Apixaban Versus Warfarin for Mechanical Heart Valve Thromboprophylaxis in a Swine Aortic Heterotopic Valve ModelHighlights
Objective—Warfarin is the current standard for oral anticoagulation therapy in patients with mechanical heart valves, yet optimal therapy to maximize anticoagulation and minimize bleeding complications requires routine coagulation monitoring, possible dietary restrictions, and drug interaction monitoring. As alternatives to warfarin, oral direct acting factor Xa inhibitors are currently approved for the prophylaxis and treatment of venous thromboembolism and reduction of stroke and systemic embolization. However, no in vivo preclinical or clinical studies have been performed directly comparing oral factor Xa inhibitors such as apixaban to warfarin, the current standard of therapy.
Approach and Results—A well-documented heterotopic aortic valve porcine model was used to test the hypothesis that apixaban has comparable efficacy to warfarin for thromboprophylaxis of mechanical heart valves. Sixteen swine were implanted with a bileaflet mechanical aortic valve that bypassed the ligated descending thoracic aorta. Animals were randomized to 4 groups: control (no anticoagulation; n=4), apixaban oral 1 mg/kg twice a day (n=5), warfarin oral 0.04 to 0.08 mg/kg daily (international normalized ratio 2–3; n=3), and apixaban infusion (n=4). Postmortem valve thrombus was measured 30 days post-surgery for control-oral groups and 14 days post-surgery for the apixaban infusion group. Control thrombus weight (mean) was significantly different (1422.9 mg) compared with apixaban oral (357.5 mg), warfarin (247.1 mg), and apixiban 14-day infusion (61.1 mg; P<0.05).
Conclusions—Apixaban is a promising candidate and may be a useful alternative to warfarin for thromboprophylaxis of mechanical heart valves. Unlike warfarin, no adverse bleeding events were observed in any apixaban groups.
In middle-aged adults, the decision to surgically implant a mechanical or a bioprosthetic valve balances the risk of bleeding, mainly associated with mechanical valves and requirement for long-term anticoagulation, with the risk of reoperation secondary to structural valve deterioration as reported with bioprosthetic valves.1 Because of the associated risk of thromboembolic events and stroke, the American College of Chest Physicians Evidence-Based Clinical Practice Guidelines recommend lifelong anticoagulation for patients with mechanical heart valves versus no long-term anticoagulation.2 Although low-molecular-weight heparin can be used to achieve anticoagulation, the vitamin K antagonist, warfarin, is currently the only oral anticoagulant recommended for patients with aortic and mitral mechanical heart valves. Moreover, achieving optimal anticoagulation with warfarin while minimizing the risk of major bleeding is challenged by the need for frequent monitoring and targeting of the international normalized ratio (INR), genetic polymorphism alterations in warfarin metabolism, multiple food and drug interactions, comorbid medical conditions, and age. In a recent study involving 546 patients treated with warfarin using targeted INR therapy after receiving aortic and mitral valves, the incidence of major bleeding was 4.4 and 4.6 per 100 patent-years, respectively.3 As alternatives to warfarin, multiple novel oral anticoagulants have been developed and are currently approved to reduce the risk of stroke, systemic embolism, and thromboembolic events secondary to atrial fibrillation and surgery, respectively, and for the treatment of venous thromboembolism.4–7
See accompanying editorial on page 743
Because of the risk of bleeding and challenges associated with warfarin management, oral alternatives to vitamin K antagonists for mechanical valve anticoagulation are needed. To date, only dabigatran, a direct thrombin inhibitor, has been evaluated and compared with the standard of care, warfarin, in human patients with mechanical aortic and mitral valves. The RE-ALIGN study (Randomized Phase II Study to Evaluate the Safety and Pharmacokinetics of Dabigatran Etexilate Oral in Patients after Heart Valve Replacement) was halted early, as dabigatran, although adjusted to a trough plasma level >50 ng/mL, was associated with an increased rate of thromboembolic events, valve thrombosis, and major bleeding complications compared with warfarin.8 Unlike direct thrombin inhibitors, factor Xa inhibitors have not been tested in a clinical setting for mechanical heart valve thromboprophylaxis. On the basis of its earlier sequence in the coagulation cascade, ability to greatly trigger the generation of thrombin, by 1000-fold, and limited platelet activation by factor Xa compared with thrombin, factor Xa may be a preferred anticoagulation target compared with factor IIa.9 However, clinical trials of oral factor Xa inhibitors compared directly to factor IIa inhibitors regarding thromboprophylaxis efficacy have not been performed.
