Thromboelastometry and TEG have recently received increased attention as in vitro methods of assessing hemostasis. In theory, TEM and TEG should have several possible advantages over plasma-based coagulation assays because the techniques incorporate the contributions of cellular components, assess clot formation over time, can detect conditions of hyper- and hypocoagulability, and evaluate clot dissolution.
Despite the theoretical advantages of TEM and TEG over traditional plasma-based coagulation assessments, the clinical usefulness of the techniques for human and veterinary populations remains to be demonstrated. Descriptions of clinical and experimental use of TEM and TEG in human medicine are widespread, but these methods have yet to achieve the standard of care for most hemostatic disorders.1,2 Descriptions of TEG in the veterinary medical literature include its use to detect the prevalence of abnormalities in specific dog populations3–6 and a single report7 in which hypocoagulability detected via TEG was correlated with clinical evidence of bleeding in dogs.
By their nature, results of TEM and TEG are postulated to be more reflective of the in vivo hemostatic capabilities of an individual, compared with results of plasma-based coagulation testing, and are therefore likely to provide better diagnostic and therapeutic guidance. Both techniques are reported to be sensitive to abnormalities of coagulation factors, platelet function, platelet number, and the interaction of platelets with the developing fibrin clot.1 The purpose of the study reported here was to evaluate which characteristics of blood samples collected from dogs influence the results obtained via TEM. We expected that coagulation results obtained via TEM for whole blood samples would be correlated with concentrations of coagulation factors and platelets within the blood samples and that the concentration of RBCs within the whole blood samples would have a marked influence on the results obtained via TEM.
Materials and Methods
Animals—Healthy pet dogs were volunteered by their owners for inclusion in the study. All dogs were not receiving any medication other than routine heartworm and flea prophylaxis. A variety of purebred and mixed-breed dogs were included, but Greyhounds were excluded due to their breed-specific abnormalities in coagulation.8 This project was approved by the Institutional Animal Care and Use Committee, and appropriate written consent from owners was obtained.
Blood sample collection and processing—Food was withheld from each dog for 12 hours prior to blood sample collection. Each dog underwent atraumatic venipuncture of a jugular vein once by use of a 21-gauge collection set with a vacuum tube adaptor.a At each sample collection, 10 mL of blood was collected directly into each of 3 vacuum tubes. The order of sample collection was as follows: blood collected in a tube containing no anticoagulant (from which serum was obtained), blood collected into a tube containing EDTA (for cell count assessments), and blood collected into a tube containing citrate anticoagulant (for TEM procedures and to provide plasma for other analyses [eg, coagulation tests and assessments of coagulation factor concentrations]). This approach was used to expose collected blood samples to the citrate anticoagulant rapidly and to minimize tissue factor contamination of citrated samples. Citrated whole blood samples were collected in 3.2% citrate (ratio, 9:1) into prewarmed (37°C) siliconized glass tubes,b then mixed thoroughly by gentle repeated inversion. Tubes were prewarmed so that the temperature of each blood sample would be 37°C at the time TEM was initiated.9
Citrated whole blood samples were centrifuged at 2,500 × g for 10 minutes at room temperature (25°C) for collection of platelet-poor plasma. Aliquots of plasma were placed into 1.5-mL tubes and flash frozen on 100% ethanol and dry ice. Plasma was stored at −80°C for up to 90 days until plasmatic coagulation testing was performed in batches.
TEM—Thromboelastometryc was performed (according to the manufacturer's instructions) within 30 minutes after blood sample collection by use of methods previously described for canine samples.9 Prewarmed (37°C) citrated whole blood samples (each 300 μL) were added to the clotting reagents in the cup by use of the supplied electronic pipette. All clotting reactions contained 20 μL of a manufacturer-supplied concentrated calcium chloride reagentd and 20 μL of either supplied tissue factore or ellagic acidf reagent. Clotting reactions were followed for 1 hour. Data obtained from the manufacturer's modeling software included CT, the lag period from initiation of the reaction until an amplitude of 2 mm is recorded; CFT, the duration necessary for amplitude to increase from 2 to 20 mm; α angle, the angle of change of the amplitude over the course of the CFT; and MCF, the greatest amplitude achieved. The TEM variables, CT, CFT, α angle, and MCF are comparable to reaction time (R), clot kinetics (K), α angle, and MA, respectively, as reported elsewhere for TEG.
Plasma coagulation tests and assessment of coagulation factor concentrations—Prothrombin timeg and aPTTh were performed by use of manufacturer-recommended methods. Plasma fibrinogen concentrations were measured by use of the method of Claussi in reference to a human fibrinogen standard.j To measure extrinsic and common pathway factor concentrations (FVII, FX, FV, and prothrombin), a 10-fold dilution of each plasma sample with Owren-Koller bufferk was prepared; 50 μL of each diluted test sample was subsequently clotted with 50 μL of factor-deficient human plasmal and 100 μL of PT reagent. To measure intrinsic pathway factor concentrations (FXII, FXI, FIX, and FVIII), a 10-fold dilution of each plasma sample with Owren-Koller buffer was prepared; 50 μL of each diluted test sample was subsequently clotted with 50 L of factor-deficient human plasma and 50 μL of aPTT reagent, incubated for 180 seconds, followed by the addition of 50 μL of 25mM CaCl2. Factor concentrations were calculated in reference to a standard curve generated from serial dilutions of pooled normal canine plasma. Pooled normal canine citrated plasma was locally prepared by mixing equal volumes of plasma from 30 healthy dogs; for each dog, results of a CBC, serum biochemical analysis, and PT and aPTT assessments were within reference intervals. All tests were evaluated at 37°C and performed on a mechanical coagulometer.m For analyses evaluating coagulation factor mass in the TEM reaction, plasma coagulation factor concentrations were corrected for Hct by use of the following formula: F% × ([100 − Hct]/50), where F% is the plasma coagulation factor result obtained in reference to pooled normal plasma (defined as 100%).
