Anticoagulation therapy is mandatory in patients with pulmonary embolism to prevent significant morbidity and mortality. The mainstay of therapy has been vitamin-K antagonist therapy bridged with parenteral anticoagulants. The recent approval of new oral anticoagulants (NOACs: apixaban, dabigatran, and rivaroxaban) has generated significant interest in their role in managing venous thromboembolism, especially pulmonary embolism due to their improved pharmacokinetic and pharmacodynamic profiles, predictable anticoagulant response, and lack of required efficacy monitoring. This paper addresses the available literature, on-going clinical trials, highlights critical points, and discusses potential advantages and disadvantages of the new oral anticoagulants in patients with pulmonary embolism.
Pulmonary embolism (PE) is a common, potentially fatal disease with an annual incidence of approximately 70 cases per 100 000 people [
The mainstay of treatment for the past fifty years has been warfarin therapy, overlapped with a parenteral anticoagulant until the vitamin K antagonist (VKA) is fully therapeutic [
Pulmonary embolism is a serious and potentially fatal complication of a venous thrombotic event (VTE). Despite clinical trials yielding similar estimates for safety and efficacy in overall treatment when comparing deep vein thrombosis (DVT) and PE, patients with PE incur additional risks not seen in patients with DVT alone. The impact of PE on mortality is striking, with significant increases in the risk of death at both 30 and 90 days after event [
Three new oral anticoagulants (apixaban, dabigatran, and rivaroxaban) are the newest to enter the anticoagulant armamentarium. The available data on each agent in the treatment of PE, their respective mechanisms of action, pharmacokinetic profiles, and potential places in therapy are discussed.
Each of the NOACs has unique pharmacokinetic and pharmacodynamic properties that clinicians must consider when selecting oral anticoagulant therapy. The comparative pharmacokinetic profiles for the NOACs in respect to warfarin are provided in Table
Comparison of the pharmacokinetic and pharmacodynamic features of oral anticoagulant therapies [
Characteristic | Warfarin | Apixaban | Dabigatran | Rivaroxaban |
---|---|---|---|---|
Mechanism of action | VKORC1 enzyme inhibitor | Direct |
Direct |
Direct |
| ||||
Prodrug | No | No | Yes | No |
| ||||
Approved indications | VTE prevention |
VTE prevention |
VTE prevention |
VTE prevention |
| ||||
FDA black box warnings | Agent can cause major or fatal bleeding; Regular INR monitoring is necessary; drugs, dietary changes, and other factors affect INR levels (FDA) | Discontinuation in patients without adequate continuous anticoagulation increases the risk of stroke (FDA) | None to date | Discontinuing places patients at an increased risk of thrombotic events; Risk of spinal/epidural hematoma during neuraxial anesthesia/spinal puncture (FDA) |
| ||||
Dosing | Variable, patient specific | Fixed, twice daily |
Fixed, twice daily |
|
| ||||
|
4 | 1–3 | 1–3 | 2–4 |
| ||||
Half-life (h) | 20–60 | 12 | 12–17 | 5–9 |
| ||||
Bioavailability (F) | 100% | 50% | 6% |
10 mg: 80–100% |
| ||||
Renal Clearance | >90% as inactive metabolites | 25% | 80%* | 36% unchanged drug* |
| ||||
CYP metabolism+ | >90% |
15% |
No | 30% |
| ||||
P-glycoprotein (P-gp)+ | No | Yes | Yes | Yes |
| ||||
Dietary considerations | Dietary vitamin K | None | None |
|
| ||||
Influence on routine coagulation assay | ||||
Protime (PT) |
|
|
|
|
| ||||
aPTT | No/ |
|
|
|
| ||||
Thrombin time (TT) | No | No |
|
No |
| ||||
Coagulation assay |
INR (Protime) | None | None | None |
| ||||
Clinically validated |
Vitamin K, FFP, PCC, Factor VIIa, aPCC | None | None | None |
EMA: European Medicines Agency; HC: Health Canada; FDA: U.S. Food and Drug Administration
AF: Atrial Fibrillation; CYP: cytochrome P450; VKORC1: C1 subunit of the vitamin K epoxide reductase enzyme; FFP: Fresh Frozen Plasma; PCC: Prothrombin Complex Concentrate.
