Hemostasis in vivo involves a complex interplay of vascular wall, blood cells, and proteins. This intricate system for regulation of the formation and dissolution of fibrin clots within the circulatory system is often affected by disease. In particular, conditions such as colic, colitis, endotoxemia, and sepsis are often associated with coagulation abnormalities in horses.1–4 Evaluation of hemostatic capacity may be helpful for successful patient assessment and management of many diseases.
Coagulation has traditionally been evaluated in vitro with plasma-based assays (eg, prothrombin time and aPTT) for analysis of the quantity and functionality of the plasma proteins that contribute to fibrin formation. Recently, thrombelastography and thromboelastometry have been used to evaluate the viscoelastic properties of a whole blood clot in real time. The kinetic nature of the assay systems also allows for observation of clot formation and lysis in real time, and results are consequently reflective of factors contributing to thrombin generation that occurs over time after initial fibrin polymerization.5
There have been numerous descriptions of the clinical and experimental application of thromboelastometry and thromboelastography in human and veterinary medicine, and recent studies have involved use of these techniques in the evaluation of blood from horses used for research purposes6–9 and from client-owned equine patients with various disease conditions.10–14,a Methodology used for performance of thromboelastography or thromboelastometry has varied widely in regard to sample handling and coagulation initiator use. Some reports15 describe immediate use of samples, but holding of samples for 30 minutes is common,6,7,12,13,16 and holding for 2 to 20 hours prior to initiation of thromboelastography or thromboelastometry has also been described.11,17 Initiation of coagulation in thromboelastography or thromboelastometry of equine whole blood samples has included use of tissue factor6,11,12,15,16,18 or contact activators11,18 and simple recalcification of citrated blood.9,10,13,14,16 Recalcification in the absence of a trigger for coagulation allows for thrombin formation as a downstream effect of activation of the contact system that occurs during sample collection and processing. Experiments with human and canine blood have indicated that continued contact activation during sample holding adversely impacts the results obtained when thromboelastography or thromboelastometry is performed in the absence of a strong coagulation trigger.19 Furthermore, multiple studies16,18 have indicated that lack of an initiating trigger results in imprecision issues and wider variability in reference ranges for thromboelastometry or thromboelastography parameters of equine blood. The objective of the study reported here was to evaluate the degree of contact activation that occurs ex vivo in citrated equine whole blood and to determine the impact, if any, of such contact activation on results obtained with recalcification-initiated thromboelastometry. We hypothesized that results of thromboelastometry would be impacted by ex vivo activation of the contact system.
Materials and Methods
ANIMALS
Eleven apparently healthy research horses (5 mares and 6 geldings of mixed breeds) were included in the study. Horses were determined to be healthy on the basis of lack of clinical signs of illness, normal physical examination results, and clinically acceptable results of a CBCb and serum biochemical analysisc performed in the clinical laboratory at the veterinary teaching hospital where the study was performed. The study protocol was approved by the Institutional Animal Care and Use Committee of the University of Illinois College of Veterinary Medicine.
BLOOD COLLECTION AND PROCESSING
Blood was collected by atraumatic venipuncture of a jugular vein with 21-gauge, 1.25-inch blood collection needles.d Samples were collected directly into evacuated tubes to expose collected blood to the citrate anticoagulant as rapidly as possible. A nonactivated sample (5 mL) was collected first and discarded, and then the citrated sample (4.5 mL) was collected. Citrated samples were collected at a 9:1 ratio in a 3.2% citrate solution in prewarmed (37°C) siliconized glass tubes,e and then mixed thoroughly by gentle repeated inversion. The citrated samples were then held at 37°C for ≤ 30 minutes.
By use of the supplied electronic pipette, whole blood aliquots (300 μL) were removed from the sample collection tube and added to 20 μL of the manufacturer-supplied concentrated calcium chloride reagent in the thromboelastometry cup at 1, 10, 20, and 30 minutes after collection. Additional aliquots (0.5 mL) for plasma production were removed immediately after collection (0 minutes) and at 8, 18, and 28 minutes after collection, then centrifuged for 1.5 minutes at 5,000 × g. Platelet-poor plasma was transferred at 2, 10, 20, and 30 minutes after collection to tubes containing corn trypsin inhibitorf (a factor XIIa inhibitor) to achieve 50 μg/mL plus SBTIg (a kallikrein inhibitor) at 50 μg/mL, or to tubes containing SBTI alone. Plasma samples were then immediately flash frozen in a 100% ethanol–dry ice bath and stored at −80°C until batch analysis. Owing to the time requirement for centrifugation of blood samples to obtain plasma, the initial sample aliquots for thromboelastometry had coagulation initiated 1 minute after venipuncture, whereas plasma was added to the inhibitor and frozen at 2 minutes after venipuncture.
