- Email: [email protected]
N) for control of bleeding after kidney trauma in a rabbit dilutional coagulopathy model
Thrombosis Research 125 (2010) 272–277
Contents lists available at ScienceDirect
Thrombosis Research j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t h r o m r e s
Regular Article
Prothrombin complex concentrate (Beriplex P/N) for control of bleeding after kidney trauma in a rabbit dilutional coagulopathy model☆ Ingo Pragst, Franz Kaspereit, Bärbel Dörr, Gerhard Dickneite ⁎ Department of Pharmacology and Toxicology, CSL Behring GmbH, Marburg, Germany
a r t i c l e
i n f o
Article history: Received 25 August 2009 Received in revised form 19 October 2009 Accepted 21 October 2009 Available online 14 November 2009 Keywords: Prothrombin complex concentrate Recombinant factor VIIa Trauma Blood coagulation disorders Hemorrhage Models, animal
a b s t r a c t Introduction: Fluid resuscitation after trauma often results in dilutional coagulopathy that may hinder control of bleeding and, once initial hemostasis has been secured, heighten risk of perioperative bleeding when further surgery is required. Since multiple coagulation factor deficiencies typically accompany fluid resuscitation, prothrombin complex concentrate (PCC) containing factors II, VII, IX and X may potentially offer greater hemostatic efficacy than coagulation factor monotherapy. Materials and methods: Anesthetized normothermic rabbits were hemodiluted 50-60% by phased blood withdrawal and infusion of hydroxyethyl starch and erythrocytes. The animals were randomly assigned to receive saline placebo, 25 IU·kg- 1 PCC (Beriplex P/N) or 180 μg·kg- 1 activated recombinant factor VII (rFVIIa; NovoSeven). Immediately thereafter, bleeding was precipitated by a standardized kidney incision. Results: PCC accelerated hemostasis compared both with saline and rFVIIa (p=0.002 for both comparisons). The median times to hemostasis in the PCC, saline and rFVIIa groups were 12, 19 and 28 min, respectively. PCC reduced blood loss by a median of 43 mL with a 95% confidence interval (CI) of 8.0-67.5 mL vs. saline and 82 mL (CI, 35.0110.0 mL) vs. rFVIIa. PCC augmented peak thrombin generation by a median of 104.1 nM (CI, 78.3-142.3 nM) compared with saline and 105.8 nM (CI, 70.7-139.5 nM ) relative to rFVIIa. At the respective 180 μg·kg- 1 and 25 IU·kg- 1 doses tested, rFVIIa displayed thrombogenicity in the Wessler stasis model, while PCC did not. Conclusions: In an animal model of dilutional coagulopathy and kidney trauma, PCC accelerated hemostasis and diminished blood loss compared with rFVIIa monotherapy. © 2009 Elsevier Ltd. All rights reserved.
Uncontrolled hemorrhage is a major challenge in trauma patients, accounting for as much as 25-30% of mortality [1]. Controlling bleeding in trauma patients is frequently complicated by major deficits in coagulation factors and platelets from blood loss, as well as dilutional coagulopathy arising when hypovolemia is remedied by infusing fluids devoid of coagulation factors, namely crystalloids, colloids and packed red blood cells. Moreover, metabolic derangements associated with major trauma such as acidosis and hypothermia can further compromise coagulation. Coagulopathic bleeding is particularly difficult to control [2]. A pressing need exists for effective agents to control refractory coagulopathic bleeding. Recombinant activated factor VII (rFVIIa) has been investigated as an adjunctive hemostatic agent for this purpose. While dramatic responses to rFVIIa in trauma patients with major Abbreviations: CI, 95% confidence interval; FII, factor II; FVII, factor VII; FIX, factor IX; FX, factor X; HES, hydroxyethyl starch; IQR, interquartile range; PCC, prothrombin complex concentrate; PT, prothrombin time; rFVIIa, recombinant factor VIIa. ☆ Presented in part at the 53rd Annual Meeting of the Society for Thrombosis and Hemostasis Research, Vienna, Austria, February 4-7, 2009. ⁎ Corresponding author. Senior Director of Preclinical Research and Development, CSL Behring GmbH, Emil von Behring-Straße 76, D-35041 Marburg, Germany. Tel.: +49 6421 39 2306; fax: +49 6421 39 5310. E-mail address: [email protected] (G. Dickneite). 0049-3848/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2009.10.011
hemorrhage have been documented in many case reports and small case series, results of controlled clinical trials have been variable [3]. In phase II clinical trials, rFVIIa significantly reduced the need for red blood cell transfusion in patients with blunt but not penetrating trauma, and in neither group was an improvement in survival attained [4,5]. One inherent limitation in replacing a single coagulation factor is the common presence of deficiencies in multiple factors after trauma and fluid resuscitation. Prothrombin complex concentrate (PCC) may afford a promising alternative to rFVIIa monotherapy for uncontrolled coagulopathic bleeding. The widely investigated PCC Beriplex P/N contains coagulation factors II (FII), VII (FVII), IX (FIX) and X (FX), as well as the anticoagulant proteins C and S. This PCC has proven highly effective for emergency reversal of coumarin oral anticoagulant therapy, an indication also involving multiple coagulation factor deficiencies [6–8]. In a rodent model of sustained coumarin anticoagulation, the hemostatic efficacy of Beriplex P/N was superior to that of rFVIIa [9]. Recently, in a porcine model of dilutional coagulopathy and spleen trauma, PCC was shown to accelerate hemostasis and increase thrombin generation compared with rFVIIa [10]. One limitation of that study was the absence of data on thrombogenic potential. The most wellestablished animal model for assessing thrombogenicity is in the rabbit [11]. So that both hemostatic efficacy and thrombogenicity could be
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
investigated simultaneously in the same animal species, a rabbit model of dilutional coagulopathy and kidney trauma has been developed. That model is described in this report, and data on the comparative hemostatic effects of PCC and rFVIIa and the thrombogenic potential of those agents are presented. Materials and methods Animals Female CHB rabbits 3-4 mo old weighing 2.8-4.0 kg (Bauer, Neuental, Germany) were housed one per cage in wire-steel cages at 21-23 °C and 50% relative humidity under a 12 h/12 h light-darkness cycle. The animals were provided tap water ad libitum and fed rabbit pellets (Deukanin®, Deutsche Tiernahrung Cremer GmbH & Co. KG, Düsseldorf, Germany). All rabbits received care in compliance with the European Convention on Animal Care, and the study was approved by the organizational Ethics Committee. Endpoints The primary study endpoints were time to hemostasis and blood loss as observed up to 30 min following a standardized kidney incision injury (Fig. 1). Time to hemostasis was defined as the interval from the kidney incision until cessation of observable bleeding or oozing. Blood loss was the volume of blood collected from the incision site by suction. Both time to hemostasis and blood loss were assessed by observers blinded to the group assignments of the rabbits. Prothrombin time (PT) was a secondary endpoint. Anesthesia A combination of 5 mg·kg- 1 i.v. ketamine (Ketavet®, Pharmacia GmbH, Erlangen, Germany) and 0.5 mg·kg- 1 i.v. xylazine 2% (Rompun®, Bayer Vital GmbH, Leverkusen, Germany) was used for induction of anesthesia. The animals were then intubated and placed on a ventilator
273
(Heyer Access, Heyer Medical AG, Bad Ems, Germany). Inhaled anesthesia was maintained with isoflurane (Isofluran CP®, CP-Pharma Handelsgesellschaft mbH, Burgdorf, Germany) at a concentration of 2.4% depending upon depth of anesthesia. A 14-gauge catheter was introduced into the carotid artery for continuous blood pressure monitoring and blood sampling and a 20-gauge catheter into the external jugular vein for hemodilution and test fluid administration. Maintenance i.v. fluid requirement was satisfied by infusion of 4 mL·kg- 1·h- 1 Ringer's lactate. Body temperature was monitored by rectal thermometry. Following a 20 min stabilization period, baseline measurements were made of hemodynamic, coagulation and hematological parameters. Hemodilution Animals were subjected to hemodilution in phases by withdrawal of 30 mL·kg- 1 blood and infusion of 30 mL·kg- 1 hydroxyethyl starch (HES) 200/0.5 (Infukoll 6%, Schwarz Pharma AG, Mannheim, Germany) prewarmed to 37 °C (Fig. 1). That procedure was repeated at 45 min. At 30 min, during the interval between the two cycles of blood withdrawal and HES infusion, the animals received 15 mL·kg- 1 salvaged erythrocytes, prepared from withdrawn whole rabbit blood by centrifugation for 10 min at 800×g, washing in normal saline and resuspension in Ringer's lactate. Treatment Animals were randomly allocated to receive i.v. infusions of isotonic saline, 25 IU·kg- 1 PCC (Beriplex® P/N, CSL Behring GmbH, Marburg, Germany) or 180 µg·kg- 1 rFVIIa (NovoSeven®, Novo Nordisk A/S, Bagsværd, Denmark) immediately prior to kidney incision injury (Fig. 1). The mean concentrations of FII, FVII, FIX and FX in Beriplex P/N are 30, 15, 30 and 40 IU·mL- 1, respectively [12]. A randomized negative control group did not undergo hemodilution. Experimental groups consisted of 5-7 rabbits each. Blood samples for determination of PT, coagulation factor and fibrinogen concentrations and platelet counts were
Fig. 1. Study procedures for hemodilution, treatment, experimental kidney trauma and assessment of hemostatic effect. Abbreviations: HES, hydroxyethyl starch; PCC, prothrombin complex concentrate; rFVIIa, activated recombinant factor VII.