#Rivaroxaban 15 mg tablet, if available, should be taken with food.
*Dose adjustment for level of renal dysfunction required.
+Potential source for drug-drug interactions—review full package insert for details.
An equally important point is the dosing schedules of the NOACs range from once to twice daily, despite relatively short half-lives. This makes compliance, appropriate counseling and monitoring imperative components to successful therapy. Although the NOACs do not have the dietary restrictions of warfarin, an important caveat for rivaroxaban is that higher doses (i.e., >10 mg/day) need to be given with food to enhance bioavailability. Additionally, the rivaroxaban package insert recommends avoiding administration via feeding tube due to concerns of decreased bioavailability. While there are no dietary restrictions with dabigatran, the dosage form cannot be crushed or chewed as bioavailability increases well above therapeutic levels. Dabigatran must also be stored in the original container to prevent capsule degradation, with the medication expiring 4 months after opening of the original container. Dabigatran therefore cannot be placed in a pill box or organizer because of its hygroscopic tendencies.
The NOACs have fewer drug interactions than warfarin; however vigilance with patient’s medication profiles and concomitant medications is necessary as the NOACs are subject to drug-drug interactions via the cytochrome p450 and p-glycoprotein systems (Table
Selected P-glycoprotein and Cytochrome P450 3A4 drug interactions with NOAC based on current FDA-approved indications [
Medications+ | Mechanism of Interaction | Dabigatran# |
Rivaroxaban |
Apixaban |
---|---|---|---|---|
Rifampin | P-glycoprotein inducer | Avoid combination | Avoid | Avoid |
| ||||
Carbamazepine, phenytoin, |
P-glycoprotein inducers and |
Avoid | Avoid | Avoid |
| ||||
Dronedarone | P-glycoprotein inhibitors |
|
Not addressed in Package Insert | Not addressed |
| ||||
Systemic ketoconazole Itraconazole | P-glycoprotein inhibitors and |
|
Avoid | Reduce dose in 1/2 for those starting on the full dose of 5 mg BID% |
| ||||
lopinavir/ritonavir, ritonavir, |
P-glycoprotein inhibitors and |
Not addressed in Package Insert | Avoid | Reduce dose in |
| ||||
Verapamil, amiodarone, quinidine, | P-glycoprotein inhibitors |
|
Not addressed in Package Insert | Reduce dose in |
| ||||
Clarithromycin | P-glycoprotein inducers and strong 3A4 inducers |
|
Unclear: refer to package insert. | Reduce dose in |
CrCl: creatinine clearance as determined by Crockcoft-Gault equation.
#Dabigatran etexilate (prodrug) uses the p-glycoprotein transport system. It is not a substrate, inducer, or inhibitor of the cytochrome p450 system.
+Listed medications are representatives of potent 3A4 inhibitors and may not be comprehensive.
%Those patients who need to start at 2.5 mg BID should avoid this combination.
Lastly, the varying degree of renal elimination and altered response in patients with evidence of renal dysfunction adds to the clinical complexity for the appropriate patient selection. Dabigatran, rivaroxaban, and apixaban have FDA approved dosing adjustments for patients with nonvalvular AF and who have compromised renal function. For dabigatran and rivaroxaban, these renal dosing recommendations, while FDA-approved, are based upon pharmacokinetic studies and have not been studied clinically. When apixaban was studied, the clinical trials took into account the patients level of renal function and so the dose adjustment in the package insert is based on validated data from the clinical trial. Therefore, the FDA approved dose for apixiban is to reduce the dose in patients with two of the following characteristics: age ≥80 years, body weight <60 kg or serum creatinine ≥1.5 mg/dL (Table
The predictable pharmacokinetic profile of the newer NOACs results in less variability in drug response thereby negating the need for individualized therapeutic monitoring as required with VKA. Although the newer agents can alter coagulation assays (Table
The lack of a reversal agent for the NOACs remains one of the most important clinical concerns with these new agents. As the NOACs pharmacologically do not induce clotting factor deficiencies as a mechanism of action, the utility of replacement products such as fresh frozen plasma (FFP) is limited. Since the NOACs represent inhibitors of the coagulation cascade, reversal will remain a challenge until an antidote is developed and tested clinically.