Swine are commonly used animal models to research thromboembolism because of their anatomic, hematologic, and coagulation similarities to humans.10 McKellar et al11 developed a heterotopic aortic valve replacement model in swine to evaluate thromboprophylaxis in mechanical heart valves. Using this model, previous studies have evaluated the thromboprophylaxis of oral dabigatran etexilate, a direct thrombin inhibitor, and oral rivaroxaban, a factor Xa inhibitor, with results demonstrating equal to superior thromboprophylaxis compared with enoxaparin, a low-molecular-weight heparin.12,13 However, there are no preclinical studies using a heterotopic aortic valve replacement model in swine comparing the thromboprophylaxis efficacy of a factor Xa inhibitor to warfarin, the mainstay of oral anticoagulant therapy used in humans with mechanical heart valves.
Apixaban, a factor Xa inhibitor, is currently approved by the US Food and Drug Administration for the treatment of deep vein thrombosis, for the prevention of thromboembolism after knee and hip replacement surgery, and to reduce the risk of systemic embolism and stroke in patients with nonvalvular atrial fibrillation. Apixaban demonstrates >30 000-fold selectivity for factor Xa compared with other human coagulation proteases with an inhibitory constant of 0.08 nmol/L for human factor Xa.14 It produces a rapid onset of inhibition of factor Xa with association rate constant of ≈20 μM−1·s−1 and inhibits free and prothrombinase and clot-bound factor Xa activity in vitro. In addition, apixaban inhibits factor Xa from rabbits, rats, and dogs, an activity that parallels its antithrombotic potency in these species.14 In humans, apixaban oral Tmax is 3 hours, and its half-life is ≈12 hours (8–15 hours). Routine monitoring is generally not advocated during apixaban therapy, the ideal administration is twice a day orally, and in humans, it is predominantly eliminated by nonrenal excretion (73% fecal and 27% renal).15
In this study, we tested the thromboprophylactic efficacy and incidence of major bleeding of apixaban compared with warfarin in a well-established aortic heterotopic valve swine model. On the basis of results in other preclinical models, we anticipated that the half-life of apixaban might be considerably shorter in swine than in humans. We, therefore, evaluated apixaban not only by oral dosing but also in a separate group of animals in which the concentration versus time profile approximated that observed in humans.
Materials and Methods
Materials and Methods are available in the online-only Data Supplement. In summary, swine were randomized to the following groups: no anticoagulation-control, oral apixaban, oral warfarin, and apixaban continuous intravenous infusion. Animals were implanted with a bileaflet mechanical aortic valve that bypassed the ligated descending thoracic aorta. End points were as follows: no anticoagulation=30 days, oral apixaban and oral warfarin=30 days, and apixaban infusion=14 days. The apixaban infusion was modeled by correlating human clinical trial plasma concentrations to swine apixaban pharmacokinetic parameters. Swine administered warfarin were monitored via prothrombin time and INR (2–3) via a mechanical clot-detection system fibrometer and thromboplastin (International Sensitivity Index=0.92). Valves were harvested at group-specific end points. Valve weights were measured and compared for statistical significance.