274
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
collected at baseline, the conclusion of hemodilution and just before kidney incision. Dosing was selected to afford adequate opportunity for showing hemostatic efficacy while avoiding excessive risk of thromboembolic events. In porcine trauma model studies, 25-35 IU·kg- 1 PCC accelerated hemostasis [10,13]. While those studies were not specifically designed to assess thrombogenicity, there was no overt clinical evidence of thromboembolic events. Preliminary measurements of thrombin generation in the rabbit suggested that 25 IU·kg- 1 would be adequate to demonstrate hemostatic efficacy. Bleeding reduction has been observed after rFVIIa doses as low as 100-150 µg·kg- 1 in rabbit trauma model studies, and patients with blunt trauma receiving 100-200 µg·kg- 1 rFVIIa in randomized clinical trials required less red blood cell transfusion without increased incidence of thromboembolic complications [4]. In the rabbit, 100-300 µg·kg- 1 rFVIIa displayed substantially less thrombogenicity than 1000 µg·kg- 1 [14].
Statistical analysis
Kidney injury
Hemodilution
At 60 min after commencement of hemodilution, a standardized renal injury was inflicted in the form of a 15 mm long and 5 mm deep scalpel incision at the lateral kidney pole (Fig. 1). The 30 min observation period for blood loss and time to hemostasis began immediately after the incision.
At the completion of phased hemodilution, the pooled median observed magnitude of hemodilution for the saline, PCC and rFVIIa groups was 56.1% (IQR, 54.8-57.6%; n = 19), based upon measured decline in hematocrit from baseline. Hemodilution resulted in median decreases of 0.13 in pH, 1.5 °C in rectal temperature, 1.00 g·L- 1 in fibrinogen concentration and 189 × 109 L- 1 in platelet count (Table 1). PT was prolonged by a median of 8.3 s in response to hemodilution (Table 1). Subsequent to hemodilution, concentrations of FII, FVII, FIX and FX were reduced to approximately 20-30% of baseline (Table 2).
Laboratory assays PT was measured with a Schnitger & Gross coagulometer using the Thromborel reagent (Dade Behring, Marburg, Germany). Plasma for coagulation factor and fibrinogen determinations was prepared from citrated blood samples and stored at –80 °C until assay. FII, FIX and FX were measured in coagulation factor-deficient plasma (Dade Behring) with a Behring Coagulation Timer (BCT®, Dade Behring). FVII was quantified with a human-specific enzyme-linked immunosorbent assay (Biozol GmbH, Eching, Germany). Fibrinogen was assayed by the Clauss method using the Behring Coagulation System (BCS®) and reagents (Dade Behring). Thrombin generation assay As described by Hemker et al. [15], thrombin generation assay (TGA) of diluted platelet-poor plasma samples was performed by calibrated automated thrombinography (CAT, Thrombinoscope B.V., Maastricht, the Netherlands) using the Thrombinoscope software version 3.0.0.29 to calculate molar thrombin content. Concentrations of recombinant relipidated tissue factor and phospholipids were 5 pM and 4 μM, respectively. Calculated assay parameters were peak thrombin generation and lag time until observable thrombin generation. Wessler model The thrombogenicity of PCC and rFVIIa was assessed in female New Zealand White rabbits using the Wessler stasis model [11]. Methodological details have been recently described elsewhere [16]. Groups of 310 anesthetized rabbits received 50, 100, 200, 300, 400 or 500 IU·kg- 1 PCC or 10, 50, 100, 180 or 300 µg·kg- 1 rFVIIa. At 10 min thereafter, an isolated jugular vein segment was allowed to fill with blood by ligation. After a 30 min period of stasis, the vein segment was excised and dissected in a Petri dish filled with sodium citrate solution. A thrombosis score was assigned on a scale of 0 to 3, with 0 denoting no clotting, 1 clotting without organized thrombus formation, 2 one or more discrete thrombi and 3 an occlusive thrombus. As an objective measure of thrombus formation, the wet weight of thrombi formed during stasis was also determined.
Descriptive statistics consisted of the median and range or interquartile range (IQR). Median differences were computed with exact 95% confidence intervals (CI) using Hodges-Lehmann estimation, and corresponding p values were calculated by exact Wilcoxon test. Time to hemostasis was analyzed by the Kaplan-Meier product limit method, and between-group differences in this endpoint were evaluated by exact logrank test. The dose-response relationships of PCC and rFVIIa with thrombosis score and thrombus weight were characterized by linear regression. R version 2.7.2 (The R Foundation for Statistical Computing, Vienna, Austria) and StatXact 7.0 (Cytel Software Corp., Cambridge, Massachusetts, USA) statistical software were used for all analyses. Results
Coagulation factors Saline infusion after hemodilution showed no effect on concentrations of FII, FVII, FIX or FX (Table 2). After administration of PCC, which contains FII, FVII, FIX and FX, the median concentrations of those four coagulation factors increased respectively to 89, 75, 38 and 181% of baseline levels. As in the pig [17], measured FIX levels in the rabbit are 2-3 fold those in humans. Consequently, the administered PCC dose produced a comparatively small FIX increase relative to the high baseline level. Conversely, measured FX level in rabbits is only approximately 40% of the corresponding human value, so that a comparatively large FX concentration rise was attained after PCC administration. Hemostasis Relative both to saline and rFVIIa, PCC significantly shortened time to hemostasis after experimental kidney trauma (Fig. 2). The median time to hemostasis was 12 min in PCC-treated rabbits compared with 19 and 28 min for the groups receiving saline and rFVIIa, respectively. In all three treatment groups time to hemostasis was retarded vs. the negative control group not subjected to hemodilution. Cessation of kidney bleeding was observed within the 30 min observation period Table 1 Effects of hemodilution on physiological and hematologic parameters. Parameter
pH Temperature (° C) Fibrinogen (g·L- 1)† Platelets (× 109 L- 1) PT (s)
Median (IQR), n = 19 Baseline
Hemodilution
Change
7.51 (7.46 to 7.54) 39.3 (39.1 to 39.6) 1.70 (1.57 to 1.75) 292 (258 to 329) 11.0 (10.9 to 11.9)
7.36 (7.33 to 7.39) 37.8 (37.5 to 38.1) 0.68 (0.59 to 0.85) 107 (89 to 118) 19.5 (18.4 to 21.3)
-0.13 (-0.16 to -0.09) - 1.5 (- 1.7 to - 1.4) -1.00 (-1.04 to -0.90) - 189 (-216 to - 166) 8.3 (7.5 to 9.8)
Median pH, rectal temperature, fibrinogen concentration, platelet count, and PT at baseline and after hemodilution and corresponding median changes. Abbreviations: IQR, interquartile range; PT, prothrombin time. † n=17.
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
275
Table 2 Coagulation factors after hemodilution and treatment with saline or PCC. Factor
II VII IX X
Group
Saline PCC Saline PCC Saline PCC Saline PCC
n
5 6 5 6 6 6 5 6
Median Percent of Baseline (Range) Hemodilution
Treatment
22.0 23.2 34.6 31.6 26.7 28.4 21.1 21.6
24.1 (23.3 to 26.7) 88.8 (71.7 to 94.3) 37.2 (33.0 to 41.4) 75.3 (41.9 to 88.4) 28.0 (18.1 to 29.7) 37.7 (29.4 to 43.9) 24.4 (22.0 to 27.2) 181 (106 to 242)
(18.0 (20.7 (32.3 (28.4 (17.7 (22.2 (11.0 (20.1
to to to to to to to to
24.6) 26.3) 38.5) 34.6) 28.1) 30.3) 23.2) 22.9)
Median concentrations of coagulation factors II, VII, IX and X as a percent of baseline after hemodilution and treatment with saline or PCC. Abbreviation: PCC, prothrombin complex concentrate.
in all animals of the PCC, saline and negative control groups. In the rFVIIa group, bleeding persisted after 30 min in 3 of 6 rabbits. Results with respect to blood loss volume generally paralleled those for time to hemostasis (Fig. 3). Thus, blood loss in the PCC-treated rabbits was significantly lower by a median of 43 mL vs. the saline group and by 82 mL vs. rFVIIa recipients. All three treatment groups lost more blood than did non-hemodiluted control animals. While blood loss was higher by a median of 38 mL after rFVIIa than saline infusion, this difference was not statistically significant. Prothrombin time PT was prolonged by a median of 7.2 s (CI, 4.7 to 9.7 s) among saline-treated rabbits and by a median of 5.2 s (CI, 2.5 to 9.8 s) after PCC infusion, as compared with the non-hemodiluted control group. rFVIIa fully normalized PT. Thrombin generation As evidenced by the saline-treated group, hemodilution reduced peak thrombin generation by a median of 56.3 nM (CI, 37.3 to 77.7 nM) compared with that in non-hemodiluted animals. rFVIIa showed no effect on peak thrombin generation vs. the saline group. In contrast, PCC augmented peak thrombin generation by a median of 104.1nM (CI, 78.3 to 142.3 nM) and 105.8 nM (CI, 70.7 to 139.5 nM) compared with saline- and rFVIIa-treated rabbits, respectively.