Currently, only summary recommendations regarding the management of bleeding or “reversal” of the NOACs can be made. Although there are published studies, many are ex vivo or animal studies that cannot reliably predict an in-vivo response and therefore recommendations remain inconclusive. In vitro publications in humans are sparse and primarily limited to case reports [
A nonactivated 4-factor PCC product (Cofact) failed to show reversal in coagulation parameters induced by dabigatran in a randomized, placebo-controlled, crossover study in healthy subjects. In this study, bleeding was not an accessible outcome [
In regard to rivaroxaban, dialysis is not a viable option due to high protein binding (92–95%) [
No data or recommendations currently exist for reversal of apixaban therapy. To complicate translation of this research into clinical practice, nonactivated 4-factor PCCs are unavailable in the United States, and the only 4-factor PCC available is the activated FEIBA product. Institutions are therefore left with the development of institution-specific policies for reversal based upon the foundation of supportive care and local measures to stop bleeding, the availability of the local medical expertise, an understanding of available literature and reversal agents, and cost.
Limited data exist to determine the safety and efficacy of apixaban for the treatment of PE. The AMPLIFY (NCT00643201) trial is a Phase III, randomized, double-blind, double-dummy, noninferiority trial ongoing to assess apixaban in the initial treatment of VTE [
AMPLIFY-EXT is a randomized, double-blind study evaluating the safety and efficacy of apixaban for the extended treatment of DVT and PE over a 12-month period following the initial 6–12 months of anticoagulation therapy [
The majority of patients on apixaban in AMPLIFY-EXT weighed greater than 60 kg (92.6%). Serum creatinines (for dose adjustment, per the package insert) were not provided; though the majority of patients on apixaban (99.8%) had a calculated creatinine clearance greater than 30 mL/min, 91.8% had a calculated creatinine clearance greater than 50 mL/min. Of the subjects on apixaban, 98.3% completed 6–12 months of prior anticoagulation therapy, as opposed to less than 6 months and more than 12 months, though no further information was provided as to the exact discontinuation date of the initial anticoagulant therapy. Patients lost to follow were classified as having the primary outcome event and negative for reaching the primary safety endpoint. Study endpoints are presented in Table
Outcomes of apixaban in AMPLIFY-EXT [
Outcome | Apixaban 2.5 mg |
Apixaban 5 mg |
Placebo |
---|---|---|---|
Primary efficacy endpoint: composite of symptomatic recurrent vte or death from any cause | 32 (3.8%) | 34 (4.2%) | 96 (11.6%) |
RR versus placebo | RR versus placebo | ||
0.33 (0.22–0.48) | 0.36 (0.25–0.53) | ||
| |||
Secondary efficacy endpoint: symptomatic recurrent VTE or death related to VTE | 14 (1.7%) | 14 (1.7%) | 73 (8.8%) |
RR versus placebo | RR versus placebo | ||
0.19 (0.11–0.33) | 0.20 (0.11–0.34) | ||
| |||
Additional endpoint added after trial initiation: composite of symptomatic recurrent VTE, death related to VTE, myocardial infarction, stroke, or death related to cardiovascular cause | 18 (2.1%) | 19 (2.3%) | 83 (10%) |
RR versus placebo | RR versus placebo | ||
0.21 (0.13–0.35) | 0.23 (0.14–0.38) | ||
| |||
Primary safety endpoint: major bleeding |
2 (0.2%) | 1 (0.1%) | 4 (0.5%) |
RR versus placebo | RR versus placebo | ||
0.49 (0.09–2.64) | 0.25 (0.03–2.24) | ||
| |||
Secondary safety endpoint: major or clinically relevant nonmajor bleeding |
27 (3.2%) | 35 (4.3%) | 22 (2.7%) |
RR versus placebo | RR versus placebo | ||
1.2 (0.69–2.10) | 1.62 (0.96–2.73) | ||
| |||
Recurrent fatal PE or death where PE could not be excluded | 2 (0.2%) | 3 (0.4%) | 7 (0.8%) |
RR versus placebo | RR versus placebo | ||
0.28 (0.06–1.35) | 0.43 (0.11–1.68) | ||
| |||
Recurrent nonFatal PE | 8 (1%) | 4 (0.5%) | 15 (1.8%) |
RR versus placebo | RR versus placebo | ||
0.53 (0.22–1.23) | 0.27 (0.09–0.82) | ||
| |||
Study conclusion | Extension with 12 months of apixaban 2.5 mg or 5 mg twice daily therapy reduced the risk of recurrent VTE and recurrent nonfatal PE, without increasing the risk of major bleeding. |
NR: not reported, LTF: lost to follow (classified as having the primary efficacy endpoint).
