Basic Science Aspects
Establishment of Methods for Performing Thrombelastography and Calibrated Automated Thrombography in Rats
del Pilar Huby, Maria; Cardenas, Jessica C.; Baer, Lisa A.; Pawelczyk, Nick S.; Salsbury, John R.; Wang, Yao-Wei W.; Matijevic, Nena; Holcomb, John B.; Wade, Charles E.
Department of Surgery and Center for Translational Injury Research, University of Texas Health Science Center, Houston, Texas
Received 27 Nov 2013; first review completed 16 Dec 2013; accepted in final form 26 Feb 2014
Address reprint requests to Charles E. Wade, PhD, 6431 Fannin St, MSB 5.204, The University of Texas Health Science Center at Houston, Houston, TX 77030. E-mail: Charles.E.Wade@uth.tmc.edu.
The authors declare no relevant conflicts of interest.
Funding for this study was provided by The University of Texas Health Science Center at Houston’;s Center for Translational Injury Research.
ABSTRACT: Rodent models of hemorrhagic shock are paramount to our understanding of the pathophysiology of this disease, the effects on coagulation and in exploring the utility of resuscitative methods for managing patients in shock. These models usually require serial blood sampling during experimentation. The lack of standardized practices for these experimental models has resulted in technical variability, discordance in the literature, and incomparable results on blood coagulation analysis between researchers, hindering substantial progress in the field of hemorrhagic shock. The aim of this study was to define the effects of cardiac puncture versus arterial catheterization on coagulation in a rat model to provide data supporting standardization of one practice over another. Blood was collected from anesthetized rats via cardiac puncture or femoral artery catheterization and hemostatic potential analyzed by thrombelastography and calibrated automated thrombography. Our data show that blood collected via cardiac puncture demonstrated hypercoagulability as indicated by faster rates of clot formation and thrombin generation, increased overall clot strength, and a greater thrombin-generating capacity when compared with blood collected via femoral artery catheter. We conclude that blood collection methods have a profound effect on hemostatic potential, and standardization of these practices is necessary to define the effects of shock on coagulation in rodents.
There has been increased interest in the use of whole-blood assays of viscoelasticity, such as thrombelastography (TEG) and thromboelastometry (TEM), in the clinical diagnosis and study of the coagulation system (1–5). These assays have been suggested to replace standard clinical laboratory tests to evaluate coagulation because they provide more detailed information and faster results at a similar cost (6). Also, several groups have advocated using TEG to both predict and guide transfusion requirements during resuscitation to control hemorrhagic shock (6–10). In addition to the TEG and TEM, evidence has shown that thrombin generation patterns in patients with coagulation disorders are directly correlated with bleeding or thrombosis (11, 12). Therefore, measuring thrombin generation via calibrated automated thrombography (CAT) has been suggested as an effective method for monitoring prophylactic anticoagulation as well as factor replacement therapy in hemophilic patients (13, 14).
Interest in these assays has also carried over to the experimental laboratory where they have been used to define experimental models of altered coagulation in a variety of species. Recently, the Silliman Laboratory put forth a standardized technique for performing TEG in rodents and provided reference ranges for native TEG after femoral artery blood draw (15). However, its subsequent publications have demonstrated changes in coagulation parameters before and after induction of hemorrhagic shock using kaolin-activated TEG (16, 17). Importantly, the source of blood collected at baseline was from cardiac puncture, whereas postshock samples were collected from a femoral artery line. This discrepancy in the model potentially compromises any relevant conclusions that could otherwise be deduced from the effects of hemorrhagic shock on coagulation status had the samples been collected from the same source.
Experimental studies of coagulation in rodents, especially in studies of shock, are usually initiated after placement of catheters for vascular access (18). Admittedly, this surgical manipulation itself could alter the “baseline” coagulation status of the animal and confound experimental results, as reported previously (19, 20). Technical issues such as these combined with inconsistencies in the current literature highlight the need for standardizing methods for serial blood collection in rodent models of hemorrhagic shock. The aim of this study was to define the effects of cardiac puncture versus arterial catheterization on coagulation in a rat model to better interpret studies of shock.
