In vitro heparinization of canine whole blood with low molecular weight heparin (dalteparin) significantly and dose-dependently prolongs heparinase-modified tissue factor-activated thromboelastography parameters and prothrombinase-induced clotting time
Lisbeth R. Jessen,
Department of Small Animal Clinical Sciences, Faculty of Life Sciences, University of Copenhagen, Copenhagen, Denmark
Correspondence Lisbeth R. Jessen, Department of Small Animal Clinical Sciences, Faculty of Life Sciences, University of Copenhagen, Dyrlaegevej 16, DK 1870 Frederiksberg C, Denmark E-mail: email@example.com
Background: Low-molecular-weight heparin (LMWH) is being used increasingly in veterinary medicine for both treatment and prophylaxis of thromboembolic disease, but no predictable patient-side method exists to monitor its effect.
Objectives: The aim of this study was to evaluate thromboelastography (TEG) and prothombinase-induced clotting time (PiCT) assays for detecting hemostatic alterations following in vitro heparinization of canine whole blood with dalteparin (Fragmin).
Methods: Citrated whole-blood samples were collected from 7 clinically healthy dogs. Dalteparin was added at concentrations of 0, 0.156, 0.625, 1.25, and 2.5 U/mL of whole blood. TEG was performed using heparinase cups with tissue factor (TF, 1:50,000) and kaolin as activators. Reaction time (R), clotting time (K), angle (α), and maximum amplitude (MA) were recorded. PiCT and anti-FXa activity were measured in plasma.
Results: With TF, increasing concentrations of dalteparin significantly prolonged R and K and significantly decreased α and MA. K, α, and MA ratios were significantly different from baseline at all dalteparin concentrations and R was significantly different from baseline at concentrations of 0.625, 1.25, and 2.5 U/mL. With kaolin, only R was significantly different from baseline at dalteparin concentrations of 0.625 and 2.5 U/mL. PiCT detected dalteparin concentrations ≤ 0.625 U/mL, with a good linear correlation (r2=.96, P<.0001).
Conclusion: These results suggest that TF-activated TEG and PiCT assays should be further evaluated as promising new methods for evaluating the effect of LMWH, using doses in the recommended clinical range and prospective clinical studies.
Low-molecular-weight heparin (LMWH) is used routinely in the treatment and prophylaxis of thromboembolic diseases in humans and the efficacy of the treatment is well documented.1–3 However, little is known about the efficacy, safety, and monitoring of LMWH in the treatment of spontaneous thromboembolic disease in dogs. The pharmacokinetics of Dalteparin (Fragmin), a 4–6 kDa LMWH, have been studied in healthy dogs after IV and SC administration.4 The antithrombotic effects of dalteparin and other LMWHs have also been reported in experimental studies using canine models of total hip replacement, deep venous thrombosis, and experimentally induced disseminated intravascular coagulation.5–7
The plasma concentration of anti-factor FXa correlates with the dose of LMWH and an anti-FXa assay is considered the gold standard for monitoring LMWH therapy. However, several studies in humans have shown a poor correlation between anti-FXa plasma concentration and antithrombotic efficacy or risk of bleeding complications.8–10 It appears that mechanisms other than FXa are involved in the hemostatic actions of LMWH, posing limitations to its use as the sole assay for monitoring therapy with LMWH.
The use of recombinant human tissue factor (TF)–activated thromboelastography (TEG) has been validated recently in dogs.11 TEG evaluates the combined effect of all intravascular components of the hemostatic system, including cells and proteins, and could potentially help to optimize, predict, and monitor the effect of treatment with LMWH.12 In humans, heparinase-modified TEG has been used to monitor hemostatic function in the presence of heparin and to provide rational transfusion guidelines during cardiac surgery and liver transplantation.13–15 Standard TEG has also proven to be a useful assay to monitor LMWH therapy during hemodialysis and thrombophilic complications associated with pregnancy.16
Prothrombinase-induced clotting time (PiCT) is a novel plasma assay for the determination of anticoagulant activity through FXa and/or FIIa inhibition. The assay is based on clot activation mediated by Russell viper venom, a specific activator of FV, which forms the prothrombinase complex with FXa, phospholipids, and calcium ions. Clinical studies are currently lacking, but in vitro studies in humans demonstrate that PiCT measures the activity of both direct and indirect thrombin inhibitors, including LMWH, in a linear fashion, suggesting that the PiCT assay is suitable for monitoring LMWH therapy in humans.17
The aim of this study was to investigate the effect of in vitro heparinization of canine whole blood with dalteparin, on heparinase modified, TF, and kaolin-activated TEG parameters and on PiCT.