Pharmacokinetics: Apixaban Intravenous Bolus
Raw plasma concentration per time data are outlined in Table 1. The curvilinear plasma concentration versus time was determined (Figure 1). Noncompartmental pharmacokinetic parameters (mean±SD) for intravenous 0.5 mg/kg bolus were as follows: t1/2=6.2±3.8 hours, area under the curve=6186.67±4981.81 ng·h/mL, clearance=118.00±72.75 mL·h−1·kg−1, volume of distribution=577.27±113.05 mL/kg, area under the first moment curve=88 048±62 619.36 ng·h2/mL, mean residual time=6.5±4.34 hours (Table 2).
Pharmacokinetics: Apixaban Oral Bolus
Raw plasma concentration per time data are outlined in Table 3. Noncompartmental pharmacokinetic parameters (mean±SD) for oral 0.5 mg/kg bolus were as follows: t1/2=5.93±1.37 hours, area under the curve=1390±564.02 ng·h/mL, clearance=397.67±144.89 mL·h−1·kg−1, area under the first moment curve=9286.33±2806.98 ng·h2/mL, mean residual time=6.93±0.87, Cmax=214.67±91.51, Tmax=1.58±0.72 hours, bioavailability=31.87±20.36% (Table 4). One animal (swine 5) developed a catheter-related complication during the crossover washout period, was removed from the study, and replaced with another animal (swine 6).
Anti-Xa Low-Molecular-Weight Heparin Anticoagulation Measurement and International Normalized Ratio
The curvilinear anti-Xa low-molecular-weight heparin (LMWH) activity versus time was determined for single intravenous apixaban 0.5 mg/kg bolus and single orally administered apixaban 0.5 mg/kg dose. Anti-Xa LMWH activity for early intravenous time points (≤2 hours) were associated with the upper limit of the LMWH calibrator (1.3 IU/mL). Both intravenous and oral apixaban exceeded the suggested lower therapeutic range for prophylaxis (0.5–1.2 IU/mL) at 4 hours up to ≈8 hours for each route of administration (Figure 2). Plasma apixaban ranges significantly predicted anti-Xa LMWH activity for both intravenous and orally administered apixaban (R2=0.8; F=27.48; P=0.0012 and R2=0.84; F=126.1; P<0.0001), respectively (Figure 3). Plasma samples >300 ng/mL were associated with the upper limit of LMWH calibrator (anti-Xa LMWH=1.3 IU/mL).
Swine administered warfarin were monitored via INR (Figure 4).
Valve Thrombus Weight and Postmortem Valve Evaluation
All control animals (no anticoagulation) had visible evidence of thrombus at 30 days postimplantation. Control valve thrombus weight (mean±SD) was significantly different (1422±676.4 mg) compared with apixaban 1 mg/kg twice a day oral (357.5±234.9 mg), warfarin (247.1±134.3 mg), and apixiban mulitstep 14-day infusion (61.1±47.2 mg; ANOVA F= 10.88; P=0.001; Figure 5; Table 5). Three out of five swine within the apixaban oral group had evidence of clot. Two out of three swine within the warfarin group had evidence of clot. No animals in the apixaban mulitstep 14-day infusion group had evidence of clot. One animal within the apixaban infusion group had evidence of a film on the bileaflet mechanical valve at the time of harvest but no discernible thrombus. The film weight was included in the analysis. Histopathology demonstrated the structure to be composed almost exclusively of fibrin consistent with an early loosely organized thrombus with scattered regions of mineralization. There was no evidence of a biofilm or bacterial colonization. Photos representing postmortem bileaflet mechanical valve for all groups including control, apixaban oral, warfarin, and apixaban infusion are shown in Figure 6.
