Skip to main content
NCBI home page
Journal of Cerebral Blood Flow & Metabolism logoLink to Journal of Cerebral Blood Flow & Metabolism
. 2013 Dec 18;34(3):495–501. doi: 10.1038/jcbfm.2013.226

Rivaroxaban does not increase hemorrhage after thrombolysis in experimental ischemic stroke

Robert Ploen 1,5, Li Sun 1,5, Wei Zhou 1, Stefan Heitmeier 2, Markus Zorn 3, Ekkehart Jenetzky 4, Roland Veltkamp 1,*
PMCID: PMC3948130  PMID: 24346690

Abstract

The management of acute ischemic stroke during anticoagulation with a novel oral anticoagulant (NOAC) is challenging because intravenous thrombolysis is contraindicated because of a putative increased risk of intracerebral hemorrhagic complications. We examined the risk of secondary postischemic hemorrhage after thrombolysis in rodents pretreated with rivaroxaban or warfarin. Mice were pretreated with either rivaroxaban (30 mg/kg), warfarin (target international normalized ratio 2 to 3) or vehicle. After 2 or 3 hours, middle cerebral artery occlusion (MCAO), mice received 9 mg/kg recombinant tissue plasminogen activator. Twenty-four hours after MCAO, secondary hemorrhage was quantified using a macroscopic hemorrhage score and hemoglobin spectrophotometry. Blood–brain barrier (BBB) permeability was measured by Evans Blue spectrofluorometry. To increase the validity of our findings, experiments were also performed using a thromboembolic model in anticoagulated rats. Infarct size did not differ among groups. Pretreatment with warfarin led to significantly more secondary hemorrhage compared with rivaroxaban and nonanticoagulated controls after 2- and 3-hour ischemia in mice as well as in rats. Blood–brain barrier permeability was significantly higher in the warfarin group compared with rivaroxaban and control. Thus, rivaroxaban in contrast to warfarin does not increase secondary hemorrhage after thrombolysis in experimental cerebral ischemia. Less effects of rivaroxaban on postischemic BBB permeability may account for this difference.

Keywords: animal models, antithrombotics, blood–brain barrier, brain ischemia, intracerebral hemorrhage, thrombolysis

Introduction

The recent introduction of direct thrombin (dabigatran)1 and factor Xa inhibitors (rivaroxaban and apixaban)2, 3 provides new opportunities for stroke prevention for millions of patients with atrial fibrillation worldwide. However, these novel oral anticoagulants (NOAC) pose a considerable challenge for the emergency management of acute ischemic stroke despite long-term anticoagulation.4, 5, 6

Even without anticoagulants, intravenous thrombolysis with recombinant tissue plasminogen activator (rtPA), the only approved therapy for ischemic stroke to date,7, 8 increases the risk of secondary hemorrhage more than fivefold.9 Effective long-term oral anticoagulation with vitamin K antagonists (i.e., international normalized ratio (INR) >2) increases the risk of intracerebral hemorrhage after thrombolytic therapy in patients and cerebral ischemia models.10, 11 The risk of thrombolysis in patients with subtherapeutic warfarin use is controversial.12, 13, 14 Current guidelines recommend excluding stroke patients with INR exceeding 1.7 from thrombolytic therapy with rtPA.7

Anticoagulation with the NOAC was associated with significantly less primary intracerebral hemorrhagic complications compared with warfarin in clinical trials.1, 2, 3 Because thrombolytic therapy in the event of an acute ischemic stroke was contraindicated in all trials of NOAC, data on the biologic effects of NOAC in the setting of cerebral ischemia and thrombolysis are very limited.6, 15 Based on the lack of experience with NOAC in thrombolysis and on extrapolation from the adverse effects of warfarin, patients under effective anticoagulation with NOAC are currently excluded from systemic thrombolysis.4, 5

So far, no data concerning the effects of thrombolysis after experimental ischemia in animals or acute ischemic stroke in patients on long-term anticoagulation with rivaroxaban have been published.6 The purpose of the present study was to assess the effect of anticoagulation with rivaroxaban on the risk of secondary hemorrhage after intravenous thrombolysis in two well-established rodent models of ischemic stroke with reperfusion.

Materials and Methods

This study was conducted in accordance with the German animal welfare law and the ARRIVE guidelines (http://www.nc3rs.org/ARRIVE). All experiments were approved by governmental animal care authorities (Regierungspräsidium Karlsruhe, Germany). Experiments were performed using male C57/Bl-6 mice weighing 23 to 28 g and male Wistar rats weighing 280 to 350 g, respectively (Charles River Laboratories, Sulzfeld, Germany). Animals were randomly assigned to receive warfarin, rivaroxaban, or no anticoagulation. A priori sample size calculation and exclusion criteria are described in online only Supplementary Information.

Drug Administration

Micronized rivaroxaban (Bayer HealthCare AG, Wuppertal, Germany) was dissolved in vehicle solution (3 mg/mL) containing 10% ethanol, 40% Solutol HS 15 (Sigma-Aldrich, Steinheim, Germany) and 50% H2O as described.16 Both mice and rats received 30 mg/kg rivaroxaban via gastric gavage 1 hour before middle cerebral artery occlusion (MCAO). This dose was chosen based on a previous study showing its anticoagulatory effectiveness in vitro and its ability to enlarge collagenase-induced intracerebral hemorrhage in vivo.17 Mice in the warfarin group received warfarin (Coumadin, Bristol-Myers-Squibb, München, Germany) in tap water (2.5 mg warfarin dissolved in 800 mL water) over a period of 3 days resulting in a dose of ∼0.4 mg/kg per day. This dose produces anticoagulation with an INR target of 2 to 318 (see below). To eliminate effects of the vehicle solution or trauma caused by the gastric gavage, all animals in the warfarin and control groups received 10 mL/kg vehicle solution via gastric gavage 1 hour before operation.