Cell counts—Complete blood countsn were performed on samples of whole blood anticoagulated with EDTA. The CBCs were performed in the clinical laboratory at the Veterinary Teaching Hospital.
Statistical analysis—Data sets were evaluated for normality by use of the Kolmogorov-Smirnov test. Due to lack of normal distribution, correlation between values was determined by Spearman rank order correlation. Multiple variable linear regression models were determined by forward sequential addition of available data sets to the model, with TEM variable as the dependent variable and individual coagulation factor concentration, platelet concentration, or Hct as independent variables. Variables that had multicolinearity and those that failed to improve performance of the model were eliminated. Values of R2 were used to choose the best model. Analyses were performed by use of statistical software.o Values of P < 0.05 were considered significant.
Results
Blood sample collection and testing—All 127 dogs underwent 1 venipuncture episode during which blood samples were collected into the 3 tubes (a tube containing no anticoagulant, a tube containing EDTA, and a tube containing citrate). However, largely because of technical constraints associated with the methods used in the study, not all tests could be performed as described for each sample. Only the test results that were generated as described were used in the data analyses; thus, the number of available data sets for comparison differed among variables.
Correlation between TF-TEM and EA-TEM variables—For TF-TEM, membrane-anchored tissue factor was used, which initiates thrombin generation via the extrinsic pathway. For EA-TEM, ellagic acid was used, which initiates thrombin generation via the contact pathway. Variables assessed via TF-TEM and EA-TEM in whole blood samples from 22 dogs were compared (Figure 1). Values of CT determined via TF-TEM and EA-TEM were weakly correlated; values of CFT, α angle, and MCF determined via TF-TEM and EA-TEM were more strongly correlated.
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM and EA-TEM in whole blood samples obtained from 22 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with either tissue factor or ellagic acid (for assessment of TF-TEM– or EA-TEM–derived variables, respectively). Spearman rank order correlation coefficients (rs) and P values are indicated. All 127 study dogs underwent 1 venipuncture episode during which blood samples were collected; however, largely because of technical constraints associated with the methods used, not all tests could be performed per protocol for each sample. Thus, the number of available data sets was < 127.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM and EA-TEM in whole blood samples obtained from 22 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with either tissue factor or ellagic acid (for assessment of TF-TEM– or EA-TEM–derived variables, respectively). Spearman rank order correlation coefficients (rs) and P values are indicated. All 127 study dogs underwent 1 venipuncture episode during which blood samples were collected; however, largely because of technical constraints associated with the methods used, not all tests could be performed per protocol for each sample. Thus, the number of available data sets was < 127.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM and EA-TEM in whole blood samples obtained from 22 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with either tissue factor or ellagic acid (for assessment of TF-TEM– or EA-TEM–derived variables, respectively). Spearman rank order correlation coefficients (rs) and P values are indicated. All 127 study dogs underwent 1 venipuncture episode during which blood samples were collected; however, largely because of technical constraints associated with the methods used, not all tests could be performed per protocol for each sample. Thus, the number of available data sets was < 127.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Correlation of results of TEM with those of routine plasma coagulation tests—For whole blood samples collected from 58 dogs, data obtained via TF-TEM were compared with results of coagulation tests performed on plasma samples obtained from the same dogs. Spearman rank order correlation values and the associated P values were calculated (Table 1). Prothrombin times were significantly correlated with CT, CFT, α angle, and MCF determined via TF-TEM but were most strongly correlated with CT values. The CT values but not other TF-TEM variables were also correlated with plasma concentrations of FVII, FV, FIX, and prothrombin. Plasma fibrinogen concentration was correlated with TF-TEM–determined CFT, α angle, and MCF but not CT.
Spearman rank order correlation coefficients (P value) for CT, CFT, α angle, and MCF determined via TF-TEM and plasma coagulation factor concentrations in whole blood samples obtained from 58 healthy dogs.*
| Variable | CT | CFT | α angle | MCF |
|---|---|---|---|---|
| PT | 0.396 (0.002) | 0.276 (0.038) | −0.265 (0.047) | −0.315 (0.019) |
| aPTT | −0.161 (0.226) | 0.052 (0.699) | −0.080 (0.547) | −0.086 (0.529) |
| FVII | −0.386 (0.003) | −0.132 (0.323) | 0.176 (0.187) | 0.168 (0.215) |
| FX | −0.127 (0.341) | −0.168 (0.208) | 0.164 (0.217) | 0.091 (0.502) |
| FV | −0.309 (0.019) | −0.168 (0.208) | 0.164 (0.217) | 0.091 (0.502) |
| Prothrombin | −0.268 (0.042) | −0.144 (0.280) | 0.139 (0.290) | 0.151 (0.265) |
| Fibrinogen | −0.169 (0.153) | −0.478 (< 0.001) | 0.447 (< 0.001) | 0.460 (< 0.001) |
| FXII | −0.300 (0.172) | −0.067 (0.762) | −0.099 (0.657) | 0.013 (0.957) |
| FXI | −0.248 (0.262) | −0.297 (0.177) | 0.210 (0.342) | 0.365 (0.111) |
| FIX | −0.524 (0.012) | −0.203 (0.199) | 0.204 (0.358) | 0.215 (0.357) |
| FVIII | −0.278 (0.206) | −0.222 (0.316) | 0.160 (0.160) | 0.576 (0.008) |
Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing citrate anticoagulant to provide plasma (assessment of coagulation factor concentrations).