Based upon information provided by the authors of the source article, apixaban 5 mg twice daily lowered the rate of recurrent, nonfatal pulmonary embolism [
Dabigatran, an oral direct thrombin inhibitor, has been FDA approved for nonvalvular atrial fibrillation and is also approved in Europe and Canada for VTE prevention. Currently, dabigatran is not approved for the treatment of VTE in Canada, Europe, or the United States of America, and applications to the respective regulatory bodies (EMA, HC, and FDA) have not been submitted. Presently four Phase III clinical trials, RE-COVER, RE-COVER-II, RE-MEDY (NCT00329238), and RE-SONATE (NCT00558259), have been completed to assess the efficacy of dabigatran in the treatment of VTE, with or without PE, though only RE-COVER has published full results [
RE-COVER is a randomized, double-blind, double-dummy, noninferiority trial conducted in 228 centers world-wide, enrolling 2564 patients [
The design of RECOVER was unique in that all patients received a parenteral anticoagulant (intravenous unfractionated heparin or subcutaneous low molecular weight heparin) for the first 6 days, while warfarin dosing was being adjusted to achieve a target INR of 2-3 by sham INR. After that time, patients continued their blinded randomized assignment of either Dabigatran (150 mg twice daily) or warfarin (dose-adjusted to a target INR of 2.0–3.0). Because parenteral anticoagulation was used during the first week, larger initial dabigatran doses, as seen in other NOACs studies, were not utilized as part of this protocol.
RE-COVER II was of similar design, enrolling 2568 patients and was conducted to confirm the previous results and gather data sufficient for subgroup analyses [
Outcomes of dabigatran in VTE studies [
Outcome | RE-COVER |
RE-COVER II |
---|---|---|
Primary efficacy endpoint: recurrent symptomatic VTE or VTE-Related Death | 2.4% versus 2.1%, HR 1.10 (95% CI 0.65–1.84) | 2.4% versus 2.2%, HR 1.08 (95% CI 0.64–1.8) |
Secondary efficacy endpoint: recurrent, nonfatal PE | 1.0% versus 0.6%, HR 1.85 (07.4–4.64) | NR |
Total days overlap between warfarin/dabigatran and parenteral anticoagulant | 5–10 days (mean = 10 days) | 5–11 days (mean NR) |
Major Bleeding |
1.6% versus 1.9%, HR 0.82 (95% CI 0.45–1.48) | 1.1% versus 1.7%, HR 0.69 (95% CI 0.36–1.32) |
Major or clinically relevant non-major bleeding |
5.6% versus 8.8%, HR 0.63 (95% CI 0.47–0.84) | NR |
Any bleeding | 16.1% versus 21.9%, HR 0.71 (95% CI 0.59–0.85) | 15.6% versus 22.1%, HR 0.67 (95% CI 0.56–0.81) |
Death | 1.6% versus 1.7%, HR 0.95 (0.53–1.79) | 1.9% in both groups |
Warfarin TTR | 60% | NR |
Study conclusion | Dabigatran is noninferior to warfarin |
Dabigatran is noninferior to warfarin |
NR: not reported, TTR: time in therapeutic range.
All RE-COVER study patients on dabigatran had a creatinine clearance above 50 mL/min (mean
Subgroup analysis did not indicate differences in the primary endpoint in relation to gender, age, or race or those presenting with initial symptomatic PE or active cancer at baseline. Likewise, the rate of recurrent VTE in patients with previous VTE was not statistically different in the two treatment groups (
Comparison data were not available in RE-COVER II as this was presented in abstract form only. Study authors note that a full publication is pending.