Assessment of Coagulation Status

In the warfarin group, INR was assessed by a point-of-care coagulometer (Coaguchek XS, Roche, Penzberg, Germany)19 before surgery to assure effective systemic anticoagulation in each individual animal (INR 2 to 3; Supplementary Information).

In separate experiments with vehicle, rivaroxaban or warfarin treatment (n=5 per group), 1 mL of citrated blood was collected by retroorbital puncture 1 hour after drug administration. INR (Innovin clotting assay, Sysmex CA-7000, Siemens Healthcare Diagnostics, Marburg, Germany) and activities of different factors (detected with reagents and BCS analyzer system from Siemens Healthcare Diagnostics, see Supplementary Information) were analyzed in the central laboratory of the University Hospital Heidelberg.

In contrast to vehicle and warfarin, the impact of rivaroxaban on in vitro coagulation is not adequately reflected by the use of INR.20 Therefore, three additional animals per group and time point received 30 mg/kg rivaroxaban or 10 ml/kg control solution without anticoagulant. After 1, 8, 16, and 24 hours in mice as well as 1, 4, and 8 hours in rats, 1 mL of whole blood was drawn from animals into a citrate tube. Platelet-poor plasma was obtained by immediate centrifugation at 2500 g for 15 minutes and stored at −80°C. Platelet-poor plasma samples were further examined by one investigator (SH) masked to group assignment for the determination of rivaroxaban plasma concentration and prothrombin time using Neoplastin plus reagent as previously described17 as well as anti-factor Xa-activity (Supplementary Information).

Moreover, in vivo bleeding time was measured in five additional mice per group using a tail vein bleeding time assay. Eight hours after application of vehicle, 30 mg/kg rivaroxaban or discontinuation of warfarin with an INR of 2 to 3, respectively, mice were anaesthetized (see below). Distal 5 mm of mice tail was cut and tails were placed in 37°C warm saline. Time until termination of bleeding was measured. Experiments were stopped if bleeding did not cease within 20 minutes. In rats, tail vein bleeding time was measured in control group and 1, 4, and 24 hours after administration of rivaroxaban. Distal 2 mm of rat tails was dissected and time until termination of bleeding was measured (up to 30 minutes).

Surgical Procedure in Mice

Anesthesia was induced with 4% halothane and maintained with 1% halothane (Sigma-Aldrich) in 70% N2O and 30% O2 via facial mask. The middle cerebral artery was occluded by an intraluminal filament (70 fine MCAO suture, tip diameter 0.19 mm, Doccol, Redlands, CA, USA) as described21 with some modifications.22 Successful occlusion of the middle cerebral artery was verified by a reduction of transcranial laser Doppler signal of at least 70%. Two or three hours after MCAO, respectively, the filament was withdrawn to allow reperfusion. Recombinant tissue plasminogen activator 9 mg/kg (Actilyse, Boehringer Ingelheim, Germany) was injected 15 minutes before filament withdrawal via a tail vein catheter. Ten percent was injected as a bolus, and the remainder was infused over 30 minutes.

Surgical Procedure in Rats

Rats were pretreated with either 30 mg/kg rivaroxaban, warfarin (INR 2 to 3), or 3 ml/kg vehicle. Anesthesia was performed as described for mice. All rats underwent transient ischemia using a previously established thromboembolic model.23 Briefly, a catheter was introduced via the external into the right internal carotid artery. Then, 12 autologous blood clots were injected.24 Two hours after MCAO, 9 mg/kg of rtPA was injected via the femoral vein catheter (10% bolus, the remainder over 30 minutes).22 Although spontaneous recanalization is rare in this model, it has a high rate of recanalization induced by rtPA.24

Neurologic Score

Neurologic evaluation was performed before, 2 and 22 hours after reperfusion using a previously described neurologic score21 with modifications.10

Preparation of Sections and Tissue Samples

Twenty-four hours after MCAO, animals were anesthetized and transcardially perfused with saline. Brains were immediately frozen in isopentane at −20°C and stored at −80°C. For infarct size measurement, 20-μm thick coronal cryosections were cut every 400 μm and subsequently stained with cresyl violet (Sigma-Aldrich). Infarct volume was calculated by multiplying measured areas (ImageJ, version 1.37 V, NIH, Bethesda, MD, USA) with the distance between sections and was corrected for brain edema by subtracting the volume difference of ischemic and nonischemic hemisphere from the infarct volume.

Quantification of Intracerebral Hemorrhage

Photographs (PowerShot A470, Canon, Tokyo, Japan) of unstained coronal sections were taken every 400 μm. Photographs were scored by two independent raters that were masked to group assignment. Hemorrhage score ranged from 0 to 4: 0=no hemorrhage; 1=single petechial hemorrhage, 2=confluent petechial hemorrhage, 3=a single space-occupying parenchymal hemorrhage encompassing<30% of infarction area, 4=multiple space-occupying parenchymal hemorrhages or single space-occupying parenchymal hemorrhage>30% of infarction area (Supplementary Figure S1).