All 127 study dogs underwent 1 venipuncture episode during which blood samples were collected; however, largely because of technical constraints associated with the methods used, not all tests could be performed per protocol for each sample. Thus, the number of available data sets was < 127.
For whole blood samples collected from 31 dogs, data obtained via EA-TEM were compared with results of coagulation tests performed on plasma samples obtained from the same dogs. Spearman rank order correlation values and the associated P values were calculated (Table 2). Prothrombin times were significantly correlated with CT, CFT, α angle, and MCF determined via EA-TEM, but aPTT results were not. The CT values but not other EA-TEM variables also correlated with plasma concentrations of FVII, FX, FV, and FIX. Plasma fibrinogen concentration was correlated with EA-TEM–determined CFT, α angle, and MCF but not CT.
Spearman rank order correlation coefficients (P value) for CT, CFT, α angle, and MCF determined via EA-TEM and plasma coagulation factor concentrations in whole blood samples obtained from 31 healthy dogs.*
| Variable | CT | CFT | α angle | MCF |
|---|---|---|---|---|
| PT | 0.454 (0.034) | 0.539 (0.009) | −0.541 (0.009) | −0.586 (0.009) |
| aPTT | 0.102 (0.640) | −0.121 (0.579) | −0.118 (0.588) | −0.014 (0.948) |
| FVII | −0.437 (0.037) | 0.324 (0.129) | 0.313 (0.143) | 0.378 (0.075) |
| FX | −0.384 (0.033) | −0.140 (0.376) | 0.142 (0.443) | 0.226 (0.228) |
| FV | −0.448 (0.019) | −0.383 (0.034) | 0.366 (0.043) | 0.353 (0.056) |
| Prothrombin | −0.323 (0.080) | −0.323 (0.076) | 0.295 (0.106) | 0.317 (0.088) |
| Fibrinogen | −0.177 (0.415) | −0.815 (, 0.001) | 0.772 (< 0.001) | 0.732 (< 0.001) |
| FXII | −0.248 (0.178) | −0.150 (0.416) | 0.147 (0.426) | 0.110 (0.564) |
| FXI | 0.023 (0.901) | −0.062 (0.736) | 0.060 (0.746) | 0.006 (0.972) |
| FIX | −0.492 (0.005) | −0.321 (0.070) | 0.297 (0.104) | 0.360 (0.051) |
| FVIII | 0.087 (0.639) | 0.094 (0.611) | −0.085 (0.646) | −0.081 (0.665) |
See Table 1 for key.
Correlation of results of TEM with platelet concentration—For 71 dogs, data obtained from citrated blood samples via TF-TEM were compared with platelet concentration in EDTA-treated blood samples collected during the same venipuncture episode (Figure 2). Platelet concentration was significantly correlated with CT, CFT, α angle, and MCF determined via TF-TEM; the correlation with CT was weakest.
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM versus platelet concentration in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing EDTA (assessment of platelet concentration). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM versus platelet concentration in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing EDTA (assessment of platelet concentration). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM versus platelet concentration in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing EDTA (assessment of platelet concentration). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
For 31 dogs, data obtained from citrated blood samples via EA-TEM were compared with platelet concentration in EDTA-treated blood samples collected during the same venipuncture episode (Figure 3). Platelet concentration was correlated with CFT, α angle, and MCF determined via EA-TEM.
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM versus platelet concentration in whole blood samples obtained from 31 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing EDTA (assessment of platelet concentration). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM versus platelet concentration in whole blood samples obtained from 31 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing EDTA (assessment of platelet concentration). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM versus platelet concentration in whole blood samples obtained from 31 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing EDTA (assessment of platelet concentration). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Correlation of results of TEM with indicators of RBC mass—For 71 dogs, data obtained from citrated blood samples via TF-TEM were compared with Hct in EDTA-treated blood samples collected during the same venipuncture episode (Figure 4). Hematocrit was significantly correlated with CT, CFT, α angle, and MCF determined via TF-TEM; the correlation with CT was weak, compared with that for CFT, α angle, and MCF. Erythrocyte concentration was similarly correlated with CT (rs = 0.383; P = 0.002), CFT (rs = 0.441; P < 0.001), α angle (rs = −0.401; P < 0.001), and MCF (rs = −0.407; P < 0.001) determined via TF-TEM. Hemoglobin concentration was correlated with CT (rs = 0.443; P < 0.001), CFT (rs = 0.545; P < 0.001), α angle (rs = −0.497; P < 0.001), and MCF (rs = −0.469; P < 0.001).
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM versus Hct in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing EDTA (assessment of Hct). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM versus Hct in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing EDTA (assessment of Hct). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM versus Hct in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing EDTA (assessment of Hct). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
For 31 dogs, data obtained from citrated blood samples via EA-TEM were compared with Hct in EDTA-treated blood samples collected during the same venipuncture episode (Figure 5). Hematocrit was significantly correlated with CFT, α angle, and MCF. Erythrocyte concentration was similarly correlated with CT (rs = 0.311; P = 0.1), CFT (rs = 0.536; P = 0.002), α angle (rs = −0.543; P = 0.002), and MCF (rs = −0.423; P = 0.020) determined via EA-TEM. Hemoglobin concentration was correlated with CT (rs = 0.209; P = 0.270), CFT (rs = 0.596; P < 0.001), α angle(rs = −0.606; P < 0.001), and MCF (rs = −0.483; P = 0.009).