Fig. 3. Blood loss after hemodilution and infusion of saline, PCC or rFVIIa. A negative control group did not undergo hemodilution. Median values are represented by horizontal lines. Abbreviations: CI, 95% confidence interval; PCC, prothrombin complex concentrate; rFVIIa, recombinant factor VIIa.
Treatment effects on thrombin generation lag time differed from those on peak thrombin. Thus, lag time in the saline and PCC groups did not significantly differ from that in non-hemodiluted controls, whereas rFVIIa markedly shortened lag time by a median of 0.89 min (CI, 0.73 to 1.07 min) vs. saline and a median of 1.06 min (CI, 0.45 to 1.12 min) vs. PCC. Thrombogenicity Over dose ranges of 50-500 IU·kg- 1 PCC and 10-300 μg·kg- 1 rFVIIa, continuous increases were observed in thrombosis score (Fig. 4(a) and (b), respectively) and thrombus weight (Fig. 4(c) and (d), respectively). In all rabbits receiving 50 IU·kg- 1 PCC, i.e. twice the dose exhibiting hemostatic efficacy after dilutional coagulopathy, both thrombosis score and thrombus weight were zero. The 180 μg·kg- 1 rFVIIa dose evaluated for hemostatic efficacy was associated with evidence of thrombogenicity. At that dose, the mean thrombosis score was 1.8 (CI, 1.5-2.1) and the mean thrombus weight 41 mg (CI, 27-55 mg), as estimated by linear regression analysis. Discussion
Fig. 2. Time to hemostasis following experimental kidney trauma in animals not subjected to hemodilution or treated following hemodilution with saline, PCC or rFVIIa. Abbreviations: PCC, prothrombin complex concentrate; rFVIIa, recombinant factor VIIa.
In this rabbit model of dilutional coagulopathy and traumatic kidney bleeding, PCC significantly decreased both time to hemostasis and blood loss compared with rFVIIa or saline. The accompanying PCC-mediated increase in peak thrombin generation provides a mechanistic basis for the observed hemostatic effects of PCC. This is the first study to evaluate the hemostatic efficacy of PCC in a rabbit trauma model. The present findings are consistent with those in a porcine model of dilutional coagulopathy and traumatic splenic bleeding, in which PCC infusion just prior to splenic incision augmented thrombin generation and accelerated hemostasis vs. rFVIIa or saline placebo [10]. In the same porcine model, PCC also shortened the time to hemostasis and diminished blood loss after either experimental bone or spleen trauma as compared with fresh frozen plasma (FFP) [13]. These three animals studies suggest that PCC could potentially prove to be a valuable adjunctive therapy for coagulopathic bleeding in trauma. Currently, FFP is the standard choice to correct dilutional coagulopathy following major trauma [18]. But its limitations, such as relatively slow correction of coagulopathy, have led to the investigation of other more convenient and rapidly effective agents such as rFVIIa. PCCs were originally developed for the treatment of hemophilia and have been
276
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
Fig. 4. (a) PCC and (b) rFVIIa dose-response relationships with thrombosis score and (c) and (d) corresponding relationships with thrombus weight. Thick lines show best-fit linear functions and dashed curves the CI of the regressions. Abbreviations: CI, 95% confidence interval; PCC, prothrombin complex concentrate; rFVIIa, recombinant factor VIIa.
used for more than 30 years to provide prompt and effective reversal of coumarin oral anticoagulant therapy [6–8]. PCC shares with rFVIIa the advantage of suitability for rapid infusion and undelayed normalization of coagulation parameters. Animal models simulating clinical traumatic injuries have primarily been developed in pigs because of similarities to humans in coagulation function and internal organ size [19]. However, the rabbit provides a useful alternative due to its intermediate size, ease of handling and relatively large circulating blood volume for its size [20]. Importantly, a well-characterized model for assessing thrombogenic potential exists in the rabbit [11]. Lastly, rabbit thromboplastin is a standard reagent in coagulation tests and may display greater activity when interacting with human FVII than swine thromboplastin [19]. While fully normalizing PT in the present study, rFVIIa at a dose of 180 µg·kg- 1 exhibited poor hemostatic efficacy in normothermic rabbits with dilutional coagulopathy and thrombocytopenia resulting from blood withdrawal and HES infusion. In previous rabbit studies, rFVIIa has exhibited hemostatic efficacy, but essential experimental parameters have differed from those in the present study [21–24]. In a study of noncoagulopathic rabbits with hepato-splenic injury, 150 µg·kg- 1 rFVIIa shortened bleeding time under both normo- and hypothermic conditions but reduced blood loss only in hypothermic animals [23]. Among rabbits with dilutional coagulopathy due to HES replacement, 100 µg·kg- 1 rFVIIa decreased ear immersion bleeding time and microvascular bleeding compared with placebo [22]. The
endpoint of two other studies was nail cuticle bleeding time, and high rFVIIa doses were administered [21,24]. In rabbits with irradiationinduced thrombocytopenia, 2000 µg·kg- 1 rFVIIa diminished bleeding time and blood loss [21]. A 5000 µg·kg- 1 rFVIIa dose was effective in reducing bleeding time of rabbits subjected to calculated hemodilution of 50% with HES 200/0.5, the same type as in the present study, but not with a higher molecular weight, more highly substituted HES solution [24]. In addition to normalizing PT, rFVIIa shortened the lag time before substantial thrombin generation. The hemostatic process is composed of an initiation phase and a propagation phase. Approximately, 5% of the total thrombin generation occurs during the initiation phase and 95% in the propagation phase [25]. PT is a measure of activity in the initiation phase [25]. The PT and lag time observations in the present study suggest that rFVIIa effectively accelerated the initiation phase. However, in light of its failure to increase peak thrombin generation, rFVIIa appeared in this model to be comparatively ineffective in sustaining the propagation phase and, accordingly, in stopping bleeding. Thrombin generation by rFVIIa was measured using platelet-poor plasma in this study, as well as in several previous studies [26–28]. Measurements in platelet-rich plasma might more faithfully simulate in vivo thrombin generation by rFVIIa, since the mechanism of rFVIIa action involves binding to activated platelets. Nevertheless, the greater observed peak thrombin generation by PCC than rFVIIa in the present study was consistent with the relative acceleration of hemostasis and reduction in blood loss after administration of PCC compared with rFVIIa. Some patients do not respond to high-dose rFVIIa [29]. Several variables such as fibrinogen and platelet levels, acid-base balance and temperature appear to influence the response to rFVIIa. Multidisciplinary task force guidelines from Israel and recommendations for the use of rFVIIa in uncontrolled bleeding call for blood component therapy to ensure fibrinogen level >0.5 g·L- 1 and platelet count > 50× 109 L- 1, correction of acidosis (pH > 7.1) and restoration of normothermia as much as feasible before rFVIIa therapy is considered [30]. While clinical investigations of PCC in trauma remain to be conducted, PCC is not currently known to be subject to the same influences as rFVIIa. In this study PCC showed hemostatic efficacy despite profound depletion of fibrinogen and platelets and a modest decline in pH. In the pig, PCC also proved to be an effective hemostatic agent during hypothermia [10,13]. Like other procoagulant agents, PCC and rFVIIa hold the potential for thrombogenicity. At the 25 IU·kg- 1 PCC and 180 µg·kg- 1 rFVIIa doses tested for hemostatic efficacy in the present study, rFVIIa displayed thrombogenicity in the Wessler model, while PCC did not. Both 25 IU·kg- 1 and 180 µg·kg- 1 are standard doses of PCC and rFVIIa, respectively, in clinical practice. In a previous study employing the Wessler model [14], rFVIIa exhibited thrombogenicity at all evaluated doses, namely, 100, 300 and 1000 µg·kg- 1, during the same 30 min