In an additional experiment with 12 animals per group, the blood volume of the ischemic hemisphere was quantified using a previously described spectrophotometric assay25 with modifications (Supplementary Information).

Blood–Brain Barrier Permeability for Evans Blue

To quantify postischemic blood–brain barrier (BBB) permeability, we used a previously established fluorometric assay to assess extravasation of albumin-bound Evans Blue.26 Four additional animals per group were operated as described above with 2-hour MCAO. Animals were briefly reanesthetized 22 hours after MCAO and 1% Evans Blue in phosphate-buffered saline (80 mg/kg; Sigma-Aldrich) was injected into the femoral vein. Two hours later, mice were transcardially perfused with saline and Evans Blue content of each hemisphere was quantified using fluorometry (Supplementary Information). To demonstrate the topographical distribution of Evans Blue extravasation, two additional animals per group were operated and photographs of coronal brain sections were taken every 400 μm.

In additional experiments (n=4 per group), we examined Evans Blue extravasation after 8-hour thromboembolic ischemia in rats. Evans Blue (2% in phosphate-buffered saline, 80 mg/kg) was injected 6 hours after thromboembolic MCAO. After transcardial perfusion, 8 hours after MCAO, the fluorometric assay of Evans Blue was performed (Supplementary Information). In previous studies, BBB permeability substantially increased at 6 hours after transient MCAO and remained unchanged until 24 hours.27, 28

Statistical Analysis

Statistical analysis was performed using SPSS (Version 19, SPSS Corporation, Chicago). All values are expressed as mean±s.d. Coagulation factor activity, INR values, infarct volume, and Evans Blue content were compared using analysis of variance with post hoc Bonferroni test. Macroscopic hemorrhage and neurologic scores were compared between groups using the non-parametric Kruskal–Wallis test and following Mann–Whitney U-test. Pearson's regression correlation was used to compare point-of-care with central lab INR, and neurologic score with infarct size. Inter-rater reliability was assessed by Cohen's Kappa. Mortality was analyzed with Pearson χ2-test. A P value <0.05 was considered statistically significant.

Results

Pharmacokinetics of Warfarin and Rivaroxaban in Mice

Point-of-care INR measurements of warfarin-treated animals correlated well with central lab INR (n=16; Pearson's r=0.959; P<0.01; Supplementary Figure S2). Treatment with warfarin led to pronouncedly decreased activities of all vitamin K-dependent coagulation factors (i.e., FII, FVII, FIX, and FX) (P<0.01; n=5; analysis of variance; Figure 1A). In the warfarin group, only mice with a central lab INR of 2.0 to 3.0 measured before MCAO were included in experiments. The target INR was reached after 3 days of pretreatment and INR remained significantly elevated for 8 hours after discontinuation of warfarin (Figures 1B and 1D).

Figure 1.

Figure 1

Open in a new tab

Effects of rivaroxaban and warfarin treatment on coagulation assays in mice. Animals were orally anticoagulated with 30 mg/kg rivaroxaban (via gastric gavage), warfarin (in drinking water for 3 days; 0.4 mg/kg per day; target INR 2.0 to 3.0) or not anticoagulated. Nonanticoagulated controls and warfarin pretreated animals received 10 mL/kg vehicle solution (10% ethanol, 40% solutol and 50% H2O) via gastric gavage. One hour after gavage, plasma samples were collected. (A) Activity of factors II, VII, IX, X, and protein C using assays with factor-deficient plasma. Activity in nonanticoagulated control mice was set as 100%, all other data are depicted as percentage compared with the value in control mice (n=5 per group). (B) International normalized ratio in the nonanticoagulated and the warfarin group. (C) Tail vein bleeding time in minutes 8 hours after gastric gavage (n=5 per group). (D) International normalized ratio in five individual mice measured before, 24, 48, and 72 hours after warfarin application and 8, 16, and 24 hours after warfarin withdrawal. Each line represents the kinetics of a single mouse (n=5). (E–G) Rivaroxaban plasma concentration, anti-Factor Xa activity and prothrombin time, respectively, measured in mice (n=3 per group and time point) 1, 8, 16, and 24 hours after application of rivaroxaban and 1 and 16 hours after application of control solution (no anticoagulation) in the control group. *P<0.05; analysis of variance.

Rivaroxaban plasma concentration rose to 0.90 mg/L at 1 hour after administration and remained detectable for 24 hours after application (Figure 1E). This resulted in an anti-factor Xa activity of 96% after 1 hour and 35% after 24 hours (Figure 1F). Prothrombin time measurements are represented in Figure 1G.

In vivo tail vein bleeding time in mice 8 hours after application of rivaroxaban or withdrawal of warfarin pretreatment was considerably increased to more than 20 minutes compared with a mean bleeding time of 1.3±0.6 minutes in nonanticoagulated controls (Figure 1C).

Pharmacokinetic of Warfarin and Rivaroxaban in Rats

Rivaroxaban treatment resulted also in a significant elevation of rivaroxaban plasma concentration, anti-factor X activity, and prothrombin time for at least 8 hours after application as well as a prolonged in vivo bleeding (Supplementary Figure S3).