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM versus Hct in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing EDTA (assessment of Hct). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM versus Hct in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing EDTA (assessment of Hct). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM versus Hct in whole blood samples obtained from 71 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing EDTA (assessment of Hct). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Correction of factor concentrations for plasma volume in whole blood samples—Because sample Hct was strongly correlated with multiple TEM variables, we hypothesized that the potential influence of Hct on TEM results could be a function of the impact of RBC mass on the volume of plasma within a whole blood sample that is loaded into the TEM cup. A TEM reaction performed on a sample with a high Hct would contain less plasma (and consequently less plasma protein mass) than a sample with a lower Hct. Consequently, plasma coagulation factor concentrations were corrected for the volume of plasma within the TEM reaction sample and reevaluated for correlation of coagulation factor concentrations with TEM variables.o
Correlation coefficients (before and after correction for plasma volume load) between coagulation factor concentrations and variables determined via TF-TEM or EA-TEM were calculated (Figures 6 and 7). Correction of the factor concentration for plasma volume load in TEM reaction samples improved the strength of correlations between most coagulation factors and variables determined via TF-TEM or EA-TEM. Following correction for plasma volume load, corrected FVII, FX, FV, and prothrombin concentrations were significantly correlated with CT, CFT, α angle, and MCF determined via either TEM assay; however, corrected plasma concentrations of FXII, FXI, and FVII correlated with a few variables determined via each type of TEM assay.
Spearman rank order correlation coefficients (rs) for CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM and plasma coagulation factor concentrations (white bars) or plasma coagulation factor concentrations corrected for plasma volume (black bars) in whole blood samples obtained from 58 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing citrate anticoagulant to provide plasma (assessment of coagulation factor concentrations). †For a given coagulation factor, a significant (P < 0.05) correlation between TF-TEM–derived variable and corrected plasma factor concentration was detected. Fg = Plasma fibrinogen (mg/dL). Pro = Plasma prothrombin (%). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Spearman rank order correlation coefficients (rs) for CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM and plasma coagulation factor concentrations (white bars) or plasma coagulation factor concentrations corrected for plasma volume (black bars) in whole blood samples obtained from 58 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing citrate anticoagulant to provide plasma (assessment of coagulation factor concentrations). †For a given coagulation factor, a significant (P < 0.05) correlation between TF-TEM–derived variable and corrected plasma factor concentration was detected. Fg = Plasma fibrinogen (mg/dL). Pro = Plasma prothrombin (%). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Spearman rank order correlation coefficients (rs) for CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM and plasma coagulation factor concentrations (white bars) or plasma coagulation factor concentrations corrected for plasma volume (black bars) in whole blood samples obtained from 58 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor (assessments of TF-TEM–derived variables) or collected into tubes containing citrate anticoagulant to provide plasma (assessment of coagulation factor concentrations). †For a given coagulation factor, a significant (P < 0.05) correlation between TF-TEM–derived variable and corrected plasma factor concentration was detected. Fg = Plasma fibrinogen (mg/dL). Pro = Plasma prothrombin (%). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Spearman rank order correlation coefficients (rs) for CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM and plasma coagulation factor concentrations (white bars) or plasma coagulation factor concentrations corrected for plasma volume (black bars) in whole blood samples obtained from 31 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing citrate anticoagulant to provide plasma (assessment of coagulation factor concentrations). See Figures 1 and 6 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Spearman rank order correlation coefficients (rs) for CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM and plasma coagulation factor concentrations (white bars) or plasma coagulation factor concentrations corrected for plasma volume (black bars) in whole blood samples obtained from 31 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing citrate anticoagulant to provide plasma (assessment of coagulation factor concentrations). See Figures 1 and 6 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Spearman rank order correlation coefficients (rs) for CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM and plasma coagulation factor concentrations (white bars) or plasma coagulation factor concentrations corrected for plasma volume (black bars) in whole blood samples obtained from 31 healthy dogs. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid (assessments of EA-TEM–derived variables) or collected into tubes containing citrate anticoagulant to provide plasma (assessment of coagulation factor concentrations). See Figures 1 and 6 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Multivariate model—All test results were evaluated for inclusion in multiple linear regression models to predict TEM variables. Best performance models for CT determined via EA-TEM included all coagulation factor concentrations and Hct; best performance models for CT determined via TF-TEM included plasma concentrations of FVII, FV, and FIX and Hct. Best performance models for CFT determined via TF-TEM or EA-TEM included plasma FV and fibrinogen concentrations, Hct, and platelet concentration. Best performance models for α angle determined via TF-TEM or EA-TEM included plasma FV, FIX, and fibrinogen concentrations, Hct, and platelet concentration. Best performance models for MCF determined via TF-TEM or EA-TEM included plasma fibrinogen concentration, Hct, and platelet concentration. Equations used for predicting TEM variables were generated (Appendix). Multivariate linear regression models and predictive performance for TF-TEM–determined variables (Figure 8) and EA-TEM–determined variables (Figure 9) were summarized.
