. First, sufficient procoagulants (e.g., fibrinogen, Factor X, etc.) must be present in order to detect changes in coagulation kinetics between samples exposed/not exposed to anti-TF antibodies. Second, the presence of anticoagulants (e.g., heparin, argatroban) may prevent detection of TF by attenuation or complete suppression of coagulation. However, it could logically be argued that if a patient is already anticoagulated to the point of eliminating hypercoagulability determined by TEG®, then perhaps the detection of elevated TF activity may be unimportant. In summary, the envisioned purpose of this TEG®-based assay of TF activity is to detect hypercoagulability (if it exists) and to identify the role played by TF in the coagulopathy.
In conclusion, the present study demonstrates that TEG®-based variables of clot initiation and velocity of propagation in conjunction with use of a TF antibody can be used to detect pathological concentrations of TF in human plasma in vitro. While promising, interactions of this TEG®-based TF assay with other pathological, comorbid conditions (e.g., upregulated contact proteins, such as bradykinin; antithrombin deficiency) remains to be determined during clinical investigation. Given the lack of other methodologies that specifically link TF concentrations with functional changes in coagulation, actuarial investigation is warranted to determine if TEG®-based monitoring could assist in the prevention of TF-initiated thromboembolic events or in the detection of TF-associated disease states (e.g., cancer, vascular disease).
|
Footnotes
|
|---|
Accepted for publication September 21, 2006.
Supported by the Department of Anesthesiology, The University of Alabama at Birmingham and Thomas Jefferson University.
Author for correspondence and reprint requests to Vance G. Nielsen, MD, Department of Anesthesiology, The University of Alabama at Birmingham, 901 South 19th St., Basic Medical Research II, Room 206, Birmingham, Alabama 35249-6810. Address e-mail to vnielsen{at}uab.edu.
|
REFERENCES
|
|---|
- He M, Wen Z, He X, et al. Observation on tissue factor pathway and some other coagulation parameters during the onset of acute cerebrocardiac thrombotic diseases. Thromb Res 2002;107:223–8.[ISI][Medline]
- Koyama T, Nishida K, Ohdama S, et al. Determination of plasma tissue factor antigen and its clinical significance. Br J Haematol 1994;87:343–7.[ISI][Medline]
- Blann AD, Amiral J, McCollum CN, Lip GYH. Differences in free and total tissue factor pathway inhibitor and tissue factor in peripheral artery disease compared to healthy controls. Atherosclerosis 2000;152:29–34.[ISI][Medline]
- Falciani M, Gori AM, Fedi S, et al. Elevated tissue factor and tissue factor pathway inhibitor circulating levels in ischaemic heart disease patients. Thromb Haemost 1998;79:495–9.[ISI][Medline]
- Asakura H, Kamikubo Y, Goto A, et al. Role of tissue factor in disseminated intravascular coagulation. Thromb Res 1995;80:217–24.[ISI][Medline]
- Ueda C, Hirohata Y, Kihara Y, et al. Pancreatic cancer complicated by disseminated intravascular coagulation associated with production of tissue factor. J Gastroenterol 2001;36:848–50.[ISI][Medline]
- Kamikura Y, Wada H, Nobori T, et al. Elevated levels of leukocyte tissue factor mRNA in patients with venous thromboembolism. Thromb Res 2005;116:307–12.[ISI][Medline]
- Han LY, Landen Jr CN, Kamat AA, et al. Preoperative serum tissue factor levels are an independent prognostic factor in patients with ovarian carcinoma. J Clin Oncol 2006;24:755–61.[Abstract/Free Full Text]
- Caine GJ, Harris AL, Christodoulos K, et al. Analysis of combination anti-angiogenesis therapy on markers of coagulation, platelet activation and angiogenesis in patients with advanced cancer. Cancer Lett 2005;219:163–7.[ISI][Medline]
- Hathaway WE, Hays T. Hypercoagulability in childhood cancer. J Pediatr Surg 1975;10:893–9.[ISI][Medline]
- Haid M. Thromboelastographic distinction of malignant from benign breast masses: a preliminary report. South Med J 1977;70:774–6.[ISI][Medline]
- Francic JL, Francic DA, Gunathilagan GJ. Assessment of hypercoagulability in patients with cancer using the Sonoclot analyzer and thrombelastography. Thromb Res 1994;74:335–46.[ISI][Medline]
- Nielsen VG, Steenwyk BL, Gurley WQ, et al. Argatroban, bivalirudin and lepirudin do not decrease clot propagation and strength as effectively as heparin activated antithrombin in vitro. J Heart Lung Transplant 2006;25:653–63.[ISI][Medline]
- Holmes MB, Schneider DJ, Hayes MG, et al. Novel, bedside, tissue factor-dependent clotting assay permits improved assessment of combination antithrombotic and antiplatelet therapy. Circulation 2000;102:2051–7.
- Butenas S, Cawthern KM, van't Veer C, et al. Antiplatelet agents in tissue factor-induced blood coagulation. Blood 2001;97:2314–22.[Abstract/Free Full Text]
- Liebman HA, Feinstein DI. Thrombosis in patients with paroxysmal noctural hemoglobinuria is associated with markedly elevated plasma levels of leukocyte-derived tissue factor. Thromb Res 2003;111:235–8.[ISI][Medline]
- Santucci RA, Erlich J, Labriola J, et al. Measurement of tissue factor activity in whole blood. Thromb Haemost 2000;83:445–54.[ISI][Medline]
- Marsik C, Quehenberger P, Mackman N, et al. Validation of a novel tissue factor assay in experimental human endotoxemia. Thromb Res 2003;111:311–15.[ISI][Medline]
- Gladner JA, Nossal R. Effects of crosslinking on the rigidity and proteolytic susceptibility of human fibrin clots. Thromb Res 1983;30:273–88.[ISI][Medline]
Top
Abstract
Introduction