Materials and Methods
This study was approved by the Small Animal Ethics and Administrative Committee at the Department of Small Animal Clinical Sciences, Faculty of Life Sciences, University of Copenhagen.
Blood samples were collected from 7 clinically healthy staff-owned dogs, 5 males and 2 females, with a mean (±SD) age of 5.9±1.7 years (range 1.9–9.5 years), at the Small Animal Veterinary Teaching Hospital, University of Copenhagen. The breeds included Dachshund (2), Labrador Retriever (1), Kooikerhondje (1), and mixed breeds (3). The body weight (BW) of the dogs ranged from 6 to 37 kg (mean±SD, 21±13.9 kg). All dogs were used for the entire study, except 1 dog, which was not tested at a dalteparin concentration of 1.25 U/mL whole blood. The dogs were considered clinically healthy based on an unremarkable physical examination and the results of a CBC, routine biochemistry profile, routine coagulation profile (prothrombin time [PT], partial thromboplastin time [aPTT], antithrombin [AT], fibrinogen, and D-dimer), and urinalysis. D-Dimer concentration was not measured in 2 dogs because of minor lipemia.
Blood collection and in vitro heparinization of canine whole blood
The dogs were fasted for 12 hours before sampling. Blood samples were collected in the morning for TF-activated TEG analyses and again in the afternoon for kaolin-activated TEG analyses. Because it was possible to run only 4 simultaneous TEG analyses, the study was divided into 2 parts. Blood samples for TEG analyses at dalteparin concentrations of 0, 0.156, 0.625, and 2.5 U/mL of whole blood were collected and run together. Blood samples for TEG analyses at a dalteparin concentration of 1.25 U/mL were collected and run on a different day 7 months later. Physical examination, CBC, biochemistry profile, coagulation profile, and urinalysis were repeated before testing the 1.25 U/mL concentration. Blood was collected by jugular venipuncture, using minimum stasis and a 21-gauge butterfly needle, and placed into 1 serum tube (5 mL), 5 citrated plastic tubes (3 mL), and 1 EDTA tube (2 mL), in that order (Vacuette, Greiner Bio-One International AG, Kremsmuenster, Austria). From each dog, 22 mL of blood was obtained in the morning and 12 mL was obtained in the afternoon of the same day. For TEG analyses at the dalteparin concentration of 1.25 U/mL whole blood, 16 mL of blood was obtained in the morning and 9 mL of blood was obtained in the afternoon.
After sampling, the 3-mL citrate tubes were carefully inverted 3 times to ensure mixing of 3.2% trisodium citrate and blood, in a 1:9 ratio. One randomly selected citrate tube was used for the coagulation profile. Serum and citrate tubes for biochemistry and coagulation profiles were centrifuged at 4000g for 120 seconds, and serum and plasma were collected within 30 minutes of sampling. The CBC, biochemistry, and coagulation profiles were all run within 1 hour of sampling.
The remaining citrate tubes were stored in an upright position at room temperature (∼21 °C) for 30 minutes for subsequent mixing with dalteparin (Fragmin, Pfizer ApS, Ballerup, Denmark) before TEG analysis. After 30 minutes of storage, the 3-mL citrate tubes were spiked with dalteparin to final concentrations of 0, 0.156, 0.625, 1.25, and 2.5 U/mL of whole blood, respectively. The corresponding median (range) plasma concentration in the samples was calculated based on the HCT as 0.3 (0.27–0.35), 1.2 (1.10–1.39), 2.38 (2.19–2.72), and 4.81 (4.34–5.55) U/mL of plasma, respectively. The concentrations of 0.156, 0.625, 1.25, and 2.5 U dalteparin/mL of whole blood corresponded to calculated IV injection doses of 12.5, 50, 100, and 200 U/kg BW.