stasis period as in the present study. The numbers and weights of thrombi observed after rFVIIa treatment in that study were comparable to those after 50-100 IU·kg- 1 doses of factor VIII inhibitor bypass activity (FEIBA). The present findings suggest that PCC may offer an attractive safety profile in trauma. Nevertheless, further investigations on the thrombogenic potential of PCC will be needed. The present rabbit model results are in conformity with existing clinical data from pharmacovigilance programs. In one such study, the estimated incidence of thrombotic events after PCC administration was 1.0 per 105 infusions (CI, 0.1-3.6 per 105 infusions) [31]. The corresponding incidence after rFVIIa administration in another pharmacovigilance study was 24.6 per 105 infusions (CI, 19.1-31.2 per 105 infusions) [32]. It should be recognized, however, that pharmacovigilance studies encompass all clinical uses rather than use specifically in trauma. Simultaneous replenishment of multiple coagulation factors with PCC in trauma embodies an attractive therapeutic concept, since trauma patients usually develop multiple factor deficiencies. This study adds to preclinical evidence that PCC can effectively control bleeding after a variety of hemorrhagic insults and under a range of conditions. These
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
encouraging findings suggest that further investigation of PCC in trauma is warranted. Conflict of interest statement The authors are employees of the study sponsor, CSL Behring. Acknowledgments Funding for this investigation was furnished by CSL Behring GmbH. The sponsor is the employer of the authors and as such played an active role in the design of the study, the collection, analysis and interpretation of data, the writing of the manuscript and the decision to submit the manuscript for publication. References [1] Pfeifer R, Tarkin IS, Rocos B, Pape HC. Patterns of mortality and causes of death in polytrauma patients—has anything changed? Injury 2009;40:907–11. [2] Vincent JL, Rossaint R, Riou B, Ozier Y, Zideman D, Spahn DR. Recommendations on the use of recombinant activated factor VII as an adjunctive treatment for massive bleeding—a European perspective. Crit Care 2006;10:R120. [3] Duchesne JC, Mathew KA, Marr AB, Pinsky MR, Barbeau JM, McSwain NE. Current evidence based guidelines for factor VIIa use in trauma: the good, the bad, and the ugly. Am Surg 2008;74:1159–65. [4] Boffard KD, Riou B, Warren B, Choong PI, Rizoli S, Rossaint R, et al. Recombinant factor VIIa as adjunctive therapy for bleeding control in severely injured trauma patients: two parallel randomized, placebo-controlled, double-blind clinical trials. J Trauma 2005;59:8–15. [5] Rizoli SB, Boffard KD, Riou B, Warren B, Iau P, Kluger Y, et al. Recombinant activated factor VII as an adjunctive therapy for bleeding control in severe trauma patients with coagulopathy: subgroup analysis from two randomized trials. Crit Care 2006;10:R178. [6] Evans G, Luddington R, Baglin T. Beriplex P/N reverses severe warfarin-induced overanticoagulation immediately and completely in patients presenting with major bleeding. Br J Haematol 2001;115:998–1001. [7] Preston FE, Laidlaw ST, Sampson B, Kitchen S. Rapid reversal of oral anticoagulation with warfarin by a prothrombin complex concentrate (Beriplex): efficacy and safety in 42 patients. Br J Haematol 2002;116:619–24. [8] Pabinger I, Brenner B, Kalina U, Knaub S, Nagy A, Ostermann H. Prothrombin complex concentrate (Beriplex® P/N) for emergency anticoagulation reversal: A prospective multinational clinical trial. J Thromb Haemost 2008;6:622–31. [9] Dickneite G. Prothrombin complex concentrate versus recombinant factor VIIa for reversal of coumarin anticoagulation. Thromb Res 2007;119:643–51. [10] Dickneite G, Dörr B, Kaspereit F, Tanaka KA. Prothrombin complex concentrate versus recombinant factor VIIa for reversal of hemodilutional coagulopathy in a porcine trauma model. J Trauma 2009;67. doi:10.1097/TA.0b013e3181b06364. [11] Wessler S, Reimer SM, Sheps MC. Biologic assay of a thrombosis-inducing activity in human serum. J Appl Physiol 1959;14:943–6. [12] Ostermann H, Haertel S, Knaub S, Kalina U, Jung K, Pabinger I. Pharmacokinetics of Beriplex P/N prothrombin complex concentrate in healthy volunteers. Thromb Haemost 2007;98:790–7.
277
[13] Dickneite G, Pragst I. Prothrombin complex concentrate vs fresh frozen plasma for reversal of dilutional coagulopathy in a porcine trauma model. Br J Anaesth 2009;102: 345–54. [14] Diness V, Bregengaard C, Erhardtsen E, Hedner U. Recombinant human factor VIIa (rFVIIa) in a rabbit stasis model. Thromb Res 1992;67:233–41. [15] Hemker HC, Giesen P. Al Dieri R, Regnault V, de Smedt E, Wagenvoord R, Lecompte T, Béguin S. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb 2003;33:4–15. [16] Dickneite G, Pragst I, Joch C, Bergman GE. Animal model and clinical evidence indicating low thrombogenic potential of fibrinogen concentrate (Haemocomplettan P). Blood Coagul Fibrinolysis. doi:10.1097/MBC.0b013e32832da1c5. [17] Dickneite G, Doerr B, Kaspereit F. Characterization of the coagulation deficit in porcine dilutional coagulopathy and substitution with a prothrombin complex concentrate. Anesth Analg 2008;106:1070–7. [18] Spahn DR, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar E, Gordini G, et al. Management of bleeding following major trauma: a European guideline. Crit Care 2007;11:R17. [19] Schreiber MA, Holcomb JB, Rojkjaer R. Preclinical trauma studies of recombinant factor VIIa. Crit Care 2005;9(Suppl 5):S25–8. [20] Blajchman MA, Lee DH. The thrombocytopenic rabbit bleeding time model to evaluate the in vivo hemostatic efficacy of platelets and platelet substitutes. Transfus Med Rev 1997;11:95–105. [21] Tranholm M, Rojkjaer R, Pyke C, Kristensen AT, Klitgaard B, Lollike K, et al. Recombinant factor VIIa reduces bleeding in severely thrombocytopenic rabbits. Thromb Res 2003;109:217–23. [22] Fattorutto M, Tourreau-Pham S, Mazoyer E, Bonnin P, Raphaël M, Morin F, et al. Recombinant activated factor VII decreases bleeding without increasing arterial thrombosis in rabbits. Can J Anaesth 2004;51:672–9. [23] Godier A, Mazoyer E, Cymbalista F, Cupa M, Samama CM. Recombinant activated factor VII efficacy and safety in a model of bleeding and thrombosis in hypothermic rabbits: a blind study. J Thromb Haemost 2007;5:244–9. [24] Viuff D, Lauritzen B, Pusateri AE, Andersen S, Rojkjaer R, Johansson PI. Effect of 470 haemodilution, acidosis, and hypothermia on the activity of recombinant factor VIIa 471 (NovoSeven®). Br J Anaesth 2008;101:324–31. [25] Mann KG, Brummel K, Butenas S. What is all that thrombin for? J Thromb Haemost 2003;1:1504–14. [26] Turecek PL, Váradi K, Keil B, Négrier C, Berntorp E, Astermark J, et al. Factor VIII inhibitorbypassing agents act by inducing thrombin generation and can be monitored by a thrombin generation assay. Pathophysiol Haemost Thromb 2003;33:16–22. [27] Livnat T, Zivelin A, Martinowitz U, Salomon O, Seligsohn U. Prerequisites for recombinant factor VIIa-induced thrombin generation in plasmas deficient in factors VIII, IX or XI. J Thromb Haemost 2006;4:192–200. [28] Szlam F, Taketomi T, Sheppard CA, Kempton CL, Levy JH, Tanaka KA. Antithrombin affects hemostatic response to recombinant activated factor VII in factor VIII deficient plasma. Anesth Analg 2008;106:719–24. [29] Meng ZH, Wolberg AS, Monroe 3rd DM, Hoffman M. The effect of temperature and pH on the activity of factor VIIa: implications for the efficacy of high-dose factor VIIa in hypothermic and acidotic patients. J Trauma 2003;55:886–91. [30] Martinowitz U, Michaelson M. Guidelines for the use of recombinant activated factor VII (rFVIIa) in uncontrolled bleeding: a report by the Israeli Multidisciplinary rFVIIa Task Force. J Thromb Haemost 2005;3:640–8. [31] Rodewald L, Scharrer I. Protein chemical investigation of 30 consecutive lots of doublevirus eliminated PCC (Beriplex P/N) and summary of safety data. Hämostaseologie 2004;24:A46. [32] Aledort LM. Comparative thrombotic event incidence after infusion of recombinant factor VIIa versus factor VIII inhibitor bypass activity. J Thromb Haemost 2004;2:1700–8.