MATERIALS AND METHODS
All experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Texas Health Science Center-Houston, and animals were maintained in accordance with the Guide for the Care and Use of Laboratory Animals. Adult male Sprague-Dawley rats (Harlan Laboratories, Indianapolis, Ind) weighing between 325 and 350 g were acclimatized for at least 3 days and housed in an accredited animal facility. Food and water were available ad libitum. Both body weight and health were monitored daily. Rats were divided into two groups by the method of blood collection: cardiac puncture ([CP] n = 10) and catheterized femoral artery ([CFA] n = 25).
Sample collection via cardiac puncture
Rats were anesthetized under 2% to 3% isoflurane in 100% O2 (Isol Vedcro Inc, St. Joseph, Mo) and placed in supine position on a heating pad (TCAT-2LV controller Thermocoupler; Physitemp, Clifton, NJ) to maintain a stable body temperature. After a 15-min stabilization period under anesthesia, 10.8 mL of blood was collected from cardiac puncture using a 20-gauge needle and a 12 cc/mL syringe prefilled with 1.2 mL of 3.2% sodium citrate (1:10 dilution). Blood was transferred back to two 4.5-mL and one 1.8-mL vacutainers containing no anticoagulant (BD Vacutainer, Franklin Lakes, NJ), gently inverted five times and placed in a vertical position for 30 min to ensure adequate anticoagulation.
Sample collection via catheterized femoral artery
Rats were anesthetized as described earlier. With no response to stimuli, the femoral superficial artery was isolated and catheterized with a 3F polyurethane catheter (Access Technologies, Skokie, Ill) and kept patent using an overnight heparin solution wash of 15 UI/mL (Heparin 1,000 UI/mL; Hospira, Lake Forest, Ill), which was flushed with saline before the experiment. After a 15-min stabilization period from the time the arterial line was placed, a 2.0-mL blood sample was obtained using a heparin-coated syringe. From the syringe, 1.8 mL of blood was transferred to a vacutainer containing 200 μL of 3.2% sodium citrate (1:10 dilution) and gently inverted five times and placed in a vertical position for 30 min to ensure adequate anticoagulation.
For both sampling methods, blood was either used fresh (TEG) or centrifuged at 3,200 RPM for 15 min at 21°C, aliquoted, and frozen at -80°C.
Thrombelastography was performed using the Thrombelastograph Hemostasis Analyzer 5000 series (Haemoscope Corporation, Niles, Ill) to measure the initiation, formation, strength, and stability of whole-blood clots. Blood samples used for this assay were run immediately after the 30-min anticoagulation period.
From the CP group (n = 10), 1.0 mL of citrated blood from the 1.8-mL vacutainer was pipetted into a kaolin vial (Haemoscope Corporation) and gently inverted five times. From the vial, 340 μL of blood was pipetted into a plastic cup with 20 μL of 0.2 M calcium chloride (Haemoscope Corporation).
From the CFA group (n = 25), 1.0 mL of citrated blood from the 1.8-mL vacutainer was pipetted into a kaolin vial (Haemoscope Corporation) and gently inverted five times. From the vial, 340 μL of blood was pipetted into a plastic heparinase cup (Haemoscope Corporation) to eliminate any heparin contaminating the sample from the washed catheter, with 20 μL of 0.2 M calcium chloride. Control experiments were performed to evaluate potential effects of residual heparin or the presence of heparinase on TEG results. These data showed negligible changes and thus confirmed both adequate flushing of heparin and that heparinase did not confound our interpretation of observations made between sampling techniques (data not shown).
Parameters recorded were split point ([SP] in minutes), reaction time ([R] in minutes), clotting time ([K] in minutes), angle ([α] in degrees), maximum amplitude ([MA] in millimeters), clot strength ([G] in dyne square centimeter), clot lysis at 30 min ([LY30] in percent), delta (Δ), the greatest clot growth secondary to thrombin generation (R-SP).
