| ||||||||||||||
|
|
|||||||||||||

From the *Department of Anesthesiology, Division of Cardiothoracic Anesthesia, Emory University School of Medicine, Atlanta, Georgia; and
Haemostasis and Transfusion Medicine, Munich University Clinic, Munich, Germany.
Address correspondence and reprint requests to Jerrold H. Levy, MD, Professor of Anesthesiology, Department of Anesthesiology, Division of Cardiothoracic Anesthesia, Emory University School of Medicine, 1364 Clifton Rd., NE, Atlanta, GA 30322. Address e-mail to jerrold.levy{at}emoryhealthcare.org.
Abstract
BACKGROUND: Recombinant activated factor VII (rFVIIa) is increasingly used for treating refractory bleeding after cardiac surgery. However, hemostasis also depends on coagulation factors, including fibrinogen, which stabilizes platelet plugs at sites of vascular injury. We compared the hemostatic effects of rFVIIa, fibrinogen, or their combination.
METHODS: Blood samples were obtained from 12 volunteers and from 7 patients after cardiopulmonary bypass (CPB). The in vitro effects of rFVIIa (1.5 µg/mL), fibrinogen (100 mg/dL), and the combination were evaluated under simulated coagulopathy in volunteer plasma using heparin (0.1 U/mL) or tissue plasminogen activator (0.1 µg/mL). Hemostatic interventions were compared using thromboelastometry, which measures clotting time (CT, s), angle of thrombus formation, and maximal clot firmness (MCF, mm). The ThrombinoscopeTM was used to quantitate thrombin generation after addition of fibrinogen and/or rFVIIa.
RESULTS: In heparinized volunteer plasma, rFVIIa shortened CT (1st and 3rd quartiles) from 663 (522–736) to 435 (397–531) s, but it did not affect MCF. Fibrinogen increased MCF from 26.0 (24.4–26.7) to 30.5 (26.3–31.5) mm without affecting CT. The combination of rFVIIa and fibrinogen in heparinized samples was most effective in improving CT to 359 (324–522) s and MCF to 29 (27.8–31.0) mm. In tissue plasminogen activator-treated volunteer plasma, fibrinolysis increased by more than 45% by the addition of rFVIIa. After CPB, both CT and MCF were most improved with coadministration of rFVIIa and fibrinogen. Thrombinoscope evaluation demonstrated that rFVIIa decreased the lag time and increased peak thrombin generation, whereas fibrinogen had no effect.
CONCLUSION: The onset of fibrin formation and thrombin generation were shortened after rFVIIa addition, but fibrin clot strength was only increased after fibrinogen supplementation. In vitro clot formation was most improved by using both rFVIIa and fibrinogen in whole blood after CPB.
In surgical patients after prolonged cardiopulmonary bypass (CPB), hemostatic imbalance often results from consumptive loss of coagulation factors, hemodilution, hypothermia, residual anticoagulation, and fibrinolysis.1–4 Under these circumstances, most procoagulant plasma factors are reduced by 40%–50% from baseline, and platelets are often reduced in number and function. Recombinant factor VII (rFVIIa) is increasingly used for refractory bleeding after CPB5–9 beyond its Food and Drug Administration approved indication for the treatment of hemophilia A in patients with Factor VIII inhibiting antibodies. However, it has been suggested the hemostatic efficacy of rFVIIa is limited in the presence of dilutional hypofibrinogenemia,10 and enhanced with fibrinogen supplementation.11 We hypothesized that coadministration of rFVIIa and fibrinogen would be more effective for improving thrombus formation than either respective agent alone. Thus, we investigated the in vitro coagulation effects of rFVIIa and fibrinogen concentrate using viscoelastic measurements of fibrin formation and fluorogenic measurements of thrombin generation.
METHODS
The study procedures were approved by the Emory University human subject research review board and were performed after receiving written informed consent. Blood was obtained from 12 healthy volunteers and seven patients undergoing combined valve and coronary artery bypass grafting surgery requiring at least 2 h of CPB. The volunteers had no history of aspirin ingestion or use of other medications that might interfere with platelet function over the preceding 2 wk.