Contents lists available at ScienceDirect
Thrombosis Research j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t h r o m r e s
Regular Article
Prothrombin complex concentrate (Beriplex P/N) for control of bleeding after kidney trauma in a rabbit dilutional coagulopathy model☆ Ingo Pragst, Franz Kaspereit, Bärbel Dörr, Gerhard Dickneite ⁎ Department of Pharmacology and Toxicology, CSL Behring GmbH, Marburg, Germany
a r t i c l e
i n f o
Article history: Received 25 August 2009 Received in revised form 19 October 2009 Accepted 21 October 2009 Available online 14 November 2009 Keywords: Prothrombin complex concentrate Recombinant factor VIIa Trauma Blood coagulation disorders Hemorrhage Models, animal
a b s t r a c t Introduction: Fluid resuscitation after trauma often results in dilutional coagulopathy that may hinder control of bleeding and, once initial hemostasis has been secured, heighten risk of perioperative bleeding when further surgery is required. Since multiple coagulation factor deficiencies typically accompany fluid resuscitation, prothrombin complex concentrate (PCC) containing factors II, VII, IX and X may potentially offer greater hemostatic efficacy than coagulation factor monotherapy. Materials and methods: Anesthetized normothermic rabbits were hemodiluted 50-60% by phased blood withdrawal and infusion of hydroxyethyl starch and erythrocytes. The animals were randomly assigned to receive saline placebo, 25 IU·kg- 1 PCC (Beriplex P/N) or 180 μg·kg- 1 activated recombinant factor VII (rFVIIa; NovoSeven). Immediately thereafter, bleeding was precipitated by a standardized kidney incision. Results: PCC accelerated hemostasis compared both with saline and rFVIIa (p=0.002 for both comparisons). The median times to hemostasis in the PCC, saline and rFVIIa groups were 12, 19 and 28 min, respectively. PCC reduced blood loss by a median of 43 mL with a 95% confidence interval (CI) of 8.0-67.5 mL vs. saline and 82 mL (CI, 35.0110.0 mL) vs. rFVIIa. PCC augmented peak thrombin generation by a median of 104.1 nM (CI, 78.3-142.3 nM) compared with saline and 105.8 nM (CI, 70.7-139.5 nM ) relative to rFVIIa. At the respective 180 μg·kg- 1 and 25 IU·kg- 1 doses tested, rFVIIa displayed thrombogenicity in the Wessler stasis model, while PCC did not. Conclusions: In an animal model of dilutional coagulopathy and kidney trauma, PCC accelerated hemostasis and diminished blood loss compared with rFVIIa monotherapy. © 2009 Elsevier Ltd. All rights reserved.
Uncontrolled hemorrhage is a major challenge in trauma patients, accounting for as much as 25-30% of mortality [1]. Controlling bleeding in trauma patients is frequently complicated by major deficits in coagulation factors and platelets from blood loss, as well as dilutional coagulopathy arising when hypovolemia is remedied by infusing fluids devoid of coagulation factors, namely crystalloids, colloids and packed red blood cells. Moreover, metabolic derangements associated with major trauma such as acidosis and hypothermia can further compromise coagulation. Coagulopathic bleeding is particularly difficult to control [2]. A pressing need exists for effective agents to control refractory coagulopathic bleeding. Recombinant activated factor VII (rFVIIa) has been investigated as an adjunctive hemostatic agent for this purpose. While dramatic responses to rFVIIa in trauma patients with major Abbreviations: CI, 95% confidence interval; FII, factor II; FVII, factor VII; FIX, factor IX; FX, factor X; HES, hydroxyethyl starch; IQR, interquartile range; PCC, prothrombin complex concentrate; PT, prothrombin time; rFVIIa, recombinant factor VIIa. ☆ Presented in part at the 53rd Annual Meeting of the Society for Thrombosis and Hemostasis Research, Vienna, Austria, February 4-7, 2009. ⁎ Corresponding author. Senior Director of Preclinical Research and Development, CSL Behring GmbH, Emil von Behring-Straße 76, D-35041 Marburg, Germany. Tel.: +49 6421 39 2306; fax: +49 6421 39 5310. E-mail address: [email protected] (G. Dickneite). 0049-3848/$ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2009.10.011
hemorrhage have been documented in many case reports and small case series, results of controlled clinical trials have been variable [3]. In phase II clinical trials, rFVIIa significantly reduced the need for red blood cell transfusion in patients with blunt but not penetrating trauma, and in neither group was an improvement in survival attained [4,5]. One inherent limitation in replacing a single coagulation factor is the common presence of deficiencies in multiple factors after trauma and fluid resuscitation. Prothrombin complex concentrate (PCC) may afford a promising alternative to rFVIIa monotherapy for uncontrolled coagulopathic bleeding. The widely investigated PCC Beriplex P/N contains coagulation factors II (FII), VII (FVII), IX (FIX) and X (FX), as well as the anticoagulant proteins C and S. This PCC has proven highly effective for emergency reversal of coumarin oral anticoagulant therapy, an indication also involving multiple coagulation factor deficiencies [6–8]. In a rodent model of sustained coumarin anticoagulation, the hemostatic efficacy of Beriplex P/N was superior to that of rFVIIa [9]. Recently, in a porcine model of dilutional coagulopathy and spleen trauma, PCC was shown to accelerate hemostasis and increase thrombin generation compared with rFVIIa [10]. One limitation of that study was the absence of data on thrombogenic potential. The most wellestablished animal model for assessing thrombogenicity is in the rabbit [11]. So that both hemostatic efficacy and thrombogenicity could be
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
investigated simultaneously in the same animal species, a rabbit model of dilutional coagulopathy and kidney trauma has been developed. That model is described in this report, and data on the comparative hemostatic effects of PCC and rFVIIa and the thrombogenic potential of those agents are presented. Materials and methods Animals Female CHB rabbits 3-4 mo old weighing 2.8-4.0 kg (Bauer, Neuental, Germany) were housed one per cage in wire-steel cages at 21-23 °C and 50% relative humidity under a 12 h/12 h light-darkness cycle. The animals were provided tap water ad libitum and fed rabbit pellets (Deukanin®, Deutsche Tiernahrung Cremer GmbH & Co. KG, Düsseldorf, Germany). All rabbits received care in compliance with the European Convention on Animal Care, and the study was approved by the organizational Ethics Committee. Endpoints The primary study endpoints were time to hemostasis and blood loss as observed up to 30 min following a standardized kidney incision injury (Fig. 1). Time to hemostasis was defined as the interval from the kidney incision until cessation of observable bleeding or oozing. Blood loss was the volume of blood collected from the incision site by suction. Both time to hemostasis and blood loss were assessed by observers blinded to the group assignments of the rabbits. Prothrombin time (PT) was a secondary endpoint. Anesthesia A combination of 5 mg·kg- 1 i.v. ketamine (Ketavet®, Pharmacia GmbH, Erlangen, Germany) and 0.5 mg·kg- 1 i.v. xylazine 2% (Rompun®, Bayer Vital GmbH, Leverkusen, Germany) was used for induction of anesthesia. The animals were then intubated and placed on a ventilator
273
(Heyer Access, Heyer Medical AG, Bad Ems, Germany). Inhaled anesthesia was maintained with isoflurane (Isofluran CP®, CP-Pharma Handelsgesellschaft mbH, Burgdorf, Germany) at a concentration of 2.4% depending upon depth of anesthesia. A 14-gauge catheter was introduced into the carotid artery for continuous blood pressure monitoring and blood sampling and a 20-gauge catheter into the external jugular vein for hemodilution and test fluid administration. Maintenance i.v. fluid requirement was satisfied by infusion of 4 mL·kg- 1·h- 1 Ringer's lactate. Body temperature was monitored by rectal thermometry. Following a 20 min stabilization period, baseline measurements were made of hemodynamic, coagulation and hematological parameters. Hemodilution Animals were subjected to hemodilution in phases by withdrawal of 30 mL·kg- 1 blood and infusion of 30 mL·kg- 1 hydroxyethyl starch (HES) 200/0.5 (Infukoll 6%, Schwarz Pharma AG, Mannheim, Germany) prewarmed to 37 °C (Fig. 1). That procedure was repeated at 45 min. At 30 min, during the interval between the two cycles of blood withdrawal and HES infusion, the animals received 15 mL·kg- 1 salvaged erythrocytes, prepared from withdrawn whole rabbit blood by centrifugation for 10 min at 800×g, washing in normal saline and resuspension in Ringer's lactate. Treatment Animals were randomly allocated to receive i.v. infusions of isotonic saline, 25 IU·kg- 1 PCC (Beriplex® P/N, CSL Behring GmbH, Marburg, Germany) or 180 µg·kg- 1 rFVIIa (NovoSeven®, Novo Nordisk A/S, Bagsværd, Denmark) immediately prior to kidney incision injury (Fig. 1). The mean concentrations of FII, FVII, FIX and FX in Beriplex P/N are 30, 15, 30 and 40 IU·mL- 1, respectively [12]. A randomized negative control group did not undergo hemodilution. Experimental groups consisted of 5-7 rabbits each. Blood samples for determination of PT, coagulation factor and fibrinogen concentrations and platelet counts were
Fig. 1. Study procedures for hemodilution, treatment, experimental kidney trauma and assessment of hemostatic effect. Abbreviations: HES, hydroxyethyl starch; PCC, prothrombin complex concentrate; rFVIIa, activated recombinant factor VII.