Infarct Severity, Mortality, and Physiologic Parameters

In mice, mean relative reduction of cerebral blood flow measured by laser Doppler flowmetry (data not shown) and infarct volumes (Supplementary Table S1) did not differ significantly among groups in all experiments. Mortality within the first 24 hours did not differ among groups. Mortality was 7%, 13%, and 33% (protocol with 2-hour MCAO; P=0.14; Pearson's χ2-test) and 31%, 14%, 31% (protocol with 3-hour MCAO; P=0.90), in the nonanticoagulated control, rivaroxaban and warfarin groups, respectively.

In rats, blood pressure, heart rate, and arterial blood gases did not differ significantly between groups 10 minutes before and 2.5 hours after MCAO (Supplementary Table S2). Mortality in rats within the first 24 hours after MCAO as well as infarct volume did not significantly differ between groups (Supplementary Table S1).

Secondary Intracerebral Hemorrhage after Thrombolysis in Mice

In mice receiving thrombolysis after 2-hour MCAO with subsequent reperfusion, warfarin increased macroscopic hemorrhage on unstained coronal cryosections compared with nonanticoagulated control (mean hemorrhage score 2.1±0.9 versus 1.0±1.1; P<0.05; Mann–Whitney). In contrast, hemorrhage severity did not differ between mice pretreated with rivaroxaban (1.1±0.7; P>0.5) and nonanticoagulated control (Figure 2B). There was a high inter-rater reliability of hemorrhage scores between the two masked raters (Cohen's Kappa=0.816; P<0.001).

Figure 2.

Figure 2

Open in a new tab

Effects of pretreatment with oral anticoagulation on secondary hemorrhage after ischemia in mice. (A) Experimental protocol of 2-hour transient middle cerebral artery occlusion (MCAO) with recombinant tissue plasminogen activator (rtPA)-associated reperfusion. Fifteen mice per group were orally anticoagulated with 30 mg/kg rivaroxaban (via gastric gavage), warfarin (international normalized ratio 2.0 to 3.0) or not anticoagulated. (B) Macroscopic hemorrhage scored by a masked rater. Depicted are individual scores, mean and s.d. *P<0.05; n=15 per group; Kruskal–Wallis. (C) Photographs of representative unstained brain sections. (D) Calculated intracerebral blood volume (ipsilateral minus contralateral hemisphere) as measured by hemoglobin spectrophotometry in a separate experiment. *P<0.05; n=12 per group; analysis of variance.

Repetition of the experiment and using the spectrophotometric hemoglobin assay also revealed significantly more bleeding in mice pretreated with warfarin than in the rivaroxaban and control groups (Figure 2D).

We further examined secondary hemorrhage after 3-hour transient MCAO with thrombolysis. Again, the severity of secondary hemorrhage as measured by the macroscopic hemorrhage score did not differ between rivaroxaban and control (1.4±0.6 and 1.5±0.7; P=0.4) but was significantly increased in animals anticoagulated with warfarin (2.3±0.8; P<0.05; Figure 3B). Inter-rater reliability was high (Kappa=0.86; P<0.001).

Figure 3.

Figure 3

Open in a new tab

Effects of oral anticoagulation on secondary hemorrhage after prolonged ischemia in mice. (A) Experimental protocol of 3-hour transient middle cerebral artery occlusion (MCAO) with recombinant tissue plasminogen (rtPA)-associated reperfusion. (B) Macroscopic hemorrhage scored by a masked rater. Bars indicate mean and s.d. (C) Neurologic score evaluated 24 hours after ischemia. *P<0.05; n=15 per group; Kruskal–Wallis.

Secondary Intracerebral Hemorrhage after Thrombolysis in Rats

To increase the validity of our findings, we examined the effects of anticoagulation with rivaroxaban on secondary hemorrhage in a second species (rats) and another (thromboembolic) ischemia model. Rats in the control and rivaroxaban group had significantly lower macroscopic hemorrhage scores compared with animals pretreated with warfarin (Figure 4).

Figure 4.

Figure 4

Open in a new tab

Effects of oral anticoagulation on secondary hemorrhage after thromboembolic ischemia in rats. (A) Experimental protocol of thromboembolic middle cerebral artery occlusion (MCAO) and recombinant tissue plasminogen (rtPA) induced reperfusion in rat. Rats received 30 mg/kg rivaroxaban, warfarin (international normalized ratio 2 to 3) or no anticoagulation. (B) Macroscopic hemorrhage scored by a masked rater. Bars indicate mean and s.d. (C) Representative photographs of unstained brain sections. *P<0.05; Kruskal–Wallis.

Neurologic Score in Mice

The neurologic score after 24 hours correlated well with the infarct size (Pearson's r=0.74; P<0.001) but not with the hemorrhage score. There was no significant difference of the neurologic score among groups in all experiments (Supplementary Table S1; for 3-hours protocol with rtPA; see Figure 3C).

Effects of Anticoagulation on Blood–Brain Barrier Permeability

In separate mice (n=4 per group), Evans Blue extravasation in the ischemic hemisphere was quantified by fluorometry. There was no difference between nonanticoagulated control (73.9±109.1 ng) and mice pretreated with rivaroxaban (57.4±40.5 ng). In contrast, mice pretreated with warfarin had substantially more Evans Blue extravasate (339.3±177.0 ng; P<0.05, analysis of variance). Thus, only warfarin led to a pronounced increase of postischemic BBB permeability after thrombolysis (Figure 5B). Evans Blue extravasation was inhomogeneous and localized within the infarct (Figure 5C).

Figure 5.