Scatterplots of measured and predicted values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM in whole blood samples obtained from 21, 57, 22, and 70 healthy dogs, respectively. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor for measurement of TF-TEM–derived variables. Predicted values were derived by use of linear model equations. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of measured and predicted values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM in whole blood samples obtained from 21, 57, 22, and 70 healthy dogs, respectively. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor for measurement of TF-TEM–derived variables. Predicted values were derived by use of linear model equations. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of measured and predicted values of CT (A), CFT (B), α angle (C), and MCF (D) determined via TF-TEM in whole blood samples obtained from 21, 57, 22, and 70 healthy dogs, respectively. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with tissue factor for measurement of TF-TEM–derived variables. Predicted values were derived by use of linear model equations. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of measured and predicted values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM in whole blood samples obtained from 23, 23, 23, and 30 healthy dogs, respectively. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid for measurement of EA-TEM–derived variables. Predicted values were derived by use of linear model equations. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of measured and predicted values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM in whole blood samples obtained from 23, 23, 23, and 30 healthy dogs, respectively. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid for measurement of EA-TEM–derived variables. Predicted values were derived by use of linear model equations. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Scatterplots of measured and predicted values of CT (A), CFT (B), α angle (C), and MCF (D) determined via EA-TEM in whole blood samples obtained from 23, 23, 23, and 30 healthy dogs, respectively. Blood samples were collected into tubes containing citrate anticoagulant, recalcified, and activated with ellagic acid for measurement of EA-TEM–derived variables. Predicted values were derived by use of linear model equations. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.789
Discussion
Consistent results with TEM or TEG requires that coagulation be initiated with a strong activator of coagulation.9 In the present study, coagulation in the whole blood samples was initiated via either the extrinsic pathway (ie, TF-TEM) which uses a tissue factor trigger (similar to coagulation initiated for measurement of plasma PT) or the intrinsic pathway (ie, EA-TEM) which uses a contact activator (similar to coagulation initiated for measurement of plasma aPTT). Direct comparison of the TEM variables determined by use of the 2 activators indicated strong and significant correlation between the CFT, α angle, or MCF values and weaker yet significant correlation for the CT values. The correlations between CFT, α angle, or MCF determined via TF-TEM and EA-TEM was expected because these variables represent the amplification and propagation phases of coagulation that occur on the activated platelet surface and should be independent of the activator used to generate the initial thrombin in the initiation phase.10 In contrast, CT represents the initiation phase of thrombin generation, during which small amounts of thrombin are generated to prime the system for the amplification and propagation phases.10 Coagulation time should be primarily dependent on both the strength of the activating trigger applied and the concentrations of coagulation factors available to generate downstream enzymes.9 The unexpected weak but significant correlation between TF-TEM–determined CT and EA-TEM–determined CT could be a reflection of the clinical normalcy of the blood samples used and the codependency of production of factors used in the 2 pathways. Because the multiple coagulation factors are produced in the liver through similar mechanisms, factor concentrations (for vitamin K–dependent factors in particular) are generally codependent in clinically normal individuals. Given that no specific individual coagulation factor was limited in the blood samples from healthy dogs, TF-TEM–determined CT and EA-TEM–determined CT would both be a function of potentially codependent factor concentrations and, consequently, appear to be correlated. It is less likely that CT values determined via the 2 methods would be as correlative in blood samples from individuals with isolated or disparate factor deficiencies. This concept is supported by results of a TEG study11 of plasma samples from humans, which indicated that isolated coagulation factor deficiencies affect TEG variables in both a factor-dependent and activation-dependent fashion.
Comparison of plasma coagulation factor concentrations to TEM results in the present study indicated that, for blood samples from healthy dogs (containing coagulation factor concentrations primarily within reference intervals), TEM results were not generally highly correlated with individual factor concentrations. This lack of correlation could be a reflection of the fact that in clinically normal dogs, coagulation factor availability is not limiting and therefore not the primary determinant of the initial rate of thrombin generation. Previous studies involving blood samples from dogs with deficiencies of FVII,12 FVIII,13,14 or FIX14 have revealed marked hypocoagulability, as determined via TEM or TEG, in the face of severe factor deficiency
However, there were a few notable exceptions to the lack of apparent correlation between coagulation factor concentrations and TEM results in the present study. Plasma concentration of FVII was highly correlated with CT determined via TF-TEM. Because TF-TEM uses a tissue factor trigger, the initial rate of thrombin generation is dependent on the concentration of tissue factor–FVII complex that forms. When tissue factor is in excess, CT would be expected to be related to the limiting quantity of FVII molecules within the blood sample. Similarly, FIX concentration was highly correlated with CT determined via EA-TEM, although interestingly, other participating factors in the intrinsic pathway (FXII, FVIII, and FXI) were not significantly correlated with CT determined via EA-TEM. The apparent impact of FIX concentration on EA-TEM–determined CT in blood samples from dogs could be explained if FIX concentration were the limiting factor in formation of the intrinsic tenase complex in this species and other intrinsic factors were available in vast excess, which would therefore have less impact on the initial rate of thrombin generation.
Another potential explanation for the lack of consistent correlation between coagulation factor concentrations and TEM results, particularly for CT values, could be the confounding effect of Hct on the mass of plasma proteins included in the TEM reaction. When individual coagulation factor concentrations were corrected for plasma volume in the TEM cup, the correlation between corrected factor concentration and TEM results improved considerably for most factors, resulting in significant associations for most coagulation factors and TEM variables. The improvement in correlation that occurred as a result of correction for plasma volume load in the TEM reaction sample supported the hypotheses that RBC mass is a confounder that artifactually impacts whole blood TEM results.