Thrombelastometry (ROTEM®)
Platelet-poor plasma (PPP) was obtained by centrifugation of the volunteer blood samples at 3000g for 15 min. Coagulopathy in these samples was simulated in vitro by the addition of heparin (Elkins-Sinn, Inc., Cherry Hill, NJ) or tissue plasminogen activator (tPA) (Alteplase®, Genentech, South San Francisco, CA). The final concentration of heparin and tPA in PPP was 0.1 U/mL and 0.1 µg/mL, respectively. The respective model was used to simulate: (1) residual or rebound heparin effect (0.05–0.1 U/mL) that may be observed after protamine administration, and (2) increased fibrinolysis associated with CPB procedures.1 In the third experiment, blood from the surgical patients was obtained 15 min after protamine administration after CPB. Whole blood samples were collected into glass Vacutainer® tubes (Becton Dickinson and Company, Franklin Lakes, NJ) containing 3.2% citrate (1:9 in volume). Four-channel ROTEM® (Pentapharm, Munich, Germany) was performed for both volunteer plasma samples and whole blood surgery patients studies using 340 µL of PPP or whole blood with kaolin activation and 10 µL of 0.4M CaCl2. PPP or whole blood samples were pretreated with either saline (control) or a hemostatic intervention. The hemostatic intervention consisted of fibrinogen (Hemocomplettan® P, CSL Behring, Marburg, Germany), rFVIIa (NovoSeven®, NovoNordisk, Princeton, NJ), or the combination of both drugs in 10.0 µL of volume. The final concentration of added fibrinogen and rFVIIa in the sample was 100 mg/dL and 1.5 µg/mL, respectively.
The following ROTEM variables were obtained: clotting time (CT; s), angle ([
]°) of thrombus formation, and maximal clot firmness (MCF; mm) (Fig. 1A). In addition, the percent decrease in clot firmness after MCF was calculated to estimate fibrinolysis (Lysis index; LI %) in tPA-treated plasma using the following formula: LI % = 100 x (MCF – CF30)/MCF, where CF30 is the clot firmness at 30 min after MCF was achieved. Aprotinin (50 kallikrein inhibitory units per milliliter) was added in some samples to neutralize tPA-induced activation of plasmin.
|
Thrombin Generation Assay (ThrombinoscopeTM)
Blood samples were collected from the seven CPB patients and blood counts including platelet were performed using coulter counter (Beckman-Coulter, Fullerton, CA). Next, blood samples were centrifuged at 200g for 10 min to obtain platelet-rich plasma (PRP) and the platelet count was again determined. The supernatant (PRP) was transferred to a clean polypropylene tube and the remaining blood sample was recentrifuged for 20 min at 2000g to obtain PPP. Because platelets become highly concentrated in PRP during the centrifugation, PRP was diluted with PPP until the platelet count of the diluted PRP sample was equal to the platelet count in the original whole blood sample. For example, when PRP contains 400 x 103 platelets per mm3 and whole blood sample contains 200 x 103 platelets per mm3, 0.4 mL of PRP was diluted with 0.4 mL of PPP (approximately zero platelets) to obtain 0.8 mL of diluted PRP with approximately 200 x 103 platelets per mm3 verified by platelet count. Diluted PRP was pretreated with either saline (control) or 1 of 3 interventions: fibrinogen (final concentration, 100 mg/dL), rFVIIa (final concentration, 1.5 µg/mL), or the combination of rFVIIa and fibrinogen. The effects of rFVIIa or fibrinogen on thrombin generation were evaluated in PRP using a thrombin generation assay (ThrombinoscopeTM Synapse, BV, Maastricht, The Netherlands).
The concentration of generated thrombin is estimated from the changes in fluorescence intensity when thrombin cleaves the fluorogenic substrate, Z-Gly-Gly-Arg-AMC (benzyloxycarbonyl-Gly-Gly-Arg- 7-amido-4-methylcoumarin, Bachem Bioscience, King of Prussia, PA) as previously described (Fig. 1B).10 For thrombin generation measurements, 80 µL of PRP followed by 20 µL of an activator (Biodis, Signes, France) was added to microtiter plate wells, the plate was incubated for 2–3 min at 37°C, and then 20 µL of the substrate buffer was added. A thrombin calibrator with known thrombin-like activity was monitored in parallel sample wells to allow for calculation of generated thrombin in nM. The progress of the reaction was continuously monitored for 70 min at 37°C with a fluorescence reader (Fluoroscan Ascent, Thermo Labsystems, Franklin, MA) equipped with a 390 nm excitation filter and a 460 nm emission filter. The lag time and peak level of thrombin generation were obtained from the Thrombinoscope (Fig. 1B).