Figure 5

Open in a new tab

Effects of oral anticoagulation on blood–brain barrier permeability in mice. (A) Experimental protocol to assess blood–brain barrier permeability using Evans blue extravasation. Two-hour transient middle cerebral artery occlusion (MCAO) with recombinant tissue plasminogen (rtPA)-associated reperfusion was performed in four mice per group. Evans blue was intravenously injected 22 hours after MCAO and 2 hours before euthanasia. (B) Difference of Evans blue content (ischemic minus nonischemic hemisphere) as measured by fluorometry of homogenized hemispheres. *P<0.05; n=4 per group; analysis of variance. (C) Topography and extent of Evans blue extravasation in unstained brains in additional experiments (n=2 per group).

In rats with thromboembolic MCAO, pretreatment with warfarin resulted in a significant increase of postischemic BBB permeability after thrombolysis compared with control, whereas Evans Blue extravasation in the rivaroxaban group did not differ from nonanticoagulated control (Supplementary Figure S4).

Discussion

Our study provides two major new findings: (1) Rivaroxaban does not increase the risk of secondary hemorrhage after thrombolysis in experimental cerebral ischemia whereas warfarin exacerbates hemorrhagic transformation. (2) Rivaroxaban in contrast to warfarin does not increase BBB permeability after ischemia and thrombolysis in mice and rats.

The present as well as previous studies have consistently demonstrated that effective anticoagulation with warfarin (i.e., INR>2.0) increases secondary hemorrhage after thrombolysis in various models of focal cerebral ischemia and recanalization in mice and rats.10, 11, 22 This is in accordance with an increased risk of hemorrhagic complications of thrombolysis in stroke patients during therapeutic anticoagulation with VKA.12 The present study is the first to establish a model of ischemic stroke and thrombolysis during oral anticoagulation with rivaroxaban. Herein, anticoagulation with rivaroxaban does not cause an excess risk of intracranial hemorrhage after thrombolysis. Similar findings have been described for dabigatran in previous studies with22, 29 and without thrombolysis.30, 31

It is unlikely that the different effects of warfarin and rivaroxaban on hemorrhagic transformation after cerebral ischemia were caused by a different intensity of anticoagulation. Owing to the different modes of action, a direct comparison between both agents using the same in vitro coagulation testing is of limited value.20 On the one hand, warfarin caused a prolonged increase of the INR and an unselective depression of the vitamin K-dependent coagulation factors II, VII, IX, X and protein C. However, we applied a very high dose of rivaroxaban resulting in peak plasma levels of 0.90 mg/L in mice. This exceeds human peak plasma levels after a once-daily dose of 20 mg rivaroxaban about threefold.32, 33 Using the Neoplastin prothrombin time assay that is sensitive to rivaroxaban,34 30 mg/kg rivaroxaban prolong the prothrombin time two- to threefold for at least 4 hours.17 In the present study, rivaroxaban reduced the activity of coagulation factors II and X (Figure 1A). Elevated values of rivaroxaban concentration and anti-factor Xa activity remain detectable even 24 hours after rivaroxaban application (Figures 1E and 1F). The in vivo bleeding assay showed substantial prolongation of tail vein bleeding time even 8 hours after administration of either warfarin or rivaroxaban (Figure 1C). Discontinuation of anticoagulation with both rivaroxaban and warfarin results in normalization of prothrombin time and INR within 16 hours. The pretreatment for 3 days in the warfarin group were performed because of the delayed onset of action of warfarin. These findings underline that rivaroxaban also had a profound systemic anticoagulatory effect in our experiments.

Pleiotropic effects of rtPA on BBB disruption and neurovascular cell death are involved in hemorrhagic transformation of the infarct.35 In the present study, warfarin but not rivaroxaban exacerbated BBB leakage after thrombolysis. The rather widespread effect of warfarin on multiple coagulation factors compared with the selective effect of rivaroxaban may be important for the risk and severity of brain hemorrhage. Beyond their role in the coagulation cascade, coagulation factors function as potent signaling molecules in the neurovascular unit. Reportedly, factor Xa can induce inflammatory signaling via endothelial-based protease-activated receptors 1 and 2, thereby increasing endothelial permeability.36, 37 Rivaroxaban can attenuate this inflammatory process in cultured human atrial cells.38 On the contrary, the vitamin K-dependent factors-activated protein C and Factor VII have protective effects on the BBB, which are mediated by the activation of the endothelial protein C receptor and downregulation of matrix metalloproteinases.39, 40 The activity of both factors was substantially reduced by warfarin but not by rivaroxaban in the present study. Activated protein C can reduce rtPA-induced secondary hemorrhage as well as infarct size in mice.41 Whether these mechanisms underlie the different effect of warfarin and rivaroxaban on secondary intracerebral hemorrhage remains to be elucidated.

Clinicians will be confronted with an increasing number of patients suffering a stroke despite anticoagulation with rivaroxaban.4, 6 Although present and previous data justify withholding thrombolytic treatment in patients with warfarin in the therapeutic INR range, our findings may suggest that hemorrhagic adverse events do not occur more frequently in patients anticoagulated with rivaroxaban than in nonanticoagulated stroke patients. Interestingly, high doses of rivaroxaban17 and dabigatran42 are necessary to increase intracerebral hemorrhage volume in murine intracerebral hemorrhage that is induced by striatal collagenase injection. In contrast, the same high dose (30 mg/kg) of rivaroxaban did not cause excess secondary hemorrhagic transformation after thrombolysis in the present study. Although the underlying molecular mechanisms remain to be elucidated, these data indicate a lower risk of intracranial hemorrhagic complications of new OAC compared with warfarin. Nevertheless, no clinical experience with thrombolysis in rivaroxaban-treated patients has been reported in the literature so far. Thus, we caution against the uncritical transfer of our experimental findings into clinical practice.