274
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
collected at baseline, the conclusion of hemodilution and just before kidney incision. Dosing was selected to afford adequate opportunity for showing hemostatic efficacy while avoiding excessive risk of thromboembolic events. In porcine trauma model studies, 25-35 IU·kg- 1 PCC accelerated hemostasis [10,13]. While those studies were not specifically designed to assess thrombogenicity, there was no overt clinical evidence of thromboembolic events. Preliminary measurements of thrombin generation in the rabbit suggested that 25 IU·kg- 1 would be adequate to demonstrate hemostatic efficacy. Bleeding reduction has been observed after rFVIIa doses as low as 100-150 µg·kg- 1 in rabbit trauma model studies, and patients with blunt trauma receiving 100-200 µg·kg- 1 rFVIIa in randomized clinical trials required less red blood cell transfusion without increased incidence of thromboembolic complications [4]. In the rabbit, 100-300 µg·kg- 1 rFVIIa displayed substantially less thrombogenicity than 1000 µg·kg- 1 [14].
Statistical analysis
Kidney injury
Hemodilution
At 60 min after commencement of hemodilution, a standardized renal injury was inflicted in the form of a 15 mm long and 5 mm deep scalpel incision at the lateral kidney pole (Fig. 1). The 30 min observation period for blood loss and time to hemostasis began immediately after the incision.
At the completion of phased hemodilution, the pooled median observed magnitude of hemodilution for the saline, PCC and rFVIIa groups was 56.1% (IQR, 54.8-57.6%; n = 19), based upon measured decline in hematocrit from baseline. Hemodilution resulted in median decreases of 0.13 in pH, 1.5 °C in rectal temperature, 1.00 g·L- 1 in fibrinogen concentration and 189 × 109 L- 1 in platelet count (Table 1). PT was prolonged by a median of 8.3 s in response to hemodilution (Table 1). Subsequent to hemodilution, concentrations of FII, FVII, FIX and FX were reduced to approximately 20-30% of baseline (Table 2).
Laboratory assays PT was measured with a Schnitger & Gross coagulometer using the Thromborel reagent (Dade Behring, Marburg, Germany). Plasma for coagulation factor and fibrinogen determinations was prepared from citrated blood samples and stored at –80 °C until assay. FII, FIX and FX were measured in coagulation factor-deficient plasma (Dade Behring) with a Behring Coagulation Timer (BCT®, Dade Behring). FVII was quantified with a human-specific enzyme-linked immunosorbent assay (Biozol GmbH, Eching, Germany). Fibrinogen was assayed by the Clauss method using the Behring Coagulation System (BCS®) and reagents (Dade Behring). Thrombin generation assay As described by Hemker et al. [15], thrombin generation assay (TGA) of diluted platelet-poor plasma samples was performed by calibrated automated thrombinography (CAT, Thrombinoscope B.V., Maastricht, the Netherlands) using the Thrombinoscope software version 3.0.0.29 to calculate molar thrombin content. Concentrations of recombinant relipidated tissue factor and phospholipids were 5 pM and 4 μM, respectively. Calculated assay parameters were peak thrombin generation and lag time until observable thrombin generation. Wessler model The thrombogenicity of PCC and rFVIIa was assessed in female New Zealand White rabbits using the Wessler stasis model [11]. Methodological details have been recently described elsewhere [16]. Groups of 310 anesthetized rabbits received 50, 100, 200, 300, 400 or 500 IU·kg- 1 PCC or 10, 50, 100, 180 or 300 µg·kg- 1 rFVIIa. At 10 min thereafter, an isolated jugular vein segment was allowed to fill with blood by ligation. After a 30 min period of stasis, the vein segment was excised and dissected in a Petri dish filled with sodium citrate solution. A thrombosis score was assigned on a scale of 0 to 3, with 0 denoting no clotting, 1 clotting without organized thrombus formation, 2 one or more discrete thrombi and 3 an occlusive thrombus. As an objective measure of thrombus formation, the wet weight of thrombi formed during stasis was also determined.
Descriptive statistics consisted of the median and range or interquartile range (IQR). Median differences were computed with exact 95% confidence intervals (CI) using Hodges-Lehmann estimation, and corresponding p values were calculated by exact Wilcoxon test. Time to hemostasis was analyzed by the Kaplan-Meier product limit method, and between-group differences in this endpoint were evaluated by exact logrank test. The dose-response relationships of PCC and rFVIIa with thrombosis score and thrombus weight were characterized by linear regression. R version 2.7.2 (The R Foundation for Statistical Computing, Vienna, Austria) and StatXact 7.0 (Cytel Software Corp., Cambridge, Massachusetts, USA) statistical software were used for all analyses. Results
Coagulation factors Saline infusion after hemodilution showed no effect on concentrations of FII, FVII, FIX or FX (Table 2). After administration of PCC, which contains FII, FVII, FIX and FX, the median concentrations of those four coagulation factors increased respectively to 89, 75, 38 and 181% of baseline levels. As in the pig [17], measured FIX levels in the rabbit are 2-3 fold those in humans. Consequently, the administered PCC dose produced a comparatively small FIX increase relative to the high baseline level. Conversely, measured FX level in rabbits is only approximately 40% of the corresponding human value, so that a comparatively large FX concentration rise was attained after PCC administration. Hemostasis Relative both to saline and rFVIIa, PCC significantly shortened time to hemostasis after experimental kidney trauma (Fig. 2). The median time to hemostasis was 12 min in PCC-treated rabbits compared with 19 and 28 min for the groups receiving saline and rFVIIa, respectively. In all three treatment groups time to hemostasis was retarded vs. the negative control group not subjected to hemodilution. Cessation of kidney bleeding was observed within the 30 min observation period Table 1 Effects of hemodilution on physiological and hematologic parameters. Parameter
pH Temperature (° C) Fibrinogen (g·L- 1)† Platelets (× 109 L- 1) PT (s)
Median (IQR), n = 19 Baseline
Hemodilution
Change
7.51 (7.46 to 7.54) 39.3 (39.1 to 39.6) 1.70 (1.57 to 1.75) 292 (258 to 329) 11.0 (10.9 to 11.9)
7.36 (7.33 to 7.39) 37.8 (37.5 to 38.1) 0.68 (0.59 to 0.85) 107 (89 to 118) 19.5 (18.4 to 21.3)
-0.13 (-0.16 to -0.09) - 1.5 (- 1.7 to - 1.4) -1.00 (-1.04 to -0.90) - 189 (-216 to - 166) 8.3 (7.5 to 9.8)
Median pH, rectal temperature, fibrinogen concentration, platelet count, and PT at baseline and after hemodilution and corresponding median changes. Abbreviations: IQR, interquartile range; PT, prothrombin time. † n=17.
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
275
Table 2 Coagulation factors after hemodilution and treatment with saline or PCC. Factor
II VII IX X
Group
Saline PCC Saline PCC Saline PCC Saline PCC
n
5 6 5 6 6 6 5 6
Median Percent of Baseline (Range) Hemodilution
Treatment
22.0 23.2 34.6 31.6 26.7 28.4 21.1 21.6
24.1 (23.3 to 26.7) 88.8 (71.7 to 94.3) 37.2 (33.0 to 41.4) 75.3 (41.9 to 88.4) 28.0 (18.1 to 29.7) 37.7 (29.4 to 43.9) 24.4 (22.0 to 27.2) 181 (106 to 242)
(18.0 (20.7 (32.3 (28.4 (17.7 (22.2 (11.0 (20.1
to to to to to to to to
24.6) 26.3) 38.5) 34.6) 28.1) 30.3) 23.2) 22.9)
Median concentrations of coagulation factors II, VII, IX and X as a percent of baseline after hemodilution and treatment with saline or PCC. Abbreviation: PCC, prothrombin complex concentrate.
in all animals of the PCC, saline and negative control groups. In the rFVIIa group, bleeding persisted after 30 min in 3 of 6 rabbits. Results with respect to blood loss volume generally paralleled those for time to hemostasis (Fig. 3). Thus, blood loss in the PCC-treated rabbits was significantly lower by a median of 43 mL vs. the saline group and by 82 mL vs. rFVIIa recipients. All three treatment groups lost more blood than did non-hemodiluted control animals. While blood loss was higher by a median of 38 mL after rFVIIa than saline infusion, this difference was not statistically significant. Prothrombin time PT was prolonged by a median of 7.2 s (CI, 4.7 to 9.7 s) among saline-treated rabbits and by a median of 5.2 s (CI, 2.5 to 9.8 s) after PCC infusion, as compared with the non-hemodiluted control group. rFVIIa fully normalized PT. Thrombin generation As evidenced by the saline-treated group, hemodilution reduced peak thrombin generation by a median of 56.3 nM (CI, 37.3 to 77.7 nM) compared with that in non-hemodiluted animals. rFVIIa showed no effect on peak thrombin generation vs. the saline group. In contrast, PCC augmented peak thrombin generation by a median of 104.1nM (CI, 78.3 to 142.3 nM) and 105.8 nM (CI, 70.7 to 139.5 nM) compared with saline- and rFVIIa-treated rabbits, respectively.