Statistics
Based on previous studies with thrombelastometry and thrombin generation assay, the sample size of 6 was needed to detect 20% change from the controls in MCF variable and peak thrombin level with a β
0.8 and an
< 0.05.11,12 The variables of thromboelastometry (CT,
°, and MCF), and thrombin generation assay (lag time and peak thrombin level) were compared among different hemostatic interventions against the control (no intervention) using the Kruskal– Wallis H-test, followed by the Mann–Whitney U-test with Bonferroni's correction using the SPSS 15.0 (SPSS Inc., Chicago, IL). Data are expressed as the median (25%–75% quartiles) or % changes. P
0.05 was considered significant. A value of P < 0.05 was considered significant.
RESULTS
Thrombelastometry
Thromboelastometric results are listed in Table 1 and Figures 2 and 3. In volunteer plasma samples treated with heparin (0.1 U/mL), the addition of rFVIIa significantly decreased the time to onset of clot formation, or CT, and it increased the
° of clot formation when compared with the control sample (Table 1). The addition of fibrinogen did not affect CT or the
° of clot formation but it did increase MCF. When both agents were added, all three variables were significantly improved over the control sample.
|
|
|
The platelet count and fibrinogen level (mean ± sd) from the blood obtained from surgical patients after CPB were 70 ± 45 x 103 mm–3 and 154 ± 51 mg/dL, respectively, (CPB duration, 169 ± 57 min). The thromboelastometry tracings in whole blood samples with in vitro addition of rFVIIa or fibrinogen after CPB are shown in Figure 3. The addition of rFVIIa increased CT but not MCF, whereas the addition of fibrinogen induced increases in the
° of clot formation and MCF (Table 1). The combination of rFVIIa and fibrinogen resulted in increases in CT,
°, and MCF compared with control CPB samples also obtained after CPB. Because all the patients received aprotinin during CPB, the addition of tPA did not cause fibrinolysis (data not shown).
Thrombin Generation in PRP After CPB
In PRP obtained from surgical patients after CPB, the addition of rFVIIa, or the combination of rFVIIa and fibrinogen, shortened the lag time of thrombin generation by 60.3% and 63.1%, respectively (Fig. 4). There was a nonsignificant increase in lag time (13.6%, P = 0.1 versus control) when fibrinogen was added to the PRP samples compared with the controls (Fig. 4). Peak thrombin levels were increased by 60.6%, 6.8%, and 83.1% relative to the control when rFVIIa, fibrinogen, or both were added respectively (P < 0.01 versus control for rFVIIa, and rFVIIa plus fibrinogen) (Fig. 4).
|
DISCUSSION
In this study, we found that FVIIa or fibrinogen differentially affects in vitro thrombus formation in the presence of low concentration heparin (0.1 U/mL), tPA (0.1 µg/mL), or CPB-induced coagulopathy. More rapid thrombus formation was demonstrated by shorter CT and a smaller
° of thrombus formation assessed with thromboelastometry after rFVIIa addition, whereas fibrinogen improved clot strength (MCF) in heparinized and post-CPB samples. The combination of rFVIIa and fibrinogen had additive effects on CT, thromboelastometry
°, and MCF (Table 1, Fig. 3) than either agent alone. In the presence of tPA in blood from volunteers, neither rFVIIa nor fibrinogen improved clot stability based on the lysis index (Table 1; Fig. 2). Clot stability, as measured by MCF, was restored only after the addition of the antifibrinolytic drug aprotinin.
Although data from thromboelastometry are more reflective of hemostasis due to low-shear conditions (<0.1 s–1), our findings shed light on hemostatic mechanisms associated with the administration of rFVIIa and fibrinogen. The faster onset of thrombus formation (indicated by lower CT), but the lack of an increase in clot strength (indicated by MCF) or peak thrombin level with the addition of rFVIIa to blood samples from patients after CPB, suggests that its hemostatic effects are due to more rapid initiation of thrombin generation rather than increased clot strength. Rapid thrombin generation may be particularly important to achieve hemostasis in the presence of active bleeding in injured vasculature because key proteases (e.g., FXa and thrombin) have a short half-life (15 s to 1 min),13 and fluid transport mechanisms are slow.14 In high shear conditions, platelets promote arterial hemostasis by forming primary hemostatic plugs and presenting negatively charged phospholipid surfaces to support efficient thrombin generation.15 Lower platelet count and reduced aggregation after CPB16 results in limited phospholipid availability at the injury site, hence slower thrombin generation. The end-point of the thrombin generation assay used in this study is thrombin-mediated cleavage of a synthetic substrate (Z-GGR-AMC), whereas thromboelastometry monitors thrombin-mediated conversion of fibrinogen to fibrin. Nevertheless, the results from thromboelastometry in this study corroborate the fluorogenic measurements of thrombin generation. Our present data demonstrate that rFVIIa, but not fibrinogen, decreases the lag time, indicating an increased rate of thrombin generation in thrombocytopenic post-CPB blood (Fig. 4). The peak thrombin level was improved most by combined rFVIIa and fibrinogen, followed by rFVIIa only (Fig. 4), implications that are important to consider when treating bleeding.