In the present study, plasma fibrinogen concentration was strongly and highly significantly correlated with CFT, α angle, and MCF, regardless of activator used. This finding is consistent with results of previous studies,11,15 which indicate that fibrinogen availability is a major contributor to values of these TEG or TEM variables. In a study11 of TEG performed on plasma samples from humans, α angle and MA were highly related to fibrinogen concentration within the range of 100 to 300 mg/dL, but R was not affected unless fibrinogen concentration was very low (< 75 mg/dL). The fibrinogen concentration was a main predictive variable of MA even in humans with combined vitamin K–dependent factor deficiencies.15 Maximum clot firmness is directly related to fibrinogen concentration in blood samples from humans with congenital fibrinogen deficiency and has been advocated as a surrogate marker for efficacy of fibrinogen replacement treatment.15
Platelet concentration has also been previously reported to impact TEM or TEG variables measured in blood samples from humans and other animals.16–19 In vitro adjustment of platelet concentration in human platelet-rich plasma samples impacted K, α angle, and particularly MA determined via TEG.17–19 Results of a study16 of human platelet-rich plasma suggested that the critical platelet concentration necessary to cause K and MA to be outside reference ranges was < 66,000 platelets/μL. In vitro adjustment of platelet concentration (by use of platelet depletion) in whole blood samples obtained from healthy humans and those with peripheral artery disease yielded a strong correlation between log platelet concentration and both K and MA.17 Platelet concentration has also been reported to correlate with MA in whole blood samples from parturient women.18 In the healthy dogs of the present study, platelet concentration correlated with CFT, α angle, and MCF, regardless of whether coagulation was initiated via the extrinsic or intrinsic pathway. This correlation was evident despite the fact that most samples contained platelet concentrations within the reference interval. Correlation between platelet concentration and TEM or TEG variables has not been specifically reported for dogs with thrombocytopenia or thrombocytosis, but a categorization of clinically ill dogs by use of G (clot stability; a mathematical transformation of MA representing shear elastic modulus strength) indicated that low, normal (ie, within reference range), and high G values were significantly associated with low, normal, and high platelet concentrations.7 It is reasonable to expect that platelet concentration would not strongly influence the CT (or R) because that value represents the initiation phase of coagulation. In contrast, CFT (or K), α angle, and MCF (or MA) represent the phases of coagulation that occur on the surface of platelets10 and would therefore be influenced by platelet concentration.
Results of the present study additionally indicated that Hct, erythrocyte concentration, and hemoglobin concentration were correlated with CT determined via TF-TEM and with CFT, α angle, and MCF determined via TF-TEM or EA-TEM. This finding is consistent with a few previous reports19–23 of the relationships of erythrocyte variables to results of TEG or TEM performed on blood samples from both humans and other animals. In vitro TEG studies that used autologous plasma or crystalloid dilution of whole blood samples from dogs,p rabbits,19 and humans20,21 have revealed apparent hypercoagulability associated with anemia. In vivo manipulation of RBC mass has also resulted in similar effects on TEM and TEG results.22,23,q Phlebotomy (removing 30% of blood volume followed by saline [0.9% NaCl] solution replacement) in humans was associated with a correlative increase in coagulability detected via TEG.22 In clinically normal dogs, phlebotomy with plasma replacement (resulting in decrease in mean Hct from 52.8% to 47.1%) was associated with a significant increase in coagulability detected via TEM that impacted CT, CFT, α angle, and MCF.q Conversely, polycythemia has been associated with apparent hypocoagulability detected via TEG; an Hct of 85% in a transgenic mouse with excess erythropoietin production was associated with marked hypocoagulability, as measured via TEG.23
In the present study, we hypothesized that the apparent relationship between Hct and coagulability as measured in vitro via TEM might be an artifact of the plasma volume loaded into the TEM cup. When a defined volume of whole blood is used for coagulation testing, RBCs act as a functional diluent for plasma, limiting the plasma volume included in the TEM reaction. Because plasma coagulation proteins are distributed only to the liquid component of the blood, a whole blood sample with a higher Hct is functionally more dilute with regard to plasma proteins than is a sample containing fewer RBCs. Blood samples with a high Hct consequently provide a lower mass of coagulation factors for inclusion in the TEM reaction and vice versa.