Our study has strengths and limitations. Middle cerebral artery occlusion in all experiments caused extensive infarction in the middle cerebral artery territory. Thus, our experiments were designed to predispose the animals to additional hemorrhage due to anticoagulation. Our findings were robust and independent of species (mouse, rat), experimental paradigm (filament versus thromboembolic MCAO), and timing of recanalization (2 versus. 3 hours). A limitation of our study is that only young rodents without significant comorbidities were studied. The different anticoagulation protocol was chosen because of the delayed onset of the anticoagulatory effect of warfarin. In contrast, there is a rapid onset of therapeutic action of rivaroxaban in mice17 and man33 with high plasma concentrations and a prolongation of prothrombin time already 1 hour after oral application.

In conclusion, the present study suggests that pretreatment with rivaroxaban does not increase the risk of secondary hemorrhage after thrombolysis in ischemic stroke. Warfarin and rivaroxaban have profoundly different effects on postischemic blood–brain barrier damage. Cautious evaluation of our findings in the clinical setting is warranted.

The study was investigator-initiated. It was planned and performed independently of Bayer HealthCare AG, except for determination of rivaroxaban plasma concentration, anti-factor Xa activity and prothrombin time, which were measured by SH. The work was supported by an unrestricted grant from Bayer HealthCare AG (Leverkusen, Germany) to RV. RV has been an investigator in the ROCKET-AF and the RELY trials. He has received speaker's honoraria, travel support, and consulting fees from Bayer, Boehringer Ingelheim, and BMS Pfizer. SH is an employee of Bayer Pharma AG. All authors have read and approved the submitted manuscript. The manuscript has not been submitted elsewhere or published elsewhere in whole or in part. The remaining authors declare no conflict of interest.

Footnotes

Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (http://www.nature.com/jcbfm)

The work was supported by an unrestricted grant from Bayer HealthCare AG (Leverkusen, Germany) to RV.

Supplementary Material

Supplementary Information
Click here for additional data file. (483.4KB, pdf)