Previous in vitro data suggest that rFVIIa compensates for thrombocytopenia by enhancing both local platelet accumulation,17 and thrombin-mediated fibrin formation.18 However, clot formation is also dependent on fibrinogen for clot strength. Therefore, restoring fibrinogen levels before administering rFVIIa for life-threatening bleeding should be considered because the hemostatic efficacy of rFVIIa ultimately depends on increases on thrombin-mediated clot formation. This is supported by our data, and by other reports that with dilutional hypofibrinogenemia, the effect of rFVIIa may be limited.10 Further, fibrinogen supplementation (250 mg/kg) has been shown to confer hemostatic effects in a porcine model of bleeding after traumatic liver laceration.11
The lack of an apparent antifibrinolytic effect of rFVIIa was unexpected. Prior studies suggest that rFVIIa increases the generation of thrombin-activated fibrinolysis inhibitor (TAFI), making fibrin clot less susceptible to plasmin.19 Other data suggest, although, that TAFI has minimal antifibrinolytic effects in hemophilic plasma containing rFVIIa.20 The activation of TAFI is a late process during blood coagulation because thrombin preferentially binds to fibrinogen and factor XIII before it binds to TAFI unless plasma thrombin concentration reaches a high level (>150 nM).21 Because TAFI cleaves the binding site (on the fibrin surface) of tPA and plasminogen, rather than directly antagonizing plasmin,22 it is plausible that rapid fibrin formation with rFVIIa might paradoxically increase tPA–plasminogen interaction on fibrin surfaces, resulting in the early clot lysis by plasmin. In agreement with our data, Nielsen et al. have shown that thrombus formation and breakdown via tissue factor activation is more rapid than via the intrinsic pathway (celite) activation in the presence of tPA.23 Neutralization of plasminogen activator inhibitor-1 by faster formation of Xa and thrombin may also contribute to a rapid fibrinolysis.24 The concentration of tPA that we used was 0.1 µg/mL (approximately 1.7 nM), which is well below the reported plasma levels (approximately 150 nM) during tPA therapy.25,26 Our results, therefore, suggest that the addition of an antifibrinolytic agent may be necessary in the profibrinolytic condition to achieve maximal clot stabilization.
Although platelet and factor transfusions and surgical hemostasis are therapies for postoperative bleeding, rFVIIa, fibrinogen, or their combinations are increasingly being reported for treating refractory bleeding. However, transfusions are not without risks for adverse events, including an increasing recognition of transfusion related acute lung injury, especially after factor and platelet transfusions.27–29 Our data show that rFVIIa and fibrinogen induce specific changes in thromboelastometry variables (Fig. 4), and thus appropriate therapies can be selected.30–32
In conclusion, we have demonstrated that the combination of rFVIIa and fibrinogen improve the onset and stability of thrombus formation. These data suggest that the treatment of bleeding in surgical patients after CPB with rFVIIa might be optimized by first normalizing fibrinogen levels.
Footnotes
Accepted for publication November 20, 2007.
Supported by the Department of Anesthesiology, Emory University School of Medicine, Atlanta, Georgia.
Dr. Jerrold H. Levy, Section Editor for Hemostasis and Transfusion Medicine, was recused from all editorial decisions related to this manuscript.
REFERENCES
This article has been cited by other articles:
![]() |
T. Lang, K. Johanning, H. Metzler, S. Piepenbrock, C. Solomon, N. Rahe-Meyer, and K. A. Tanaka The Effects of Fibrinogen Levels on Thromboelastometric Variables in the Presence of Thrombocytopenia Anesth. Analg., March 1, 2009; 108(3): 751 - 758. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Sniecinski, F. Szlam, E. P. Chen, S. O. Bader, J. H. Levy, and K. A. Tanaka Antithrombin Deficiency Increases Thrombin Activity After Prolonged Cardiopulmonary Bypass Anesth. Analg., March 1, 2008; 106(3): 713 - 718. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Szlam, T. Taketomi, C. A. Sheppard, C. L. Kempton, J. H. Levy, and K. A. Tanaka Antithrombin Affects Hemostatic Response to Recombinant Activated Factor VII in Factor VIII Deficient Plasma Anesth. Analg., March 1, 2008; 106(3): 719 - 724. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|