The concept that lower Hct may be associated with artifactual hypercoagulability (or alternatively that higher Hct may be associated with artifactual hypocoagulability) detected via TEM or TEG is supported by clinical findings in healthy humans. On the basis of TEG findings, blood samples from women appear to be hypercoagulable, compared with blood samples from men, and women typically have lower Hct.24 However, a similar sex difference was not described for dogs,25 for which circulating RBC mass does not appear to differ between males and females. Increased hypercoagulability also develops with aging in humans, and this change in TEG values is partly attributable to the development of lower RBC mass with age.26 Alternative explanations for the influence of Hct on TEM or TEG results include the possibility that RBC mass truly impacts coagulation in vivo or that there is a potential influence of too much or too little citrate on the plasma component of a blood sample. A true in vivo effect is unlikely because abnormalities of RBC mass in vivo seem to be associated with effects on coagulation opposite to those indicated via TEG or TEM; polycythemia vera is generally associated with a thrombotic tendency,27 whereas anemia is often associated with a bleeding tendency.28 It is also unlikely that citrate concentration in the plasma component of a sample is the cause of the changes detected via TEM because the relationship between RBC mass and TEM results persisted when TEM was performed on canine whole blood samples in the absence of citrate or when citrate concentration was corrected for sample Hct.q
Thromboelastography has revealed apparent hypercoagulability in blood samples from dogs with diseases such as parvoviral enteritis,5 neoplasia,3 and immune-mediated hemolytic anemia.6 In those studies, the reported data were not evaluated for any specific association between RBC mass indices and TEG results, but concurrent anemia is common in inflammatory and neoplastic processes, suggesting that the results indicative of hypercoagulability were obtained from samples containing fewer than normal RBCs. However, many of those diseases have been clinically associated with evidence of true in vivo hypercoagulability with signs of disseminated intravascular coagulation or macrothrombosis, suggesting that hypercoagulability is indeed present in vivo. Additionally, although Greyhounds have a breed-specific higher erythrocyte concentration and their blood samples appear markedly hypocoagulable via TEG,8 clear evidence of a bleeding tendency has been reported for this breed.29
In light of the probable impact of RBC mass on TEM and TEG values, interpretation of TEM and TEG results in the face of abnormal circulating RBC mass may be difficult and could severely limit the application of this technology in populations of ill animals. It is likely that hypercoagulability detected via TEM or TEG in dogs with polycythemia is reflective of true in vivo hypercoagulability and, alternatively, that hypocoagulability detected via TEM or TEG in dogs with anemia is reflective of true in vivo hypocoagulability. However, the most common clinical circumstance where TEM or TEG would be theoretically useful in canine populations would be to identify hypercoagulability in at-risk populations such as dogs with neoplasia, sepsis, or immune-mediated hemolytic anemia, all diseases wherein anemia is common. Evidence of hypercoagulability, as detected via TEM or TEG, should be interpreted with caution in such populations. Further studies comparing TEM or TEG data to alternative indicators of in vivo hyper- or hypocoagulability in anemic (or polycythemic) dogs are warranted.
ABBREVIATIONS
| aPTT | Activated partial thromboplastin time |
| CFT | Clot formation time |
| CT | Coagulation time |
| EA-TEM | Thromboelastometry with ellagic acid |
| FV | Coagulation factor V |
| FVII | Coagulation factor VII |
| FVIII | Coagulation factor VIII |
| FIX | Coagulation factor IX |
| FX | Coagulation factor X |
| FXI | Coagulation factor XI |
| FXII | Coagulation factor XII |
| MA | Maximum amplitude |
| MCF | Maximum clot firmness |
| PT | Prothrombin time |
| TEG | Thromboelastography |
| TEM | Thromboelastometry |
| TF-TEM | Thromboelastometry with tissue factor |
| rs | Spearman correlation coefficient (rho) |
Vaculet blood collection sets, 21 gauge, 3.4 × 12 inches, Excel International, Los Angeles, Calif.
Buffered sodium citrate vacutainer, silicone coated glass, 4.5-mL tube, BD, Franklin Lakes, NJ.
ROTEM, Pentapharm GMbH, Munich, Germany.
star-TEM, Pentapharm GMbH, Munich, Germany.
ex-TEM, Pentapharm GMbH, Munich, Germany.
in-TEM, Pentapharm GMbH, Munich, Germany.
Neoplastine CL+, Diagnostic Stago, Assiernes, France.
Sta-APTT, Diagnostic Stago, Assiernes, France.
Fibri-prest-automate, Diagnostic Stago, Assiernes, France.
Unicalibrator human plasma, Diagnostic Stago, Assiernes, France.
Owren-Koller buffer, Diagnostic Stago, Assiernes, France.
Human factor immunodepleted plasmas, Haematologic Technologies, Essex Junction, Vt.
Start 4 coagulometer, Diagnostic Stago, Assiernes, France.
CELL-DYN 3700 hematology analyzer, with manual leucocyte differential analysis and blood smear evaluation, Abbott Diagnostics, Abbott Park, Ill.
SigmaStat, version 2.03, SPSS Inc, Chicago, Ill.
Vilar P, Hansell J, Westendorf N, et al. Effects of Hct on thromboelastography tracings in dogs (abstr). J Vet Intern Med 2008;22:774.
Smith SA, McMichael MA, Galligan AJ, et al. Effect of in vivo reduction in red cell mass on results on canine whole blood thromboelastometry (abstr). J Thromb Haemost 2009;7(S2):PP-TH-255.
References
1 Ganter MT, Hofer CK. Coagulation monitoring: current techniques and clinical use of viscoelastic point-of-care coagulation devices. Anesth Analg 2008; 106:1366–1375.
2 Shima M, Matsumoto T, Ogiwara K. New assays for monitoring haemophilia treatment. Haemophilia 2008; 14(suppl 3): 83–92.
3 Kristensen AT, Wiinberg B, Jessen LR, et al. Evaluation of human recombinant tissue factor-activated thromboelastography in 49 dogs with neoplasia. J Vet Intern Med 2008; 22:140–147.
4 Wiinberg B, Jensen AL, Johansson PI, et al. Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular coagulation. J Vet Intern Med 2008; 22:357–365.
5 Otto CM, Rieser TM, Brooks MB, et al. Evidence of hypercoagulability in dogs with parvoviral enteritis. J Am Vet Med Assoc 2000; 217:1500–1504.
6 Sinnott VB, Otto CM. Use of thromboelastography in dogs with immune-mediated hemolytic anemia: 39 cases (2000–2008). J Vet Emerg Crit Care 2009; 19:484–488.
7 Wiinberg B, Jensen AL, Rozanski E, et al. Tissue factor activated thromboelastography correlates to clinical signs of bleeding in dogs. Vet J 2009; 179:121–129.
8 Vilar P, Couto CG, Westendorf N, et al. Thromboelastographic tracings in retired racing greyhounds and in non-greyhound dogs. J Vet Intern Med 2008; 22:374–379.