After addition of dalteparin, each tube was gently inverted 3 times to ensure adequate mixing and a 340-μL sample from each citrate tube was used for TEG analysis. The remaining blood from each citrate tube was centrifuged at 4000g for 120 seconds, and plasma was stored at −80 °C for PiCT and anti-FXa activity.
Canine citrated whole-blood samples were activated with a solution of recombinant human TF (Innovin, Dade Behring, Marburg, Germany) at a final dilution of 1:50,000 (0.119 pmol/L) and analyzed as described previously.16 Kaolin activation was performed according to the protocol recommended by the manufacturer (Haemoscope Corporation, Niles, IL, USA). TEG analyses were performed using a TEG 5000 Hemostasis Analyzer (Haemoscope) with continuous transfer of data from the analyzer to a computer. All TEG analyses were performed using prewarmed heparinase cups (Haemoscope) and run for 90 minutes. The following parameters were recorded: reaction time (R), clotting time (K), angle (α), and maximum amplitude (MA).
Prothrombinase-induced clotting time
The Pefakit PiCT assay (Pentapharm Ltd., Basel, Switzerland) was performed according to the protocol recommended by the manufacturer. Two milliliters of deionized water was added to the activator reagent containing FXa, Russell viper venom, phospholipids, 4-2-hydroxyethyl-1-piperazine-ethanesulfonic acid, and mannitol, and stored for 15 minutes at room temperature. After storage, a 50-μL plasma sample was mixed with 50 μL of mannitol and the sample was incubated for 180 second at 37 °C. Finally, each sample was mixed with 50 μL of an activator reagent containing 25 mM CaCl2. The clotting time was measured using an automated coagulometric analyzer (ACL 9000, Instrumentation Laboratories, Barcelona, Spain). All measurements were run in duplicate.
Anti-factor Xa activity
Anti-FXa activity was measured at Novo Nordisk A/S, Maaloev, Denmark. The samples were shipped on dry ice and the transit time was 1 hour. Anti-FXa activity was measured with a chromogenic substrate (HemoIL-Heparin, Instrumentation Laboratories) on an ACL 300R analyzer (Instrumentation Laboratories) in all samples, except those containing 1.25 U/mL, which were stored improperly. The assay was calibrated with normal human plasma calibrators containing known amounts of LMWH (Tinzaparin, Innohep, LEO Pharma Nordic, Ballerup, Denmark).
Routine hematology and coagulation tests
Activated aPTT, PT, fibrinogen concentration, and AT activity were measured using standard assays and reagents (Instrumentation Laboratories) on the ACL 9000 analyzer. A plasma pool from 5 healthy dogs was used as a reference. D-Dimer concentration was measured with an immunofiltration assay (D-Dimer Single Test) on the NycoCard READER II (Medinor A/S, Brøndby, Denmark).18
Platelet count and HCT were measured on an ADVIA 120 automated hematology system (Siemens Health Care Diagnostics Inc, Deerfield, IL, USA)
Nonparametric statistical analyses were performed. All ratios were analyzed as paired data when comparing measurements at different dalteparin concentrations and were reported as median and range. Results from TF- and kaolin-activated TEG were analyzed separately. Results for TEG, anti- FXa, and PiCT assays were converted into ratios to compare the relation of a value at a defined dalteparin concentration (fold-increase) with the baseline value of the same individual. The relationship of PiCT and anti-FXa results with dalteparin concentration was examined by linear regression. The intra-assay analytical coefficient of variation (CV) of the PiCT assay was calculated by a pooled variance estimate of the duplicate determinations. Friedman analysis was applied to anti-FXa, PiCT, and TEG results to test the hypothesis that no significant changes were observed across dalteparin concentrations. If the hypothesis was rejected, groups were compared with the baseline by use of a Wilcoxon's signed rank test for paired observations, to assess the sensitivity of the parameter for detecting the effect of dalteparin. Statistical significance was set at P<.05. Computer software programs (GraphPad Prism v. 4.01, GraphPad Software, La Jolla, CA, USA; SAS v. 9.1, SAS, Inc., Cary, NC, USA) were used for statistical analysis.