Thrombin generation was performed using CAT to measure hemostatic potential (Thrombinoscope BV, Maastricht, the Netherlands), as developed by Hemker et al. (13). Frozen plasma samples from the CP (n = 10) and CFA (n = 32) groups were thawed in 37°C water baths.. Thrombin generation was measured using a plate reader equipped with both 390/460-nm fluorescent filter and dispenser (Thermolab Systems OY, Helsinki, Finland). As recommended by Stago, only half of a 96-well plate (Immulon 2HB; Thermo Scientific, Rochester, NY) was used because animal plasmas tend to have short lag times and rapid thrombin generation (21). Samples were run in duplicate, with two wells of 20 μL of PPP-low trigger solution (1 pM tissue factor [TF] and 4 μmol/L phospholipids) and two wells of 20 μL of calibrator solution (Diagnostica Stago, NJ). Before addition of plasma to wells, 360 μL of plasma from both CP and CFA groups was treated by pipetting up and down three times into a heparinase cup (to eliminate the heparin effect) (Haemoscope Corporation). Immediately after mixing the plasma in a heparinase cup, 80 μL was pipetted into each well. The plate containing plasma and CAT reagent or calibrator was placed into a fluorometer to warm to 37°C. Thrombin generation was initiated by autodispensing 20 μL of FluCa solution containing the fluorogenic substrate and calcium (Diagnostica Stago) to each well. Thrombin generation was monitored at 37°C for 60 min at 20-s intervals. Thrombinoscope software (Thrombinoscope BV) was used to compare the thrombin to calibrator activity, thus calculating thrombin generation over time. Parameters recorded were lag time (in minutes), endogenous thrombin potential (ETP), peak height (in nanomolar), time to peak ([ttpeak] in minutes), and start of tail (in minutes).
Data were analyzed using SigmaPlot version 11.0 (Systat Software Inc., Chicago, Ill). Data were analyzed by one-way analysis of variance using the Student-Newman-Kuels post hoc method. Statistical significance was defined as P < 0.05.
Thrombelastography values and calculated reference ranges for the CP and CFA groups are listed in Table 1. Significantly lower values were observed in the CFA group compared with those in the CP group for SP (4.5 [0.7] vs. 1.9 [0.6]), R (5.3 [0.8] vs. 2.2 [0.7]), α (57.8 [8.5] vs. 80.3 [1.5]), and Δ (0.82 [0.25] vs. 0.24 [0.03]). Also, the K value was significantly higher in the CFA group compared with the CP group (2.6 [0.9] vs. 0.8 ). A difference of 58% was observed in SP, 59% in R time, 69% in K time, and 28% in α. The clot strength was also significantly lower in the CFA group compared with the CP group (MA 58.0 [4.9] vs. 73.3 [1.7] and G 7.1 [1.5] vs. 13.8 [1.2]). A difference of 21% was seen in the MA and 49% in the G. Reference ranges were calculated as ±2 SDs of the mean values.
Calibrated automated thrombography values and calculated reference ranges for the CP and CFA groups are listed in Table 2. Excluding start of tail, values for lag time and ttPeak were significantly higher in the CFA group compared with those in the CP group (lag time, 3.4 [0.5] vs. 1.9 [0.3] and ttPeak 5.6 [0.7] vs. 3.8 [0.4]). Also, the ETP and peak height were significantly lower in the CFA group compared with those in the CP group (ETP, 227.9 [63.3] vs. 389.4 [38.3] and peak height 59.5 [23.0] vs. 120.6 [20.9]). Thus, thrombin generation was significantly prolonged and reduced in the CFA group as compared with the CP group. A difference of 44% was seen in the lag time, 42% in the ETP, 51% in peak height, and 32% in the ttPeak. There was no significant difference observed in the start of tail. Reference ranges were calculated as ±2 SDs of the mean values.