THROMBOELASTOMETRY
Thromboelastometry was performed with an in vitro viscoelasticity monitoring systemh in accordance with the manufacturer's instructions. Citrated whole blood (300 μL) was added at each designated time to the thromboelastometry cup as described. Variables measured with the manufacturer's modeling software included coagulation time (the lag period from initiation of the reaction until an amplitude of 2 mm was recorded), clot formation time (the duration necessary for amplitude to increase from 2 to 20 mm), α angle (the angle of change of amplitude during the clot formation time), and maximum clot firmness (the greatest measured amplitude). These thromboelastometry parameters are comparable to R (reaction time), K (clot formation time), α angle, and maximum amplitude, respectively, for thromboelastography. Clotting reactions were followed until maximum clot firmness was achieved.
MEASUREMENT OF PLASMA CONTACT ENZYME ACTIVITY
Plasma samples containing SBTI were thawed at 37°C for 5 minutes, then diluted 10-fold into 125mM imidazole (pH, 7.0). Diluted samples (60 μL) were mixed with 60 μL of human factor XII–deficient plasma,i and clotting was initiated with 60 μL of 25mM CaCl2 containing 75μM sonicated phospholipidj (10% phosphatidylserine, 40% phosphatidylethanolamine, and 40% phosphatidylcholine) vesicles. Plasma clotting times were quantified in 96-well ultralow-binding polystyrene microplates.k Clotting was evaluated by monitoring the change in turbidity (light absorbance at 405 nm) for 1 hour at room temperature (approx 25°C) with a microplate reader.l Clotting times were calculated through use of analytical softwarem by fitting a line to the steepest segment of the absorbance curve and determining its intersection with the initial baseline light absorbance at 405 nm (representing the lag phase prior to clot formation). Factor XII procoagulant activity was calculated in reference to a standard curve for purified human factor XIIa.n Factor XI procoagulant activity was determined in a similar manner for equine plasma samples that contained SBTI and corn trypsin inhibitor, but with human factor XI–deficient plasmao and a standard curve for purified human factor XIa used as the basis for calculations.p
STATISTICAL ANALYSIS
The effect of holding time on thromboelastometry variables and plasma enzyme activity was evaluated by 1-way repeated-measures ANOVA on ranks, with the Tukey test used to determine P values when a significant difference was found. Correlation between factor XII or factor XI procoagulant activity and coagulation time was determined with Spearman rank-order correlation analysis. Analyses were performed with statistical software.q Values of P ≤ 0.05 were considered significant.
Results
THROMBOELASTOMETRY PARAMETERS AND FIBRINOGEN
All 4 measured thromboelastometry parameters were affected by sample holding prior to initiation of coagulation with recalcification of the citrated sample, although significant differences were not found for maximum clot firmness at all time points and effects on this variable were less substantial overall (Figure 1). Samples became progressively more coagulable over the 30-minute holding period, as indicated by significant decreases in coagulation time and clot formation time and increased α angle at each time point, compared with earlier time points, and increased maximum clot firmness at 10 and 20 minutes, compared with that at 1 minute. A representative set of thromboelastometry curves from a single sample is provided (Figure 2). Fibrinogen concentrations in samples were within the reference range and did not change over time.
Recalcification-initiated thromboelastometry parameters measured in whole blood samples collected from 11 healthy adult mixed-breed horses. Values were obtained immediately after collection (0 minutes) and after storage at 37°C for 10, 20, or 30 minutes. A—Coagulation time (CT). B—Clot formation time (CFT). C—α Angle. D—Maximum clot firmness (MCF). Data points represent the mean and error bars represent SD (n = 11 samples/time point). *†Values are significantly (*P < 0.001; †P < 0.01; ‡P < 0.05) different between time points.