There was no evidence of thromboembolic events, severe or complete valve or graft thrombosis, or hemorrhage in the apixaban treatment groups. The majority of swine in the warfarin treatment group remained within a therapeutic INR without complications, with the exception of 2 animals that were removed from study because of major bleeding complications including pulmonary hemorrhage, hemothorax, and mutlifocal subcutaneous hemorrhage. A mechanical clot-detection system (fibrometer) was used with a low International Sensitivity Index thromboplastin, which may have improved accuracy in monitoring INR values, thus reducing warfarin-associated bleeding complications.16
There was no consistent pathological change noted across animals of all groups with tissues provided. Thrombosis was not observed, and hemorrhage was not a significant feature of any of the animals. Minimal to mild tubular basophilic area was noted in the kidneys of one animal within the apixaban oral treatment group and one animal in the apixaban infusion group. One animal in the warfarin oral group had a fibrinopurulent pleuritis. Minimal lymphocytic infiltrates were noted in the coronary artery adventitia in one animal in the apixaban oral treatment group and were considered a background change in this animal.
The goal of this study was to determine the preclinical efficacy of apixaban regarding thromboprophylaxis of mechanical heart valves in a swine surgical model. Swine were chosen for this model because of their similar coagulation properties and cardiovascular anatomy and physiology compared with humans. The study was designed to provide a comparison to current standard therapy, warfarin, for anticoagulation and stroke prevention in human patients implanted with a mechanical heart valve. Results from this study demonstrated that apixaban has comparable efficacy to warfarin in reducing mechanical valve thrombosis in swine. This in vivo preclinical study demonstrates the thromboprophylactic efficacy of apixaban in direct comparison to warfarin for mechanical heart valves.
Apixaban has a high degree of selectivity for factor Xa, no active metabolites, a predictable dose response, and pharmacokinetic profile with limited renal excretion, minimal drug and food interactions, and a reduced need for titration and therapeutic monitoring, making it a favorable alternative to warfarin for the prevention of stroke in patients with atrial fibrillation.15 Its bioavailability in humans is ≈50%, which is greater than the mean bioavailability of 32% seen in swine with the current study. The half-life of apixaban in humans is ≈12 hours versus a 1.6-hour oral half-life in swine.
In the current study, an oral 0.5 mg/kg dose in swine was associated with anti-Xa LMWH plasma levels >0.6 IU/mL at 4 hours. However, we chose to use a higher dose (1 mg/kg) to ensure adequate anti-Xa activity when using twice a day oral dosing.
The use of orally administered direct factor Xa inhibitors for mechanical valve thromboembolic prophylaxis is promising especially when compared with the current standard of care, warfarin, which is associated with bleeding complications, frequent anticoagulation monitoring and titration, dietary restrictions, and drug interactions including associated healthcare costs for testing, monitoring, and treatment for complications. As a result, the decision to use a bioprosthetic valve with a risk of premature valve failure must be weighed versus lifelong warfarin therapy recommended in patients implanted with mechanical valves.
Dabigatran, an oral direct thrombin inhibitor, demonstrated favorable thrombotic reduction in a porcine heterotopic aortic mechanical valve model.12 In that study, dabigatran was as effective as enoxaparin dosed at 2.5 mg/kg SC BID (goal to maintain anti-Xa levels >0.6 IU/mL at 4 hours after administration) in preventing valve thrombosis compared with the no anticoagulation group.12 Dabigatran was also compared with warfarin in the RE-ALIGN study in 2 groups of patients implanted with aortic or mitral mechanical heart valves (implantation within 7 days or after 3 months).8 In the RE-ALIGN trial, dabigatran dosing was based on previous doses used in patients with atrial fibrillation to maintain a trough level >50 ng/mL, whereas warfarin was dosed to an INR (2–3 or 2.5–3.5) based on thromboembolic risk. The study was halted early because of a higher incidence of thromboembolic ischemia or stroke (5% patients receiving dabigatran versus no patients receiving warfarin) and major bleeding events (dabigatran 4% versus warfarin 2% in patients receiving dabigatran within 7 days of surgery).8 Of interest, the majority of major bleeding occurred in the immediate postsurgical period, which is generally characterized by inflammation, enhanced thrombogenicity, and platelet reactivity. In the RE-ALIGN trial, the authors hypothesized that the higher incidence of thromboembolic complications was possibly because of activation of the coagulation cascade secondary to tissue factor released during surgery and activation of the intrinsic pathway via contact with the sewing ring and valve leaflets of mechanical heart valves. In this setting, warfarin may have been more efficacious as it targets multiple pathways of the clotting cascade versus dabigatran, which is limited to thrombin inhibition.