9 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.
10 Smith SA. The cell-based model of coagulation. J Vet Emerg Crit Care 2009; 19:3–10.
11 Nielsen VG, Cohen BM, Cohen E. Effects of coagulation factor deficiency on plasma coagulation kinetics determined via thrombelastography: critical roles of fibrinogen and factors II, VII, X and XII. Acta Anaesthesiol Scand 2005; 49:222–231.
12 Callan MB, Aljamali MN, Margaritis P, et al. A novel missense mutation responsible for factor VII deficiency in research Beagle colonies. J Thromb Haemost 2006; 4:2616–2622.
13 Jiang H, Lillicrap D, Patarroyo-White S, et al. Multiyear therapeutic benefit of AAV serotypes 2, 6, and 8 delivering factor VIII to hemophilia A mice and dogs. Blood 2006; 108:107–115.
14 Margaritis P, Roy E, Aljamali MN, et al. Successful treatment of canine hemophilia by continuous expression of canine FVIIa. Blood 2009; 113:3682–3689.
15 Peyvandi F. Results of an international, multicentre pharmacokinetic trial in congenital fibrinogen deficiency. Thromb Res 2009; 124(suppl 2): S9–S11.
16 Oshita K, Az-ma T, Osawa Y, et al. Quantitative measurement of thromboelastography as a function of platelet count. Anesth Analg 1999; 89:296–299.
17 Bowbrick VA, Mikhailidis DP, Stansby G. Influence of platelet count and activity on thromboelastography parameters. Platelets 2003; 14:219–224.
18 Beilin Y, Arnold I, Hossain S. Evaluation of the platelet function analyzer (PFA-100) vs. the thromboelastogram (TEG) in the parturient. Int J Obstet Anesth 2006; 15:7–12.
19 Nielsen VG, Baird MS. Extreme hemodilution in rabbits: an in vitro and in vivo Thrombelastographic analysis. Anesth Analg 2000; 90:541–545.
20 Ruttmann TG, Lemmens HJ, Malott KA, et al. The haemodilution enhanced onset of coagulation as measured by the thrombelastogram is transient. Eur J Anaesthesiol 2006; 23:574–579.
21 Ruttmann TG, James MF, Viljoen JF. Haemodilution induces a hypercoagulable state. Br J Anaesth 1996; 76:412–414.
22 Ng KF, Lam CC, Chan LC. In vivo effect of haemodilution with saline on coagulation: a randomized controlled trial. Br J Anaesth 2002; 88:475–480.
23 Shibata J, Hasegawa J, Siemens HJ, et al. Hemostasis and coagulation at a hematocrit level of 0.85: functional consequences of erythrocytosis. Blood 2003; 101:4416–4422.
24 Gorton HJ, Warren ER, Simpson NA, et al. Thromboelastography identifies sex-related differences in coagulation. Anesth Analg 2000; 91:1279–1281.
25 Bauer N, Eralp O, Moritz A. Establishment of reference intervals for kaolin-activated thromboelastography in dogs including an assessment of the effects of sex and anticoagulant use. J Vet Diagn Invest 2009; 21:641–648.
26 Ng KF. Changes in thrombelastograph variables associated with aging. Anesth Analg 2004; 99:449–454.
27 Finazzi G, Barbui T. Evidence and expertise in the management of polycythemia vera and essential thrombocythemia. Leukemia 2008; 22:1494–1502.
28 Lier H, Krep H, Schroeder S, et al. Preconditions of hemostasis in trauma: a review. The influence of acidosis, hypocalcemia, anemia, and hypothermia on functional hemostasis in trauma. J Trauma 2008; 65:951–960.
29 Lara-Garcia A, Couto CG, Iazbik MC, et al. Postoperative bleeding in retired racing Greyhounds. J Vet Intern Med 2008; 22:525–533.
Appendix
Equations used in multivariate linear modeling of data for CT, CFT, α angle, and MCF determined via TF-TEM and EA-TEM in whole blood samples obtained from healthy dogs.
| Variable | Equation |
|---|---|
| TF-TEM–derived variable | |
| CT | (1.683×Hct) + (0.243×FVII) − (0.172×FX) − (0.222×FV) − (0.475×FIX) + 21.976 |
| CFT | (4.627×Hct) − (0.167×Plat) + (0.0116×FV) − (0.274×Fg) − 3.397 |
| α angle | (−0.377×Hct) + (0.00607×Plat) + (0.047×Fg) − (0.0129×FIX) + (0.018×FV) + 74.555 |
| MCF | (−0.318×Hct) + (0.0356×Plat) + (0.0452×Fg) + 53.039 |
| EA-TEM−derived variable | |
| CT | (0.0398×Hct) − (0.285×FVII) + (0.782×FX) − (0.196×FV) − (0.169×Pro) − (0.033×Fg) + (0.0895×FXII) + (0.304×FXI) − (0.684×FIX) + (0.0947×FVIII) + 187.872 |
| CFT | (6.427×Hct) − (0.126×Plat) − (0.427×FV) − (0.167×Fg) − 86.93 |
| α angle | (−0.841×Hct) + (0.0244×Plat) + (0.0216•Fg) − (0.0213×FIX) + (0.0666×FV) + 94.423 |
| MCF | (−0.598×Hct) + (0.0443×Plat) + (0.0134×Fg) + 71.894 |
Fg = Plasma fibrinogen concentration (mg/dL). Pro = Plasma prothrombin concentration (%). Plat = Platelet concentration (× 103 platelets/μL).