Citation: American Journal of Veterinary Research 76, 2; 10.2460/ajvr.76.2.122
Representative set of thromboelastometry curves from a single blood sample of a healthy horse. Whole blood was collected into prewarmed evacuated tubes containing a citrate anticoagulant and held at 37°C for 1, 10, 20, or 30 minutes, after which coagulation was initiated with recalcification.
Citation: American Journal of Veterinary Research 76, 2; 10.2460/ajvr.76.2.122
PLASMA CONTACT ENZYME ACTIVITY
Factor XII procoagulant activity (Figure 3) and factor XI procoagulant activity (Figure 4) increased significantly in equine plasma during the whole blood sample holding period. Factor XII procoagulant activity was significantly increased at 20 minutes, compared with 2 minutes, and at 30 minutes, compared with 2 and 10 minutes. Factor XII procoagulant activity in the plasma sample was linearly related to whole blood coagulation time on a log-log scale and had a significant (P < 0.001) negative correlation (r = −0.662) with coagulation time. Factor XI procoagulant activity was significantly increased at 30 minutes, compared with 2 or 10 minutes, and similarly had a significant (P < 0.001) negative correlation (r = −0.642) with coagulation time.
Effect of sample holding time on plasma factor XII procoagulant activity (FXII:C) and relationship between plasma factor XII procoagulant activity and whole blood coagulation time. A—At 2, 10, 20, and 30 minutes after whole blood sample collection, plasma was mixed with SBTI to prevent additional kallikrein-mediated generation of factor XIIa, flash frozen, and stored at −80°C for later batch analysis. Factor XII procoagulant activity increased significantly during the sample holding period. B—Whole blood coagulation time measured by thromboelastometry was significantly (P < 0.001) negatively (r = −0.662) correlated with plasma factor XII procoagulant activity. Symbols represent aliquots obtained at 2 (black triangles), 10 (gray circles), 20 (gray squares), and 30 (white triangles) minutes after blood sample collection. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 76, 2; 10.2460/ajvr.76.2.122
Effect of sample holding time on plasma factor XI procoagulant activity (FXI:C) and relationship between plasma factor XI procoagulant activity and whole blood coagulation time (CT). A—At 2, 10, 20, and 30 minutes after whole blood sample collection, plasma was mixed with SBTI and corn trypsin inhibitor to prevent additional factor XIIa–mediated generation of factor XIa, flash frozen, and stored at −80°C for later batch analysis. Factor XI procoagulant activity increased significantly during the sample holding period. B—Whole blood coagulation time measured by thromboelastometry was significantly (P < 0.001) negatively (r = −0.642) correlated with factor XI coagulant activity in the plasma sample. See Figures 1 and 3 for remainder of key.
Citation: American Journal of Veterinary Research 76, 2; 10.2460/ajvr.76.2.122
Discussion
Previous reports of use of whole blood thromboelastography and thromboelastometry in veterinary patients have included a wide variety of methodological approaches, including different blood collection methods (evacuated glass tubes and collection needles, syringes, or a combination of syringes and evacuated tubes), types of sample (citrated vs native), coagulation triggers (tissue factor, contact activators, or recalcification alone), and different handling methods after collection. Early reports19,20 of the use of thromboelastography or thromboelastometry for assessment of canine and equine blood indicated that whole blood samples became progressively hypercoagulable during sample holding. This change was initially attributed to ex vivo platelet activation secondary to venipuncture21 or to continued thrombin generation during storage.20–22 These early reports resulted in widespread adoption of a 30-minute holding period for animal blood prior to initiation of thromboelastography or thromboelastometry. However, results of our previous study20 to evaluate canine and human whole blood samples with thromboelastometry found that the shift toward hypercoagulability during sample holding is primarily attributable to activation of the contact system, which is more efficient in canine blood than in human blood. On the basis of data collected in the present study, it appears that equine blood undergoes an even more profound ex vivo activation of the contact pathway during sample holding. Coagulation kinetics for whole blood thromboelastometry were markedly affected by sample holding, and evaluation of the plasma obtained from whole blood samples of healthy horses indicated that both factor XII and factor XI were activated ex vivo to a marked degree. Furthermore, the significant negative correlations found between coagulation time and factor XII and factor XI procoagulant activities indicated that the observed hypercoagulability was a function of contact pathway activation.