Fig. 3. Blood loss after hemodilution and infusion of saline, PCC or rFVIIa. A negative control group did not undergo hemodilution. Median values are represented by horizontal lines. Abbreviations: CI, 95% confidence interval; PCC, prothrombin complex concentrate; rFVIIa, recombinant factor VIIa.
Treatment effects on thrombin generation lag time differed from those on peak thrombin. Thus, lag time in the saline and PCC groups did not significantly differ from that in non-hemodiluted controls, whereas rFVIIa markedly shortened lag time by a median of 0.89 min (CI, 0.73 to 1.07 min) vs. saline and a median of 1.06 min (CI, 0.45 to 1.12 min) vs. PCC. Thrombogenicity Over dose ranges of 50-500 IU·kg- 1 PCC and 10-300 μg·kg- 1 rFVIIa, continuous increases were observed in thrombosis score (Fig. 4(a) and (b), respectively) and thrombus weight (Fig. 4(c) and (d), respectively). In all rabbits receiving 50 IU·kg- 1 PCC, i.e. twice the dose exhibiting hemostatic efficacy after dilutional coagulopathy, both thrombosis score and thrombus weight were zero. The 180 μg·kg- 1 rFVIIa dose evaluated for hemostatic efficacy was associated with evidence of thrombogenicity. At that dose, the mean thrombosis score was 1.8 (CI, 1.5-2.1) and the mean thrombus weight 41 mg (CI, 27-55 mg), as estimated by linear regression analysis. Discussion
Fig. 2. Time to hemostasis following experimental kidney trauma in animals not subjected to hemodilution or treated following hemodilution with saline, PCC or rFVIIa. Abbreviations: PCC, prothrombin complex concentrate; rFVIIa, recombinant factor VIIa.
In this rabbit model of dilutional coagulopathy and traumatic kidney bleeding, PCC significantly decreased both time to hemostasis and blood loss compared with rFVIIa or saline. The accompanying PCC-mediated increase in peak thrombin generation provides a mechanistic basis for the observed hemostatic effects of PCC. This is the first study to evaluate the hemostatic efficacy of PCC in a rabbit trauma model. The present findings are consistent with those in a porcine model of dilutional coagulopathy and traumatic splenic bleeding, in which PCC infusion just prior to splenic incision augmented thrombin generation and accelerated hemostasis vs. rFVIIa or saline placebo [10]. In the same porcine model, PCC also shortened the time to hemostasis and diminished blood loss after either experimental bone or spleen trauma as compared with fresh frozen plasma (FFP) [13]. These three animals studies suggest that PCC could potentially prove to be a valuable adjunctive therapy for coagulopathic bleeding in trauma. Currently, FFP is the standard choice to correct dilutional coagulopathy following major trauma [18]. But its limitations, such as relatively slow correction of coagulopathy, have led to the investigation of other more convenient and rapidly effective agents such as rFVIIa. PCCs were originally developed for the treatment of hemophilia and have been
276
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
Fig. 4. (a) PCC and (b) rFVIIa dose-response relationships with thrombosis score and (c) and (d) corresponding relationships with thrombus weight. Thick lines show best-fit linear functions and dashed curves the CI of the regressions. Abbreviations: CI, 95% confidence interval; PCC, prothrombin complex concentrate; rFVIIa, recombinant factor VIIa.
used for more than 30 years to provide prompt and effective reversal of coumarin oral anticoagulant therapy [6–8]. PCC shares with rFVIIa the advantage of suitability for rapid infusion and undelayed normalization of coagulation parameters. Animal models simulating clinical traumatic injuries have primarily been developed in pigs because of similarities to humans in coagulation function and internal organ size [19]. However, the rabbit provides a useful alternative due to its intermediate size, ease of handling and relatively large circulating blood volume for its size [20]. Importantly, a well-characterized model for assessing thrombogenic potential exists in the rabbit [11]. Lastly, rabbit thromboplastin is a standard reagent in coagulation tests and may display greater activity when interacting with human FVII than swine thromboplastin [19]. While fully normalizing PT in the present study, rFVIIa at a dose of 180 µg·kg- 1 exhibited poor hemostatic efficacy in normothermic rabbits with dilutional coagulopathy and thrombocytopenia resulting from blood withdrawal and HES infusion. In previous rabbit studies, rFVIIa has exhibited hemostatic efficacy, but essential experimental parameters have differed from those in the present study [21–24]. In a study of noncoagulopathic rabbits with hepato-splenic injury, 150 µg·kg- 1 rFVIIa shortened bleeding time under both normo- and hypothermic conditions but reduced blood loss only in hypothermic animals [23]. Among rabbits with dilutional coagulopathy due to HES replacement, 100 µg·kg- 1 rFVIIa decreased ear immersion bleeding time and microvascular bleeding compared with placebo [22]. The
endpoint of two other studies was nail cuticle bleeding time, and high rFVIIa doses were administered [21,24]. In rabbits with irradiationinduced thrombocytopenia, 2000 µg·kg- 1 rFVIIa diminished bleeding time and blood loss [21]. A 5000 µg·kg- 1 rFVIIa dose was effective in reducing bleeding time of rabbits subjected to calculated hemodilution of 50% with HES 200/0.5, the same type as in the present study, but not with a higher molecular weight, more highly substituted HES solution [24]. In addition to normalizing PT, rFVIIa shortened the lag time before substantial thrombin generation. The hemostatic process is composed of an initiation phase and a propagation phase. Approximately, 5% of the total thrombin generation occurs during the initiation phase and 95% in the propagation phase [25]. PT is a measure of activity in the initiation phase [25]. The PT and lag time observations in the present study suggest that rFVIIa effectively accelerated the initiation phase. However, in light of its failure to increase peak thrombin generation, rFVIIa appeared in this model to be comparatively ineffective in sustaining the propagation phase and, accordingly, in stopping bleeding. Thrombin generation by rFVIIa was measured using platelet-poor plasma in this study, as well as in several previous studies [26–28]. Measurements in platelet-rich plasma might more faithfully simulate in vivo thrombin generation by rFVIIa, since the mechanism of rFVIIa action involves binding to activated platelets. Nevertheless, the greater observed peak thrombin generation by PCC than rFVIIa in the present study was consistent with the relative acceleration of hemostasis and reduction in blood loss after administration of PCC compared with rFVIIa. Some patients do not respond to high-dose rFVIIa [29]. Several variables such as fibrinogen and platelet levels, acid-base balance and temperature appear to influence the response to rFVIIa. Multidisciplinary task force guidelines from Israel and recommendations for the use of rFVIIa in uncontrolled bleeding call for blood component therapy to ensure fibrinogen level >0.5 g·L- 1 and platelet count > 50× 109 L- 1, correction of acidosis (pH > 7.1) and restoration of normothermia as much as feasible before rFVIIa therapy is considered [30]. While clinical investigations of PCC in trauma remain to be conducted, PCC is not currently known to be subject to the same influences as rFVIIa. In this study PCC showed hemostatic efficacy despite profound depletion of fibrinogen and platelets and a modest decline in pH. In the pig, PCC also proved to be an effective hemostatic agent during hypothermia [10,13]. Like other procoagulant agents, PCC and rFVIIa hold the potential for thrombogenicity. At the 25 IU·kg- 1 PCC and 180 µg·kg- 1 rFVIIa doses tested for hemostatic efficacy in the present study, rFVIIa displayed thrombogenicity in the Wessler model, while PCC did not. Both 25 IU·kg- 1 and 180 µg·kg- 1 are standard doses of PCC and rFVIIa, respectively, in clinical practice. In a previous study employing the Wessler model [14], rFVIIa exhibited thrombogenicity at all evaluated doses, namely, 100, 300 and 1000 µg·kg- 1, during the same 30 min