Results for TF-activated α from 1 dog and all TF-activated TEG results from another dog were excluded for samples containing 2.5 U dalteparin/mL due to flattening of the TEG tracings such that values could not be obtained. TF-activated TEG results from 1 dog were excluded for samples containing 0.625 U/mL due to technical error. The results for kaolin-activated TEG in samples with 1.25 U/mL were excluded due to technical problems with the batch of kaolin. Results of routine coagulation tests, platelet count, and HCT were within reference limits for our laboratory on both days of the study in all dogs (Table 1).
Table 1. Results of coagulation and hematology tests for 7 clinically healthy dogs.
Aggregates observed on blood smear.
ND indicates not done; Fib, fibrinogen; AT, antithrombin.
Days 1 and 2 refer to the days on which the in vitro studies were conducted.
When TF was used as an activator, increasing concentrations of dalteparin significantly prolonged R and K and significantly decreased α and MA. When kaolin was used as an activator, only R was significantly different from baseline at high dalteparin concentrations. Median baseline values and ratios for TF- and kaolin-activated TEG results were tabulated (Table 2). The results for TF- and kaolin-activated TEG parameters at different dalteparin concentrations are shown in Figure 1 and typical TEG tracings are shown in Figure 2.
Table 2. Changes in tissue factor (TF)- and kaolin-activated thromboelastography values in whole blood samples from clinically healthy dogs following in vitro heparinization with dalteparin at concentrations of 0.156, 0.625, 1.25, and 2.5 U/mL of whole blood.
Median baseline PiCT was 24.5 second (23.9–28.9 second) (Table 3). Increasing concentrations of dalteparin increased PiCT values significantly and dose-dependently (r2=.96, P<.0001, Friedman analysis). At or above a dalteparin concentration of 1.25 U/mL, clotting was not achieved within the analysis time of 220 second. The intra-assay CV was 2.3% at 0 U/mL dalteparin (PiCT, 27.9±SD second); 5.9% at 0.156 U/mL (PiCT, 41.3±SD); and 7.5% at 0.625 U/mL (PiCT, 98.64±SD second). The median PiCT baseline value and ratios were tabulated (Table 3). PiCT values at different dalteparin concentrations are shown in Figure 3.
Table 3. Changes in prothrombinase induced clotting time (PiCT) and anti-FXa activity in whole blood samples from clinically healthy dogs following in vitro heparinization with dalteparin at concentrations of 0.156, 0.625, and 2.5 U/mL of whole blood.
The median baseline value for anti-Xa activity was 0.07 U/mL of plasma (0.03–0.25 U/mL) (Table 3). Increasing concentrations of dalteparin increased anti-FXa activity significantly and dose-dependently. A close correlation was found between added and measured anti-FXa activities up to a dalteparin concentration of 2.5 U/mL (r2=.96) (Figure 4).
The results of this study show that spiking citrated canine whole blood with increasing concentrations of dalteparin affected the results of all TF-activated, heparinase-modified TEG parameters significantly and dose-dependently. A dose-dependent response was also found for R values in kaolin-activated, heparinase-modified TEG. These results suggest that TF-activated, heparinase-modified TEG could be used as a patient-side method to monitor the hemostatic effect of heparin in canine patients receiving LMWH therapy.