Although activation of the contact pathway detected in the present study occurred ex vivo as a result of exposure of the blood to artificial surfaces, it is interesting to speculate as to whether horses may have a highly responsive contact pathway in vivo. Although contact activation plays no part in hemostatic responses, recent evidence suggests that the contact pathway may contribute to pathological conditions.23 Further exploration of contact activation as it occurs in horses in vivo may be indicated.
Assembly of the contact factors factor XII, prekallikrein, and high molecular–weight kininogen readily occurs on artificial surfaces.23 The resulting production of factor XIIa and its subsequent cleavage of factor XI to factor XIa is not prevented by citrate anticoagulation because these enzymatic reactions are not calcium dependent. The amount of factor XIIa available for generation of downstream coagulation enzymes once a sample is recalcified is a function of the concentration of these factors in a given blood sample, nature of the surfaces to which the blood is exposed, temperature of the sample, duration of exposure to artificial surfaces, and rate of inhibition of enzymes generated in the sample.24 Removing blood from the vasculature is always associated with some contact activation because blood is exposed to the metal, glass, and plastic of blood collection supplies. Siliconization of citrate-containing glass tubes and use of minimally charged plastics reduce the degree of, but do not completely eliminate, contact activation.24 Maximal activation of the contact system in equine blood stored in siliconized tubes was subjectively slow over the 30-minute evaluation period, with maximal procoagulant activity detected at the 30-minute time point, which is a commonly reported sample holding period.6,7,12,13,16 In comparison, strongly anionic surfaces such as kaolin or ellagic acid can be expected to maximally activate the contact pathway within 3 to 5 minutes, a time frame generally used for aPTT assays.25 Considering that factor XIIa is slowly inhibited by factors present in plasma (primarily C1 inhibitor, less so α2-antiplasmin, α2-macroglobulin, and antithrombin),23 holding periods > 30 minutes would presumably lead to a gradual decrease in coagulability in the whole blood sample. Reports of longer holding times for equine blood support this notion.17
Maximal clot strength (maximum clot firmness) was less profoundly affected by the ex vivo contact activation than were other thromboelastometry parameters. This finding is consistent with previous reports that maximum clot firmness is the measure that is least sensitive to the kinetics of thrombin formation19 because it is primarily a function of fibrinogen concentration and platelet numbers.26
Although the technology of the in vitro viscoelasticity monitoring systemh used in the present study is not identical to thromboelastography systems, it is quite similar.5,27,28 The sample and reagent volumes used for thrombelastography and thromboelastometry are slightly different, and the available reagents, cups, and pins are quite dissimilar.5,28 It is therefore likely that, had the study reported here been performed with thrombelastography, the raw data obtained would have been somewhat different. However, we have no reason to expect that different conclusions would have been reached had the alternate system been used. A recent report28 comparing results for human plasma obtained with the 2 systems did suggest that the cups and pins supplied for the thromboelastometry systemh used in the present study might be more charged than those used with thrombelastography, causing slightly greater contact activation in the sample during generation of the thromboelastometry tracing.
The materials used for blood collection could also possibly contribute to contact activation during sample holding. All of our samples were collected in an identical manner with a blood collection system that allowed samples to be collected directly into evacuated glass tubes containing citrate anticoagulant, although a previous report29 of variables affecting thromboelastography results for canine blood samples indicated minimal to no influence of sample collection method on the results obtained in the study.