The purpose of our study was to establish proper methods and reference ranges for coagulation studies in adult male Sprague-Dawley rats. Previous studies used different techniques for obtaining blood for coagulation studies. Ignoring surgical manipulation or differences in hemostatic potential of blood sampling location can have dramatic effects on the data. Therefore, after sampling through either cardiac puncture or arterial catheter, viscoelastometry in whole blood via TEG and thrombin generation in plasma via CAT were measured and compared. Our study confirmed that the different blood collection methods have a profound effect on coagulation parameters in rodents. We found that blood collected via cardiac puncture demonstrated hypercoagulable properties, as indicated by faster rates of clot formation and thrombin generation, increased overall clot strength, and a greater thrombin-generating capacity when compared with blood collected via a femoral artery catheter.
These findings can mostly likely be explained by TF contamination in the cardiac puncture samples. Tissue factor is a transmembrane glycoprotein that, when bound to its cofactor FVIIa, serves as a potent initiator of the extrinsic coagulation pathway. Tissue factor is preferentially expressed at high levels in vital organs, such as the brain, lungs, uterus, and heart (22, 23). Importantly, constitutive expression of highly functional and procoagulant TF is found in cardiac muscle (24). Presumably, TF present in the pericardium will contaminate blood samples obtained from cardiac puncture, thus priming the blood for rapid and robust coagulation on replacement of calcium. Calibrated automated thrombography is sensitive to increased TF levels, producing shortened lag times and increased overall thrombin generation (peak and ETP) (25). Furthermore, TF contamination would result in activation of both intrinsic and extrinsic pathways in the kaolin TEG assay, which could explain the hypercoagulability of the cardiac puncture samples observed in this assay. Alternatively, the materials used to execute blood collection could affect its overall hemostatic potential, namely, the potential for activation of whole blood by metal needles and plastic syringes used for cardiac puncture compared with the polyurethane catheter for femoral artery blood.
Previous publications have attempted to standardize a method for performing coagulation assays in rodents. Wohlauer et al. (15) described the use of TEG to analyze rat blood collected from the femoral artery; however, these data were collected using native TEG, which simply adds back calcium to decalcified whole blood. Given that the use of an activator, such as kaolin, is more feasible and relevant clinically, subsequent publications from this group describe a rat model of hemorrhagic shock commonly used in our field of trauma research (15, 16). The first publication with this method used a model of serial blood sample collection where the baseline sample is taken via cardiac puncture and by femoral artery line thereafter (16). Data from this study demonstrated significant prolongation of the R and SP times and reduced thrombin generation in the second blood sample taken via femoral artery catheter after induction of shock. These data were attributed to impaired clotting factor function resulting from trauma/shock (16). Although this may be true, our data suggest that the differences between baseline and postshock could simply be caused by the hypercoagulable nature of cardiac puncture–derived blood samples compared with the femoral artery line. Interestingly, however, the subsequent publication from this laboratory using a model of serial sample collection via cardiac puncture showed no difference in the enzymatic phases of the TEG in preshock and postshock blood samples (17). This finding supports our suggestion that postshock hypocoagulability is not pathophysiologic but rather resulted from sample collection methods.
Other data in the literature describe findings in nonrodent models using femoral artery catheterization, followed by subsequent further vascular manipulation. King et al. (19) reported TEGs on swine and humans before and after pulmonary artery catheterization. The results indicated that pulmonary artery catheterization in both swine and humans produced a significant hypercoagulable state, according to the R and K times and α angle compared with femoral artery catheterization alone. In a similar study, Ryan et al. (20) reported TEG on swine and humans before and after central venous catheterization, also demonstrating a significant hypercoagulable state after central venous catheterization indicated by the same TEG variables compared with single catheter placement. Taken together, these studies reiterate the importance of acknowledging the potential effects of surgical and vascular manipulation when serial sampling is necessary and the need to standardize models for accurate data comparison.