The results of this study prove the concept that heparinase-modified, TF-activated TEG can detect increasing concentrations of dalteparin in a dose-dependent fashion. However, according to the current consensus on the clinically relevant dosage of LMWH in dogs, most concentrations of dalteparin applied in this study are supratherapeutical. The recommended dose of dalteparin for dogs in clinical practice is 100–150 U/kg SC every 8 hours. In a study by Mischke et al,19 peak anti-FXa activity 2 hours after a single injection of 100 U/kg to dogs in vivo was 0.43 U/mL of plasma. With repeated SC injections of 150 U/kg dalteparin every 8 hours, plasma anti-FXa activity was 0.77 U/mL of plasma and 0.82 U/mL at 2 hours after the third and sixth injections.20 The supratherapeutical concentrations pose a limitation in the present study with regard to the direct clinical applicability of the results, and further evaluations of the heparinase-modified, TF-activated TEG assay with doses in the recommended clinical range, followed by prospective clinical studies, are required to assess the performance and usefulness of heparinase-modified, TF-activated TEG in a clinical setting. Dalteparin concentrations of 0.625 and 1.25 U/mL of whole blood (which correspond to calculated IV injections of 50 and 100 U/kg dalteparin) were chosen as the 2 central concentrations because the effects of these IV dosages were described previously in dogs.19 However, in that study, dalteparin was injected in vivo and anti-FXa activities in plasma after 2 minutes were 0.88±0.18 and 1.86±0.17 U/mL. Thus, our study did not account for distribution or metabolism of dalteparin in vivo, and our calculated median plasma concentrations were 1.2 (1.1–1.39) and 2.38 (2.19–2.72) anti–FXa U/mL plasma at dalteparin concentrations of 0.625 and 1.25 U/mL of whole blood. The concentrations of 0.156 and 2.5 U/mL (corresponding to calculated median plasma levels of 0.3 and 4.81 U/mL of plasma) were chosen to test the lower and upper limits of the assays. At the highest dalteparin concentration, 1 dog had a completely flat TEG tracing (no values obtained), indicating that the upper limit of the assay was ∼2.5 U/mL (corresponding to a calculated median plasma level of 4.81 U/mL).
TEG should be performed in the presence of heparinase, due to the high sensitivity of TEG to even small amounts of heparin, which inhibits measurable clot formation. In a pilot study, we found that readable TEG tracings could not be obtained at a dalteparin concentration of ≥0.625 U/mL of whole blood (corresponding to a calculated median plasma level of 1.2 U/mL). This was in accordance with an in vitro study using human samples in which LMWH (enoxaparin) concentrations of 0.5 U/mL inhibited TEG analysis markedly to completely.21 In a recent study, Gerotziafas et al22 detected dose-dependent alterations in TF-activated TEG following in vitro heparinization of human blood with enoxaparin and fondaparinux. Differences in the type of LMWH and the concentration of TF (1:8000 as compared with 1:50,000 in the present study) may account for the fact that they were able to obtain diagnostic TEG tracings at concentrations up to 1 U/mL for enoxaparin and 1 μg/mL for fondaparinux.
Kaolin-activated, heparinase-modified TEG is currently used in human medicine to reverse the effect of heparin or LMWH, so as to diagnose underlying nonheparin-dependent abnormalities in clot formation and breakdown. In most studies of heparinase reversal, either no activator or kaolin is used to trigger TEG.13,14,21 In our study, when using kaolin as an activator, there was a trend toward complete reversal of the dalteparin effect, with statistically significant dose-dependent alterations only for the R value at dalteparin concentrations above 0.625 U/mL (corresponding to a calculated median plasma of 1.2 U/mL), thus confirming the findings of human studies of heparinase reversal using kaolin activation of TEG.14
In contrast, we found that when using diluted TF as an activator, heparinase did not completely reverse the anticoagulant effects of dalteparin on TEG tracings and the sensitivity was increased compared with kaolin, as significant alterations were observed from the lowest dalteparin concentration of 0.156 U/mL (corresponding to a calculated median plasma level of 0.3 U/mL) for all parameters except R. The results of this study indicate that when coagulation is triggered by diluted TF, some of the anticoagulant mechanisms of dalteparin are reversed by heparinase, thus allowing for readable TEG tracings, while other anticoagulant mechanisms seem to remain unaltered by the presence of heparinase, allowing for monitoring of the alterations in all 4 TEG parameters with increasing concentrations of dalteparin. The exact mechanisms responsible for this are not known, but one possible explanation could involve mechanisms related to the function of the TF pathway inhibitor (TFPI). Both unfractionated heparin and LMWH are known to increase the release of TFPI from endothelial cells. This effect could not be evaluated in the present in vitro study. However, LMWH also enhances the effect of TFPI already present in blood and platelets. Our results and those of Gerotziafas et al22 suggest that TF-activated TEG may help provide additional information on the hemostatic effects of LMWH not explained by anti-FXa activity.