To the authors’ knowledge, the influence of holding temperature on thromboelastometry or thromboelastography results has not been specifically addressed in previous studies of viscoelastic testing of equine blood. In a study20 of sample handling effects on thromboelastometry results for canine blood samples, collection of blood into room temperature (approx 25°C) tubes followed by holding at room temperature rapidly decreased the sample temperature to that of the ambient environment, and this may vary considerably. When a sample at room temperature is added to a thromboelastometry or thromboelastography cup at 37°C, it takes time for the sample to warm to this operating temperature, the time being dependent on sample volume and the ambient temperature.19 During this initial period, the speed of enzymatic reactions changes with the sample temperature. To eliminate this source of inconsistency, we elected to collect our samples into prewarmed tubes and to hold the samples at 37°C prior to initiation of thromboelastometry. The warmer temperature of storage may have impacted the rate of contact activation in these blood samples, potentially exaggerating the impact of storage as compared with storage at room temperature (which is more commonly described).6,7,12,13,16,29 As a general rule, enzymatic activity doubles for every 10°C increase in temperature, so we would expect that, had we stored the blood samples at a temperature of 25°C, the degree of contact activation would have been decreased by approximately 50%. Storage at cooler temperatures such as 4°C would be expected to delay contact activation, but would not be recommended because of the well-described ex vivo activation of platelets under these conditions.30 Regardless of temperature used, contact activation would be expected to occur in equine blood and consequently would influence the results obtained with recalcification for thromboelastometry and thromboelastography. Therefore, establishment of protocols for sample collection, preanalytic handling, and storage is essential for test precision and accuracy.
Aliquots of whole blood from each horse were obtained from 1 sample collected into a single evacuated tube. This may represent a study limitation in that the tubes may have been exposed to a greater degree of manipulation than in usual laboratory protocol. In addition, the study population was a convenience sample of 11 healthy horses, which may not have been representative of the equine population in our veterinary hospital and should not be extrapolated to different populations. However, selecting a true random sample of horses for this study was not logistically possible. Future studies, with larger study populations, should allow for evaluation of the effects of age, sex, and other variables on thromboelastometry results.
Although it seems rational that, to most closely resemble in vivo conditions, in vitro testing should minimally manipulate the sample, it must be recognized that clot formation in vivo is usually initiated by exposure of extravascular tissue factor,31 the contribution of which cannot be assessed by any in vitro methodology. Further, in the absence of any coagulation activator, recalcification of citrated blood allows for thrombin generation only via the contact pathway, which is irrelevant for in vivo hemostasis (although it may possibly have relevance to pathological thrombus development).23 Consequently, lack of application of an appropriate trigger means that the kinetics of clot formation are primarily dependent on ex vivo factors that have little association with the inherent hemostatic capacity of the animal from which the blood sample was collected. Recalcification of blood in the absence of a trigger is therefore not an acceptable method of assessing the hemostatic system in equine patients. Implementation of a consistent holding time prior to recalcification does not rectify the problem, considering that such an approach would affect the variable magnitude of the artifact and not eliminate the fact that the assay result was largely dependent on changes that occurred ex vivo.
The approach most likely to mimic in vivo conditions would likely be use of a tissue factor trigger. The difficulty comes in determining the optimal concentration of tissue factor to use in an assay; the amount needs to be adequate to maximally trigger thrombin formation and minimize the contribution of any ex vivo contact activation. The minimal amount of tissue factor required is likely to vary, depending on the species source of the tissue factor and of the blood being evaluated. Veterinary thromboelastography assays often use a prothrombin time reagent that contains recombinant human tissue factor.r When this reagent was diluted, a much higher concentration of the tissue factor was required to minimize the influence of factor XII activation in canine blood19 (approx 3,500-fold diluted) than in human blood (approx 17,000-fold diluted), for which contact activation ex vivo is much less profound.20 A previous report32 describing use of the same reagent diluted 3,600-fold for use with equine blood samples indicated an influence of ex vivo changes on thromboelastography results with holding times ≤ 1 hour. We would consequently expect that for equine blood, a higher concentration of tissue factor would be required to avoid the influence of ex vivo contact activation. The amount needed would be similar to, or greater than, that reported in some recent studies12,16 describing thromboelastography results for equine whole blood. Alternatively, there may be specific circumstances under which contact activation with a strong activator (eg, ellagic acid or kaolin) would be indicated. This approach, like the plasma-based aPTT, could potentially be more sensitive to deficiencies in intrinsic factors (eg, with hemophilia) and may be more sensitive for monitoring effects of anticoagulant treatment. The optimal trigger to be used for both thromboelastometry and thromboelastography testing of equine whole blood samples needs to be further evaluated in outcome-based studies of equine patients. However, our results suggested that use of simple recalcification should be eliminated for both thromboelastometry and thromboelastography performed on equine blood.