stasis period as in the present study. The numbers and weights of thrombi observed after rFVIIa treatment in that study were comparable to those after 50-100 IU·kg- 1 doses of factor VIII inhibitor bypass activity (FEIBA). The present findings suggest that PCC may offer an attractive safety profile in trauma. Nevertheless, further investigations on the thrombogenic potential of PCC will be needed. The present rabbit model results are in conformity with existing clinical data from pharmacovigilance programs. In one such study, the estimated incidence of thrombotic events after PCC administration was 1.0 per 105 infusions (CI, 0.1-3.6 per 105 infusions) [31]. The corresponding incidence after rFVIIa administration in another pharmacovigilance study was 24.6 per 105 infusions (CI, 19.1-31.2 per 105 infusions) [32]. It should be recognized, however, that pharmacovigilance studies encompass all clinical uses rather than use specifically in trauma. Simultaneous replenishment of multiple coagulation factors with PCC in trauma embodies an attractive therapeutic concept, since trauma patients usually develop multiple factor deficiencies. This study adds to preclinical evidence that PCC can effectively control bleeding after a variety of hemorrhagic insults and under a range of conditions. These
I. Pragst et al. / Thrombosis Research 125 (2010) 272–277
encouraging findings suggest that further investigation of PCC in trauma is warranted. Conflict of interest statement The authors are employees of the study sponsor, CSL Behring. Acknowledgments Funding for this investigation was furnished by CSL Behring GmbH. The sponsor is the employer of the authors and as such played an active role in the design of the study, the collection, analysis and interpretation of data, the writing of the manuscript and the decision to submit the manuscript for publication. References [1] Pfeifer R, Tarkin IS, Rocos B, Pape HC. Patterns of mortality and causes of death in polytrauma patients—has anything changed? Injury 2009;40:907–11. [2] Vincent JL, Rossaint R, Riou B, Ozier Y, Zideman D, Spahn DR. Recommendations on the use of recombinant activated factor VII as an adjunctive treatment for massive bleeding—a European perspective. Crit Care 2006;10:R120. [3] Duchesne JC, Mathew KA, Marr AB, Pinsky MR, Barbeau JM, McSwain NE. Current evidence based guidelines for factor VIIa use in trauma: the good, the bad, and the ugly. Am Surg 2008;74:1159–65. [4] Boffard KD, Riou B, Warren B, Choong PI, Rizoli S, Rossaint R, et al. Recombinant factor VIIa as adjunctive therapy for bleeding control in severely injured trauma patients: two parallel randomized, placebo-controlled, double-blind clinical trials. J Trauma 2005;59:8–15. [5] Rizoli SB, Boffard KD, Riou B, Warren B, Iau P, Kluger Y, et al. Recombinant activated factor VII as an adjunctive therapy for bleeding control in severe trauma patients with coagulopathy: subgroup analysis from two randomized trials. Crit Care 2006;10:R178. [6] Evans G, Luddington R, Baglin T. Beriplex P/N reverses severe warfarin-induced overanticoagulation immediately and completely in patients presenting with major bleeding. Br J Haematol 2001;115:998–1001. [7] Preston FE, Laidlaw ST, Sampson B, Kitchen S. Rapid reversal of oral anticoagulation with warfarin by a prothrombin complex concentrate (Beriplex): efficacy and safety in 42 patients. Br J Haematol 2002;116:619–24. [8] Pabinger I, Brenner B, Kalina U, Knaub S, Nagy A, Ostermann H. Prothrombin complex concentrate (Beriplex® P/N) for emergency anticoagulation reversal: A prospective multinational clinical trial. J Thromb Haemost 2008;6:622–31. [9] Dickneite G. Prothrombin complex concentrate versus recombinant factor VIIa for reversal of coumarin anticoagulation. Thromb Res 2007;119:643–51. [10] Dickneite G, Dörr B, Kaspereit F, Tanaka KA. Prothrombin complex concentrate versus recombinant factor VIIa for reversal of hemodilutional coagulopathy in a porcine trauma model. J Trauma 2009;67. doi:10.1097/TA.0b013e3181b06364. [11] Wessler S, Reimer SM, Sheps MC. Biologic assay of a thrombosis-inducing activity in human serum. J Appl Physiol 1959;14:943–6. [12] Ostermann H, Haertel S, Knaub S, Kalina U, Jung K, Pabinger I. Pharmacokinetics of Beriplex P/N prothrombin complex concentrate in healthy volunteers. Thromb Haemost 2007;98:790–7.
277
[13] Dickneite G, Pragst I. Prothrombin complex concentrate vs fresh frozen plasma for reversal of dilutional coagulopathy in a porcine trauma model. Br J Anaesth 2009;102: 345–54. [14] Diness V, Bregengaard C, Erhardtsen E, Hedner U. Recombinant human factor VIIa (rFVIIa) in a rabbit stasis model. Thromb Res 1992;67:233–41. [15] Hemker HC, Giesen P. Al Dieri R, Regnault V, de Smedt E, Wagenvoord R, Lecompte T, Béguin S. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb 2003;33:4–15. [16] Dickneite G, Pragst I, Joch C, Bergman GE. Animal model and clinical evidence indicating low thrombogenic potential of fibrinogen concentrate (Haemocomplettan P). Blood Coagul Fibrinolysis. doi:10.1097/MBC.0b013e32832da1c5. [17] Dickneite G, Doerr B, Kaspereit F. Characterization of the coagulation deficit in porcine dilutional coagulopathy and substitution with a prothrombin complex concentrate. Anesth Analg 2008;106:1070–7. [18] Spahn DR, Cerny V, Coats TJ, Duranteau J, Fernandez-Mondejar E, Gordini G, et al. Management of bleeding following major trauma: a European guideline. Crit Care 2007;11:R17. [19] Schreiber MA, Holcomb JB, Rojkjaer R. Preclinical trauma studies of recombinant factor VIIa. Crit Care 2005;9(Suppl 5):S25–8. [20] Blajchman MA, Lee DH. The thrombocytopenic rabbit bleeding time model to evaluate the in vivo hemostatic efficacy of platelets and platelet substitutes. Transfus Med Rev 1997;11:95–105. [21] Tranholm M, Rojkjaer R, Pyke C, Kristensen AT, Klitgaard B, Lollike K, et al. Recombinant factor VIIa reduces bleeding in severely thrombocytopenic rabbits. Thromb Res 2003;109:217–23. [22] Fattorutto M, Tourreau-Pham S, Mazoyer E, Bonnin P, Raphaël M, Morin F, et al. Recombinant activated factor VII decreases bleeding without increasing arterial thrombosis in rabbits. Can J Anaesth 2004;51:672–9. [23] Godier A, Mazoyer E, Cymbalista F, Cupa M, Samama CM. Recombinant activated factor VII efficacy and safety in a model of bleeding and thrombosis in hypothermic rabbits: a blind study. J Thromb Haemost 2007;5:244–9. [24] Viuff D, Lauritzen B, Pusateri AE, Andersen S, Rojkjaer R, Johansson PI. Effect of 470 haemodilution, acidosis, and hypothermia on the activity of recombinant factor VIIa 471 (NovoSeven®). Br J Anaesth 2008;101:324–31. [25] Mann KG, Brummel K, Butenas S. What is all that thrombin for? J Thromb Haemost 2003;1:1504–14. [26] Turecek PL, Váradi K, Keil B, Négrier C, Berntorp E, Astermark J, et al. Factor VIII inhibitorbypassing agents act by inducing thrombin generation and can be monitored by a thrombin generation assay. Pathophysiol Haemost Thromb 2003;33:16–22. [27] Livnat T, Zivelin A, Martinowitz U, Salomon O, Seligsohn U. Prerequisites for recombinant factor VIIa-induced thrombin generation in plasmas deficient in factors VIII, IX or XI. J Thromb Haemost 2006;4:192–200. [28] Szlam F, Taketomi T, Sheppard CA, Kempton CL, Levy JH, Tanaka KA. Antithrombin affects hemostatic response to recombinant activated factor VII in factor VIII deficient plasma. Anesth Analg 2008;106:719–24. [29] Meng ZH, Wolberg AS, Monroe 3rd DM, Hoffman M. The effect of temperature and pH on the activity of factor VIIa: implications for the efficacy of high-dose factor VIIa in hypothermic and acidotic patients. J Trauma 2003;55:886–91. [30] Martinowitz U, Michaelson M. Guidelines for the use of recombinant activated factor VII (rFVIIa) in uncontrolled bleeding: a report by the Israeli Multidisciplinary rFVIIa Task Force. J Thromb Haemost 2005;3:640–8. [31] Rodewald L, Scharrer I. Protein chemical investigation of 30 consecutive lots of doublevirus eliminated PCC (Beriplex P/N) and summary of safety data. Hämostaseologie 2004;24:A46. [32] Aledort LM. Comparative thrombotic event incidence after infusion of recombinant factor VIIa versus factor VIII inhibitor bypass activity. J Thromb Haemost 2004;2:1700–8.