References

  1. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361:1139–1151. doi: 10.1056/NEJMoa0905561. [DOI] [PubMed] [Google Scholar]
  2. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883–891. doi: 10.1056/NEJMoa1009638. [DOI] [PubMed] [Google Scholar]
  3. Granger CB, Alexander JH, McMurray JJV, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981–992. doi: 10.1056/NEJMoa1107039. [DOI] [PubMed] [Google Scholar]
  4. Alberts MJ, Eikelboom JW, Hankey GJ. Antithrombotic therapy for stroke prevention in non-valvular atrial fibrillation. Lancet Neurol. 2012;11:1066–1081. doi: 10.1016/S1474-4422(12)70258-2. [DOI] [PubMed] [Google Scholar]
  5. Steiner T, Böhm M, Dichgans M, Diener H, Ell C, Endres M, et al. Recommendations for the emergency management of complications associated with the new direct oral anticoagulants (DOACs), apixaban, dabigatran and rivaroxaban. Clin Res Cardiol. 2013;102:399–412. doi: 10.1007/s00392-013-0560-7. [DOI] [PubMed] [Google Scholar]
  6. Diener H, Foerch C, Riess H, Röther J, Schroth G, Weber R. Treatment of acute ischaemic stroke with thrombolysis or thrombectomy in patients receiving anti-thrombotic treatment. Lancet Neurol. 2013;12:677–688. doi: 10.1016/S1474-4422(13)70101-7. [DOI] [PubMed] [Google Scholar]
  7. Adams HP, del Zoppo GJ, Alberts MJ, Bhatt DL, Brass L, Furlan A, et al. Guidelines for the early management of adults with ischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council, Clinical Cardiology Council, Cardiovascular Radiology and Intervention Council, and the Atherosclerotic Peripheral Vascular Disease and Quality of Care Outcomes in Research Interdisciplinary Working Groups: The American Academy of Neurology affirms the value of this guideline as an educational tool for neurologists. Circulation. 2007;115:e478–e534. doi: 10.1161/CIRCULATIONAHA.107.181486. [DOI] [PubMed] [Google Scholar]
  8. Hacke W, Kaste M, Bluhmki E, Brozman M, Dávalos A, Guidetti D, et al. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med. 2008;359:1317–1329. doi: 10.1056/NEJMoa0804656. [DOI] [PubMed] [Google Scholar]
  9. Lees KR, Bluhmki E, Kummer R, von, Brott TG, Toni D, Grotta JC, et al. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet. 2010;375:1695–1703. doi: 10.1016/S0140-6736(10)60491-6. [DOI] [PubMed] [Google Scholar]
  10. Pfeilschifter W, Spitzer D, Pfeilschifter J, Steinmetz H, Foerch C, Borlongan CV. Warfarin anticoagulation exacerbates the risk of hemorrhagic transformation after rt-PA treatment in experimental stroke: therapeutic potential of PCC. PLoS ONE. 2011;6:e26087. doi: 10.1371/journal.pone.0026087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Sun L, Zhou W, Ploen R, Heiland S, Zorn M, Veltkamp R. Rapid reversal of anticoagulation prevents excessive secondary hemorrhage after thrombolysis in a thromboembolic model in rats. Stroke. 2011;42:3524–3529. doi: 10.1161/STROKEAHA.110.608059. [DOI] [PubMed] [Google Scholar]
  12. Ruecker M, Matosevic B, Willeit P, Kirchmayr M, Zangerle A, Knoflach M, et al. Subtherapeutic warfarin therapy entails an increased bleeding risk after stroke thrombolysis. Neurology. 2012;79:31–38. doi: 10.1212/WNL.0b013e31825dcdf0. [DOI] [PubMed] [Google Scholar]
  13. Xian Y, Liang L, Smith EE, Schwamm LH, Reeves MJ, Olson DM, et al. Risks of intracranial hemorrhage among patients with acute ischemic stroke receiving warfarin and treated with intravenous tissue plasminogen activator. J Am Med Ass. 2012;307:2600–2608. doi: 10.1001/jama.2012.6756. [DOI] [PubMed] [Google Scholar]
  14. Rizos T, Horstmann S, Jenetzky E, Spindler M, Gumbinger C, Möhlenbruch M, et al. Oral anticoagulants - a frequent challenge for the emergency management of acute ischemic stroke. Cerebrovasc Dis. 2012;34:411–418. doi: 10.1159/000343655. [DOI] [PubMed] [Google Scholar]
  15. Dempfle C, Hennerici MG. Fibrinolytic treatment of acute ischemic stroke for patients on new oral anticoagulant drugs. Cerebrovasc Dis. 2011;32:616–619. doi: 10.1159/000334579. [DOI] [PubMed] [Google Scholar]
  16. Perzborn E, Strassburger J, Wilmen A, Pohlmann J, Roehrig S, Schlemmer K, et al. In vitro and in vivo studies of the novel antithrombotic agent BAY 59-7939—an oral, direct Factor Xa inhibitor. J Thromb Haemost. 2005;3:514–521. doi: 10.1111/j.1538-7836.2005.01166.x. [DOI] [PubMed] [Google Scholar]
  17. Zhou W, Zorn M, Nawroth P, Bütehorn U, Perzborn E, Heitmeier S, et al. Hemostatic therapy in experimental intracerebral hemorrhage associated with rivaroxaban. Stroke. 2013;44:771–778. doi: 10.1161/STROKEAHA.112.675231. [DOI] [PubMed] [Google Scholar]
  18. Illanes S, Zhou W, Schwarting S, Heiland S, Veltkamp R. Comparative effectiveness of hemostatic therapy in experimental warfarin-associated intracerebral hemorrhage. Stroke. 2011;42:191–195. doi: 10.1161/STROKEAHA.110.593541. [DOI] [PubMed] [Google Scholar]
  19. Rizos T, Herweh C, Jenetzky E, Lichy C, Ringleb PA, Hacke W, et al. Point-of-care international normalized ratio testing accelerates thrombolysis in patients with acute ischemic stroke using oral anticoagulants. Stroke. 2009;40:3547–3551. doi: 10.1161/STROKEAHA.109.562769. [DOI] [PubMed] [Google Scholar]
  20. Samama MM, Contant G, Spiro TE, Perzborn E, Le Flem L, Guinet C, et al. Laboratory assessment of rivaroxaban: a review. Thromb J. 2013;11:11. doi: 10.1186/1477-9560-11-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Longa EZ, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke. 1989;20:84–91. doi: 10.1161/01.str.20.1.84. [DOI] [PubMed] [Google Scholar]
  22. Sun L, Zhou W, Ploen R, Zorn M, Veltkamp R. Anticoagulation with dabigatran does not increase secondary intracerebral haemorrhage after thrombolysis in experimental cerebral ischaemia. Thromb Haemost. 2013;110:153–161. doi: 10.1160/TH12-12-0942. [DOI] [PubMed] [Google Scholar]