Acknowledgments
Portions of this study were presented in abstract form at the American College of Veterinary Internal Medicine Forum, Anaheim, Calif, June 2010.
ABBREVIATIONS
| aPTT | Activated partial thromboplastin time |
| SBTI | Soy bean trypsin inhibitor |
Footnotes
Mendez J, Vilar P, Mudge M, et al. Thromboelastography (TEG) for early detection of hemostatic abnormalities in horses with colic (abstr). J Vet Intern Med 2008;22:816.
Cell-dyn 3700 Hematology Analyzer, Abbott Diagnostics, Abbott Park, Ill.
Hitachi 917 Automatic Analyzer, Roche Ltd, Basel, Switzerland.
BD Eclipse, Becton-Dickson, Franklin Lakes, NJ.
BD Vacutainer Citrate tube, 4.5 mL, Becton-Dickson, Franklin Lakes, NJ.
Corn trypsin inhibitor, Haematologic Technologies, Essex Junction, Vt.
SBTI, Sigma-Aldrich, St Louis, Mo.
ROTEM, Pentapharm GMbH, Munich, Germany.
Human factor XII–deficient plasma, Haematologic Technologies, Essex Junction, Vt.
Phospholipids, Avanti Polar Lipids, Alabaster, Ala.
Ultralow cluster microplates, Corning Inc, Corning, NY.
Spectramax Microplate Reader, Molecular Devices Corp, Sunnyvale, Calif.
SigmaPlot, version 7.101, SPSS Inc, Chicago, Ill.
Human Factor XIIa, Haematologic Technologies, Essex Junction, Vt.
Human factor XI–deficient plasma, Haematologic Technologies, Essex Junction, Vt.
Human factor XIa, Haematologic Technologies, Essex Junction, Vt.
Sigma Stat, version 2.03, SPSS Inc, Chicago, Ill.
Innovin, Dade Behring Höchs, Liederbach, Germany.
References
1. Welch RD, Watkins JP, Taylor TS, et al. Disseminated intravascular coagulation associated with colic in 23 horses (1984–1989). J Vet Intern Med 1992; 6: 29–35.
2. Barton MH, Morris DD, Norton N, et al. Hemostatic and fibrinolytic indices in neonatal foals with presumed septicemia. J Vet Intern Med 1998; 12: 26–35.
3. Dolente BA, Wilkins PA, Boston RC. Clinicopathologic evidence of disseminated intravascular coagulation in horses with acute colitis. J Am Vet Med Assoc 2002; 220: 1034–1038.
4. Pusterla N, Luff JA, Myers CJ, et al. Disseminated intravascular coagulation in a horse with Streptococcus equi subspecies zooepidemicus meningoencephalitis and interstitial pneumonia. J Vet Intern Med 2007; 21: 344–347.
5. McMichael MA, Smith SA. Viscoelastic coagulation testing: technology, applications, and limitations. Vet Clin Pathol 2011; 40: 140–153.
6. Tennent-Brown BS, Epstein KL, Whelchel DD, et al. Use of viscoelastic coagulation testing to monitor low molecular weight heparin administration to healthy horses. J Vet Emerg Crit Care 2013; 23: 291–299.
7. Blong AE, Epstein KL, Brainard BM. In vitro effects of three formulations of hydroxyethyl starch solutions on coagulation and platelet function in horses. Am J Vet Res 2013; 74: 712–720.
8. McGovern KF, Lascola KM, Smith SA, et al. Assessment of acute moderate hyperglycemia on traditional and thromboelastometry coagulation parameters in healthy adult horses. J Vet Emerg Crit Care 2012; 22: 550–557.
9. McGovern KF, Lascola KM, Smith SA, et al. The effects of hyperglycemia and endotoxemia on coagulation parameters in healthy adult horses. J Vet Intern Med 2013; 27: 347–353.
10. Mendez-Angulo JL, Mudge M, Zaldivar-Lopez S, et al. Thromboelastography in healthy, sick non-septic and septic neonatal foals. Aust Vet J 2011; 89: 500–505.
11. Mendez-Angulo JL, Mudge MC, Vilar-Saavedra P, et al. Thromboelastography in healthy horses and horses with inflammatory gastrointestinal disorders and suspected coagulopathies. J Vet Emerg Crit Care 2010; 20: 488–493.