Finally, our study has demonstrated the relative agreement in data trends between TEG and CAT. The observation that changes in enzymatic rates and overall clot strength in whole-blood TEG were emulated by the same directional changes on CAT lag time and thrombin generation performed in platelet-poor plasma suggests that TEG data are highly reflective of coagulation factor levels and the ability to generate thrombin. Previous data have shown that thrombin concentration directly dictates fibrin clot structure and stability (26). Given the clinical importance of producing and maintaining strong and stable clots during hemorrhage, the CAT assay could offer an important resource for defining the hemostatic potential of plasma samples in rodent hemorrhagic shock models.
A limitation in the present study is the use of heparinase to negate the effects of the heparin used to wash the femoral lines. Although the femoral catheters were flushed with saline, admittedly, some lingering effects of heparin could be present. Previous studies measuring the effects of heparinase on nonheparinized blood samples showed minimal changes in TEG parameters after heparinase treatment. Thus, although the likelihood that heparinase resulted in a significant influence to the data reported here is low (27), we cannot fully discount its contribution.
Coagulation studies in rodents, particularly those involving serial blood sampling, need to take into account the effects of surgical manipulation and blood collection methods on coagulation parameters. We have demonstrated that both TEG and CAT are sensitive to these factors and can have profound effects on data interpretation. Standardization of practices surrounding rodent models of trauma and shock are necessary to move forward with research on the effects of shock on coagulation.
1. Kaufmann CR, Dwyer KM, Crews JD, Dols SJ, Trask AL: Usefulness of thrombelastography in assessment of trauma patient coagulation. J Trauma 42 (4): 716–720, 1997.
2. Carroll RC, Craft RM, Langdon RJ, Clanton CR, Snider CC, Wellons DD, Dakin PA, Lawson CM, Enderson BL, Kurek SJ: Early evaluation of acute traumatic coagulopathy by thrombelastography. Transl Res 154 (1): 34–39, 2009.
3. Schochl H, Frietsch T, Pavelka M, Jambor C: Hyperfibrinolysis after major trauma: differential diagnosis of lysis patterns and prognostic value of thrombelastometry. J Trauma 67 (1): 125–131, 2009.
4. Sivula M, Pettila V, Niemi TT, Varpula M, Kuitunen AH: Thromboelastometry in patients with severe sepsis and disseminated intravascular coagulation. Blood Coagul Fibrinolysis 20 (6): 419–426, 2009.
5. Branco BC, Inaba K, Ives C, Okoye O, Shulman I, David JS, Schochl H, Rhee P, Demetriades D: Thromboelastogram evaluation of the impact of hypercoagulability in trauma patients. Shock 41 (3): 200–207, 2014.
6. Cotton BA, Faz G, Hatch QM, Radwan ZA, Podbielski J, Wade C, Kozar RA, Holcomb JB: Rapid thrombelastography delivers real-time results that predict transfusion within 1 hour of admission. J Trauma 71 (2): 407–414, 2011.
7. Kheirabadi BS, Crissey JM, Deguzman R, Holcomb JB: In vivo bleeding time and in vitro thrombelastography measurements are better indicators of dilutional hypothermic coagulopathy than prothrombin time. J Trauma 62 (6): 1352–1359, 2007.
8. White H, Zollinger C, Jones M, Bird R: Can thromboelastography performed on kaolin-activated citrated samples from critically ill patients provide stable and consistent parameters? Int J Lab Hematol 32 (2): 167–173, 2010.
9. Holcomb JB, Minei KM, Scerbo ML, Radwan ZA, Wade CE, Kozar RA, Gill BS, Albarado R, McNutt MK, Khan S, et al.: Admission rapid thrombelastography can replace conventional coagulation tests in the emergency department: experience with 1974 consecutive trauma patients. Ann Surg 256 (3): 476–486, 2012.
10. Vogel AM, Radwan ZA, Cox CS Jr, Cotton BA: Admission rapid thrombelastography delivers real-time “actionable” data in pediatric trauma. J Pediatr Surg 48 (6): 1371–1376, 2013.