Because factor Xa is an initial component of the coagulation common pathway where both the extrinsic and contact activation pathways converge, inhibition of factor Xa inhibits both coagulation pathways and the downstream exponential amplification of thrombin.17 Furthermore, in vitro and in vivo animal studies have demonstrated hypercoagulability and thrombogenesis with subtherapeutic doses of direct thrombin inhibitors versus direct Xa inhibitors, resulting from inactivation of the protein kinase C system.18–20 Thus, factor Xa inhibitors such as apixaban may provide a greater reduction in mechanical heart valve thrombosis compared with direct thrombin inhibitors such as dabigatran with minimal antagonism of existing thrombin needed to maintain hemostasis via platelet thrombin receptor interactions. Theoretically, this may reduce major bleeding in the immediate postoperative period. In addition, the favorable pharmacokinetic and clearance profile of apixaban may lend to less permutations in plasma levels with increased anticoagulation stability. In these efforts, the apixaban infusion group dose was modeled after human pharmacokinetic data based on previous studies and the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation study (ARISTOTLE).6,21 In the apixaban infusion group, swine apixaban pharmacokinetic data were used to mimic the human 5 mg PO BID area under the curve values. Total thrombus weights in the apixaban infusion group tended to be smaller than both the apixaban and warfarin anticoagulation groups possibly indicating more stable anticoagulation because of minimization of peak to trough plasma concentrations. One animal in the infusion group demonstrated evidence of a visible fibrin film on the mechanical valve leaflets at the 14-day postmortem end point.
One limitation of the study is translating and determining optimal dosing based on previous studies and indications for venous thrombosis and prevention of embolic disease and stroke. Doses of factor Xa or direct thrombin inhibitors with confirmed efficacy for low flow or stasis conditions may not correlate with similar efficacy when used in high flow and high shear environments associated with mechanical heart valves. In addition, the acute postoperative period is generally an inflammatory and prothrombotic environment characterized by a high degree of platelet activation. As a result, additional preclinical and appropriate controlled trials evaluating the use of a factor Xa inhibitor such as apixaban possibly in combination with an antiplatelet agent are needed to elucidate optimal dosing to minimize both thromboembolic complications and incidences of major bleeding. Additional limitations to the study were small sample sizes mainly in part to complexity associated with this research model, which limited power to establish noninferiority of apixaban compared with warfarin for thromboprophylaxis efficacy. However, despite the small sample size, we demonstrated significant differences in thrombus deposition between control and treatment groups, emphasizing the anticoagulant efficacy of apixaban. In addition, we used a short-term 30-day time point to measure thromboprophylaxis based on previous publications11–13,22 and used a 14-day infusion time period to focus on the immediate postsurgical period in which the risk of thromboembolism would be greatest. With regard to major bleeding complications, intracranial hemorrhage is of great concern in human patients receiving anticoagulation. Postmortem evaluation of the central nervous system to completely confirm absence of intracranial hemorrhage was not performed as the study did not demonstrate clinical evidence of major bleeding or neurological signs within the apixaban groups. Two animals in the warfarin group were removed from study because of major bleeding complications including pulmonary hemorrhage, hemothorax, and mutlifocal subcutaneous hemorrhage (bleeding complication animal 1, last INR 3.85 on day 6 post-surgery; bleeding complication animal 2, last INR 5.35 on day 15 post-surgery).
This study was not powered to test for noninferiority, yet apixaban demonstrated promising efficacy comparable to oral warfarin for preventing short-term arterial thromboembolism in swine implanted with a heterotopic mechanical heart valve. Swine that received apixaban via infusion modeled after human clinical trial pharmacokinetic data for the prevention of systemic emboli had the smallest thrombus weights compared with all groups. Although the duration of the infusion was only 14 days compared with 30 days for the oral apixaban and warfarin groups, the smaller thrombus weights seem to correlate with apixaban’s rapid in vivo activity and predictable pharmacokinetic profile.