12. Giordano A, Meazza C, Salvadori M, et al. Thromboelastometric profiles of horses affected by exercise-induced pulmonary hemorrhages. Vet Med Int [serial online] 2010; 2010: 945789. Available at: www.hindawi.com/journals/vmi/2010/945789/. Accessed Jun 19, 2014.
13. Epstein KL, Brainard BM, Giguere S, et al. Serial viscoelastic and traditional coagulation testing in horses with gastrointestinal disease. J Vet Emerg Crit Care 2013; 23: 504–516.
14. Epstein KL, Brainard BM, Gomez-Ibanez SE, et al. Thrombelastography in horses with acute gastrointestinal disease. J Vet Intern Med 2011; 25: 307–314.
15. Dunkel B, Chan DL, Boston R, et al. Association between hypercoagulability and decreased survival in horses with ischemic or inflammatory gastrointestinal disease. J Vet Intern Med 2010; 24: 1467–1474.
16. McMichael M, Smith SA, McConachie EL, et al. In-vitro hypocoagulability on whole blood thromboelastometry associated with in-vivo expansion of red cell mass in an equine model. Blood Coagul Fibrinolysis 2011; 22: 424–430.
17. Epstein KL, Brainard BM, Lopes MA, et al. Thrombelastography in 26 healthy horses with and without activation by recombinant human tissue factor. J Vet Emerg Crit Care 2009; 19: 96–101.
18. Paltrinieri S, Meazza C, Giordano, et al. Validation of thromboelastometry in horses. Vet Clin Pathol 2008; 37: 277–285.
19. Laursen SH, Andersen PH, Kjelgaard-Hansen M, et al. Comparison of components of biological variation between 3 equine thromboelastography assays. Vet Clin Pathol 2013; 42: 443–450.
20. Smith SA, McMichael M, Galligan A, et al. Clot formation in canine whole blood as measured by rotational thromboelastometry is influenced by sample handling and coagulation activator. Blood Coagul Fibrinolysis 2010; 21: 692–702.
21. Wiinberg B, Jensen AL, Rojkjaer R, et al. Validation of human recombinant tissue factor-activated thromboelastography on citrated whole blood from clinically healthy dogs. Vet Clin Pathol 2005; 34: 389–393.
22. Camenzind V, Bombeli T, Seifert B, et al. Citrate storage affects Thrombelastograph analysis. Anesthesiology 2000; 92: 1242–1249.
23. Renne T. The procoagulant and proinflammatory plasma contact system. Semin Immunopathol 2012; 34: 31–41.
24. Roche AM, James MF, Grocott MP, et al. Just scratching the surface: varied coagulation effects of polymer containers on TEG variables. Eur J Anaesthesiol 2006; 23: 45–49.
25. Tans G, Griffin JH. Properties of sulfatides in factor-XII–dependent contact activation. Blood 1982; 59: 69–75.
26. Smith SA, McMichael MA, Gilor S, et al. Correlation of hematocrit, platelet concentration, and plasma coagulation factors with results of thromboelastometry in canine whole blood samples. Am J Vet Res 2012; 73: 789–798.
27. Kol A, Borjesson DL. Application of thrombelastography/thromboelastometry to veterinary medicine. Vet Clin Pathol 2010; 39: 405–416.
28. Nielsen VG. A comparison of the Thrombelastograph and the ROTEM. Blood Coagul Fibrinolysis 2007; 18: 247–252.
29. Ralph AG, Brainard BM, Pittman JR, et al. Effects of rest temperature, contact activation, and sample technique on canine thrombelastography. J Vet Emerg Crit Care 2012; 22: 320–326.
30. Tablin F, Walker NJ, Klein SD, et al. Animal models for studies on cold-induced platelet activation in human beings. J Lab Clin Med 2000; 135: 339–346.
31. Morrissey JH. Tissue factor: a key molecule in hemostatic and nonhemostatic systems. Int J Hematol 2004; 79: 103–108.
32. Leclere M, Lavoie JP, Dunn M, et al. Evaluation of a modified thrombelastography assay initiated with recombinant human tissue factor in clinically healthy horses. Vet Clin Pathol 2009; 38: 462–466.