11. Trossaert M, Regnault V, Sigaud M, Boisseau P, Fressinaud E, Lecompte T: Mild hemophilia A with factor VIII assay discrepancy: using thrombin generation assay to assess the bleeding phenotype. J Thromb Haemost 6 (3): 486–493, 2008.
12. Besser M, Baglin C, Luddington R, van Hylckama Vlieg A, Baglin T: High rate of unprovoked recurrent venous thrombosis is associated with high thrombin-generating potential in a prospective cohort study. J Thromb Haemost 6 (10): 1720–1725, 2008.
13. Hemker HC, Giesen P, Al Dieri R, Regnault V, de Smedt E, Wagenvoord R, Lecompte T, Beguin S: Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb 33 (1): 4–15, 2003.
14. Dargaud Y, Sorensen B, Shima M, Hayward C, Srivastava A, Negrier C: Global haemostasis and point of care testing. Haemophilia 18 (Suppl 4): 81–88, 2012.
15. Wohlauer MV, Moore EE, Harr J, Gonzalez E, Fragoso M, Silliman CC: A standardized technique for performing thromboelastography in rodents. Shock 36 (5): 524–526, 2011.
16. Harr JN, Moore EE, Wohlauer MV, Droz N, Fragoso M, Banerjee A, Silliman CC: The acute coagulopathy of trauma is due to impaired initial thrombin generation but not clot formation or clot strength. J Surg Res 170 (2): 319–324, 2011.
17. Wohlauer MV, Moore EE, Droz NM, Harr J, Gonzalez E, Fragoso M, Silliman CC: Hemodilution is not critical in the pathogenesis of the acute coagulopathy of trauma. J Surg Res 173 (1): 26–30, 2012.
18. Darlington DN, Craig T, Gonzales MD, Schwacha MG, Cap AP, Dubick MA: Acute coagulopathy of trauma in the rat. Shock 39 (5): 440–446, 2013.
19. King DR, Cohn SM, Feinstein AJ, Proctor KG: Systemic coagulation changes caused by pulmonary artery catheters: laboratory findings and clinical correlation. J Trauma 59 (4): 853–857, 2005.
20. Ryan ML, Thorson CM, King DR, Van Haren RM, Manning RJ, Andrews DM, Livingstone AS, Proctor KG: Insertion of central venous catheters induces a hypercoagulable state. J Trauma Acute Care Surg 73 (2): 385–390, 2012.
21. Tchaikovski SN, Van Vlijmen BJ, Rosing J, Tans G: Development of a calibrated automated thrombography based thrombin generation test in mouse plasma. J Thromb Haemost 5 (10): 2079–2086, 2007.
22. Mackman N: Role of tissue factor in hemostasis and thrombosis. Blood Cells Mol Dis 36 (2): 104–107, 2006.
23. Mackman N: The role of tissue factor and factor VIIa in hemostasis. Anesth Analg 108 (5): 1447–1452, 2009.
24. Luther T, Mackman N: Tissue factor in the heart. Multiple roles in hemostasis, thrombosis, and inflammation. Trends Cardiovasc Med 11 (8): 307–312, 2001.
25. Ollivier V, Wang J, Manly D, Machlus KR, Wolberg AS, Jandrot-Perrus M, Mackman N: Detection of endogenous tissue factor levels in plasma using the calibrated automated thrombogram assay. Thromb Res 125 (1): 90–96, 2010.
26. Wolberg AS: Thrombin generation and fibrin clot structure. Blood Rev 21 (3): 131–142, 2007.
27. Ostrowski SR, Johansson PI: Endothelial glycocalyx degradation induces endogenous heparinization in patients with severe injury and early traumatic coagulopathy. J Trauma Acute Care Surg 73 (1): 60–66, 2012.
Shock; rodents; coagulation; blood collection; experimental model
© 2014 by the Shock Society
Highlight selected keywords in the article text.