Compared with warfarin, swine in the apixaban groups did not demonstrate adverse bleeding complications. This preclinical efficacy study directly compares the anti-Xa inhibitor, apixaban, to warfarin, the current standard of therapy for the prevention of embolic complications in human patients with mechanical heart valves. These data provide support for additional dose–response studies powered to directly compare thromboprophylaxis efficacy of apixaban to warfarin in preclinical mechanical valve models. Data from such studies may provide the foundation for future clinical studies evaluating apixaban as an alternative to warfarin in select patients with mechanical heart valves.
We would like to thank the following individuals for the participation and assistance with this study: Drs Bo Wen, Ting Zhao, and Duxin Sun, Department of Pharmaceutical Sciences, College of Pharmacy, University of Michigan (NIH-NCI P30CA046592) for assistance with pharmacokinetic assays and Terry Majors for assistance with coagulation analysis. Dr Thomas R. Meier for assistance with development of a swine jacketed-tether infusion system. Lorie Gavulic for her medical illustrations. Finally, we would like to thank the Unit for Laboratory Animal Medicine Animal Care Supervisors, Scot A. Pittman and Michael J Ream and Veterinary Technicians, Lisa A. Burlingame and Laura B. Durham for their expert technical assistance and supportive veterinary care during the course of this study.
Sources of Funding
This research was supported with funding from Bristol-Myers Squibb and the Conrad Jobst Foundation through an industry-academic alliance with the University of Michigan. Bristol-Myers Squibb colleagues were not involved in either data collection or analysis.
Dr Diaz is member of the Board of Directors of the American Venous Forum. The other authors report no conflicts.
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.116.308649/-/DC1.
- Nonstandard Abbreviations and Acronyms
- international normalized ratio
- low-molecular-weight heparin
- Received October 24, 2016.
- Accepted February 6, 2017.
- © 2017 American Heart Association, Inc.
- Zhao DF,
- Seco M,
- Wu JJ,
- Edelman JB,
- Wilson MK,
- Vallely MP,
- Byrom MJ,
- Bannon PG
- Whitlock RP,
- Sun JC,
- Fremes SE,
- Rubens FD,
- Teoh KH
- Labaf A,
- Grzymala-Lubanski B,
- Stagmo M,
- Lövdahl S,
- Wieloch M,
- Själander A,
- Svensson PJ
- Greiten LE,
- McKellar SH,
- Rysavy J,
- Schaff HV
- Hanna MS,
- Mohan P,
- Knabb R,
- Gupta E,
- Frost C,
- Lawrence JH
- Furugohri T,
- Sugiyama N,
- Morishima Y,
- Shibano T
- Lopes RD,
- Alexander JH,
- Al-Khatib SM,
- Ansell J,
- Diaz R,
- Easton JD,
- Gersh BJ,
- Granger CB,
- Hanna M,
- Horowitz J,
- Hylek EM,
- McMurray JJ,
- Verheugt FW,
- Wallentin L
- McKellar SH,
- Thompson JL 3rd.,
- Garcia-Rinaldi RF,
- Macdonald RJ,
- Sundt TM 3rd.,
- Schaff HV
Despite small sample size, apixaban demonstrated similar significant reductions in mechanical valve thrombus weights, as oral warfarin, in swine implanted with a heterotopic mechanical heart valve. Compared with warfarin, swine in the apixaban groups showed no evidence of bleeding complications.
Apixaban is a promising candidate and may be a useful alternative to warfarin for thromboprophylaxis of mechanical heart valves.
Human plasma concentrations from apixaban clinical trials were used to develop a novel targeted infusion model for swine based on comparable pharmacokinetic parameters. Additional dose–response studies are warranted to fully elucidate the thromboprophylaxis efficacy of apixaban.