Another advantage of TEG is its ability to clearly distinguish hypercoagulable from normal dogs.11 In this study, increasing concentrations of dalteparin pushed the TEG tracing to the right and narrowed the height of the tracing in a dose-dependent manner. This unique feature of the TF-activated TEG assay could potentially be utilized in targeting therapy of the hypercoagulable patient toward a normalization of the patient's TEG tracings and thereby tailor the LMWH dose in vitro to meet the requirement of the individual patient.
In this study, PiCT values were obtained at dalteparin concentrations ranging from 0 to 0.625 U/mL whole blood, corresponding to calculated median plasma levels of 0 to 1.20 U/mL plasma, which was similar to the measurable range in human plasma stated by the manufacturer of the PiCT assay. These results were also similar to those obtained previously in vitro in our laboratory, where PiCT values were obtained in the dalteparin range from 0 to 1.6 U/mL plasma in dogs.23 The excellent linear relationship between dalteparin concentration and PiCT in our study confirmed the results of in vitro studies with human samples.17
PiCT measures the combined effects on anti-FXa and FIIa of unfractionated heparin, LMWH, direct thrombin inhibitors, and other anticoagulant drugs with similar modes of action. The potential benefit of the PiCT assay as compared with the anti-FXa assay is its ability to also measure anti-FIIa activity. LMWH is thought to have very low anti-FIIa activity because the majority of LMWH fragments are too short to form the tertiary complex with AT and thrombin that is required to inactivate thrombin. However, anti-IIa activity varies among different LMWH and may not be negligible for all of them.24 A potential drawback of the PiCT assay is its inability to assess high concentrations of LMWH. The test also requires expensive equipment and special sample handling and in this respect does not provide an advantage over the currently used anti-FXa assay. Further studies are required to establish whether the PiCT assay will provide additional information on the anticoagulant and antithrombotic efficacy of LMWH therapy in vivo as compared with the gold standard chromogenic anti-FXa assay.
In this study, we observed an excellent linear relationship between dalteparin concentrations and anti-FXa levels up to a dalteparin concentration of 0.625 U/mL (corresponding to a calculated median plasma level of 1.2 U/mL). At a dalteparin concentration of 2.5 U/mL (corresponding to a calculated median plasma level of 4.81 U/mL), we obtained a median anti-FXa activity of 3.21 U/mL of plasma, which was lower than expected. In addition, we measured an unexpectedly high baseline anti-FXa activity of 0.25 U/mL in 1 dog. It has been reported that some anti-Xa assays are linear only up to 1 U/mL of plasma.25 A plausible explanation for the discrepancy between expected and measured values at the dalteparin concentration of 2.5 U/mL may be a lack of linearity at very high dalteparin concentrations. The standard curve used for calibration was obtained using human plasma and tinzaparin, which poses a limitation to the assay.
Studies in humans have demonstrated a significant correlation between anti-FXa activity and the dose of LMWH.9,26 However, several clinical studies have also demonstrated that anti-FXa activity does not accurately predict the anticoagulant or the antithrombotic effect in an individual patient, which, in combination with the special laboratory handling required, represents limitation to the use of this assay for monitoring LMWH therapy in dogs.27,28
In conclusion, TF-activated TEG using heparinase TEG cups and the PiCT assay both detect hemostatic alterations in a significant and dose-dependent manner following in vitro addition of increasing concentrations of LMWH (dalteparin) to canine whole blood. These results suggest that both TF-activated TEG and PiCT should be further evaluated as promising new methods for evaluating the effect of LMWH.
The authors thank G. Wagner and N. Errebo, Department of Small Animal Clinical Sciences, Faculty of Life Sciences, University of Copenhagen, Denmark, for expert technical assistance and H. Bach and B. Binow Sørensen, Novo Nordisk A/S, Maaloev, Denmark, for performing the anti-FXa assay.