  23. Toomey JR, Valocik RE, Koster PF, Gabriel MA, McVey M, Hart TK, et al. Inhibition of factor IX(a) is protective in a rat model of thromboembolic stroke. Stroke. 2002;33:578–585. doi: 10.1161/hs0202.102950. [DOI] [PubMed] [Google Scholar]
  24. Sun L, Zhou W, Heiland S, Marti HH, Veltkamp R. A translationally relevant thromboembolic stroke model for the study of secondary hemorrhage after thrombolysis in rats. Brain Res. 2011;1368:346–354. doi: 10.1016/j.brainres.2010.10.067. [DOI] [PubMed] [Google Scholar]
  25. Choudhri TF, Hoh BL, Solomon RA, Connolly ES, Pinsky DJ. Use of a spectrophotometric hemoglobin assay to objectively quantify intracerebral hemorrhage in mice. Stroke. 1997;28:2296–2302. doi: 10.1161/01.str.28.11.2296. [DOI] [PubMed] [Google Scholar]
  26. Uyama O, Okamura N, Yanase M, Narita M, Kawabata K, Sugita M. Quantitative evaluation of vascular permeability in the gerbil brain after transient ischemia using Evans blue fluorescence. J Cereb Blood Flow Metab. 1988;8:282–284. doi: 10.1038/jcbfm.1988.59. [DOI] [PubMed] [Google Scholar]
  27. Belayev L, Busto R, Zhao W, Ginsberg MD. Quantitative evaluation of blood–brain barrier permeability following middle cerebral artery occlusion in rats. Brain Res. 1996;739:88–96. doi: 10.1016/s0006-8993(96)00815-3. [DOI] [PubMed] [Google Scholar]
  28. Veltkamp R, Siebing DA, Sun L, Heiland S, Bieber K, Marti HH, et al. Hyperbaric oxygen reduces blood–brain barrier damage and edema after transient focal cerebral ischemia. Stroke. 2005;36:1679–1683. doi: 10.1161/01.STR.0000173408.94728.79. [DOI] [PubMed] [Google Scholar]
  29. Pfeilschifter W, Bohmann F, Baumgarten P, Mittelbronn M, Pfeilschifter J, Lindhoff-Last E, et al. Thrombolysis with recombinant tissue plasminogen activator under dabigatran anticoagulation in experimental stroke. Ann Neurol. 2012;71:624–633. doi: 10.1002/ana.23558. [DOI] [PubMed] [Google Scholar]
  30. Bohmann F, Mirceska A, Pfeilschifter J, Lindhoff-Last E, Steinmetz H, Foerch C, et al. No influence of dabigatran anticoagulation on hemorrhagic transformation in an experimental model of ischemic stroke. PLoS ONE. 2012;7:e40804. doi: 10.1371/journal.pone.0040804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gliem M, Hermsen D, van Rooijen N, Hartung H, Jander S. Secondary intracerebral hemorrhage due to early initiation of oral anticoagulation after ischemic stroke: an experimental study in mice. Stroke. 2012;43:3352–3357. doi: 10.1161/STROKEAHA.112.666818. [DOI] [PubMed] [Google Scholar]
  32. Mueck W, Lensing AW, Agnelli G, Decousus H, Prandoni P, Misselwitz F. Rivaroxaban: population pharmacokinetic analyses in patients treated for acute deep-vein thrombosis and exposure simulations in patients with atrial fibrillation treated for stroke prevention. Clin Pharmacokinet. 2011;50:675–686. doi: 10.2165/11595320-000000000-00000. [DOI] [PubMed] [Google Scholar]
  33. Kubitza D, Becka M, Voith B, Zuehlsdorf M, Wensing G. Safety, pharmacodynamics, and pharmacokinetics of single doses of BAY 59-7939, an oral, direct factor Xa inhibitor. Clin Pharmacol Ther. 2005;78:412–421. doi: 10.1016/j.clpt.2005.06.011. [DOI] [PubMed] [Google Scholar]
  34. Harenberg J, Marx S, Weiss C, Krämer R, Samama MM, Schulman S. Report of the Subcommittee of Control of Anticoagulation on the determination of the anticoagulant effects of rivaroxaban. J Thromb Haemost. 2012;10:1433–1436. doi: 10.1111/j.1538-7836.2012.04784.x. [DOI] [PubMed] [Google Scholar]
  35. Wang X, Tsuji K, Lee S, Ning M, Furie KL, Buchan AM, et al. Mechanisms of hemorrhagic transformation after tissue plasminogen activator reperfusion therapy for ischemic stroke. Stroke. 2004;35:2726–2730. doi: 10.1161/01.STR.0000143219.16695.af. [DOI] [PubMed] [Google Scholar]
  36. McLean K, Schirm S, Johns A, Morser J, Light DR. FXa-induced responses in vascular wall cells are PAR-mediated and inhibited by ZK-807834. Thromb Res. 2001;103:281–297. doi: 10.1016/s0049-3848(01)00330-9. [DOI] [PubMed] [Google Scholar]
  37. Ossovskaya VS, Bunnett NW. Protease-activated receptors: contribution to physiology and disease. Physiol Rev. 2004;84:579–621. doi: 10.1152/physrev.00028.2003. [DOI] [PubMed] [Google Scholar]
  38. Bukowska A, Zacharias I, Weinert S, Skopp K, Hartmann C, Huth C, et al. Coagulation factor Xa induces an inflammatory signalling by activation of protease-activated receptors in human atrial tissue. Eur J Pharmacol. 2013;718:114–123. doi: 10.1016/j.ejphar.2013.09.006. [DOI] [PubMed] [Google Scholar]
  39. Zlokovic BV, Griffin JH. Cytoprotective protein C pathways and implications for stroke and neurological disorders. Trends Neurosci. 2011;34:198–209. doi: 10.1016/j.tins.2011.01.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sen P, Gopalakrishnan R, Kothari H, Keshava S, Clark CA, Esmon CT, et al. Factor VIIa bound to endothelial cell protein C receptor activates protease activated receptor-1 and mediates cell signaling and barrier protection. Blood. 2011;117:3199–3208. doi: 10.1182/blood-2010-09-310706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Cheng T, Petraglia AL, Li Z, Thiyagarajan M, Zhong Z, Wu Z, et al. Activated protein C inhibits tissue plasminogen activator-induced brain hemorrhage. Nat Med. 2006;12:1278–1285. doi: 10.1038/nm1498. [DOI] [PubMed] [Google Scholar]
  42. Zhou W, Schwarting S, Illanes S, Liesz A, Middelhoff M, Zorn M, et al. Hemostatic therapy in experimental intracerebral hemorrhage associated with the direct thrombin inhibitor dabigatran. Stroke. 2011;42:3594–3599. doi: 10.1161/STROKEAHA.111.624650. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
Click here for additional data file. (483.4KB, pdf)

Articles from Journal of Cerebral Blood Flow & Metabolism are provided here courtesy of SAGE Publications

RESOURCES