In RE-COVER, dabigatran 150 mg twice daily was noninferior, but not superior to dose-adjusted warfarin managed to a time in therapeutic range (TTR) of 60 percent. The rate of recurrent PE was higher in the dabigatran group (1% versus 0.6%, HR 1.85, 95% CI 0.74–4.64), though not statistically different. The dabigatran 150 mg twice daily dose was also utilized in the RE-LY atrial fibrillation study. Noting that the 18 113 patients were not randomized to correct for VTE risk factors, nor were they reported, the rates of PE development in the dabigatran and warfarin groups were 0.15% per year and 0.09% per year respectively (Relative Risk 1.61, 95% CI 0.76–3.42,
The RE-MEDY (NCT00329238) and RE-SONATE (NCT00558259) trials have both been completed, with data collection terminating in October 2010 and February 2011, respectively. RE-MEDY is a randomized, double-blind, multicenter, parallel-group, active controlled study to compare the efficacy and safety of dabigatran 150 mg twice daily to dose-adjusted warfarin (INR target 2.0–3.0) for the secondary prevention of VTE. A total of 2867 patients in 275 study locations were followed for a total of 36 months to the primary outcome of composite recurrent VTE or VTE death at 18 and 36 months [
RE-SONATE is a randomized, double-blind, multicenter, parallel-group, prevention study to compare dabigatran 150 mg twice daily versus matching placebo in the long-term prevention of recurrent, symptomatic VTE in patients with history of DVT or PE who completed 6–18 months of treatment with VKA therapy. Total study enrollment is 1353 patients in 147 study locations. The primary study endpoint is time to centrally confirmed VTE over the 6-month study period [
Bleeding remains a main safety concern with dabigatran and the NOACs. In the setting of acute VTE, warfarin showed slightly higher rates of nonmajor bleeding, though no difference in major bleeding outcomes was found [
Rivaroxaban, an oral direct Factor Xa inhibitor, has been FDA approved for both nonvalvular atrial fibrillation and VTE prevention and in Canada, Europe, and the United States for VTE treatment. Presently, three Phase III clinical trials, EINSTEIN-DVT, EINSTEIN-PE, and EINSTEIN-Extension, have been completed to assess the efficacy of rivaroxaban treatment in VTE [
EINSTEIN-DVT was an open label, randomized, multicenter, noninferiority trial enrolling 3449 patients with symptomatic DVT without PE [
In EINSTEIN-DVT, patients were either randomized to either rivaroxaban (15 mg orally twice daily for 3 weeks, followed by 20 mg daily for 3 to 12 months) or a parenteral anticoagulant (enoxaparin 1 mg/kg of body weight subcutaneously twice daily) overlapping with a vitamin K antagonist such as warfarin or acenocoumarol (dose-adjusted to a target INR of 2.0–3.0) [
Outcomes of rivaroxaban in VTE Studies [
Outcome | EINSTEIN-DVT |
EINSTEIN-PE |
---|---|---|
Primary efficacy endpoint: noninferiority | ||
Recurrent symptomatic VTE | 2.1% versus 3.0%, | 2.1% versus 1.8%, |
HR = 0.68 (95% CI 0.44–1.04) | HR = 1.12 (95% CI 0.75–1.68) | |
|
|
|
# recurrent events that were |
25 versus 24 | 32 versus 27 |
| ||
Secondary efficacy endpoint: | ||
All-cause mortality | 2.2% versus 2.9% | 2.4% versus 2.1% |
HR = 0.67 (95% CI 0.44–1.02) | HR = 1.13 (95% CI 0.77–1.65) | |
|
|
|
Net clinical benefit | 2.9% versus 4.2% | 3.4% versus 4.0% |
HR = 0.67 (95% CI 0.47–0.95) | HR = 0.85 (95% CI 0.63–1.14) | |
|
|
|
| ||
Primary safety outcome: | ||
Major bleeding or clinically relevant |
8.1% versus 8.1% | 10.3% versus 11.4% |
HR = 0.97 (99% CI 0.76–1.22) | HR = 0.90 (95% CI 0.76–1.07) | |
|
|
|
Major bleeding |
0.8% versus 1.2% | 1.1% versus 2.2% |
HR = 0.65 (95% CI 0.33–1.30) | HR = 0.49 (95% CI 0.31–0.79) | |
|
|
|
Clinically relevant nonmajor |
7.36% versus 7.0% | 9.5% versus 9.8% |
(statistics NR) | (statistics NR) | |
Warfarin TTR | 57.7% | 62.7% |
Study conclusion | Rivaroxaban is noninferior to warfarin. Rivaroxaban is not superior to warfarin. | Rivaroxaban is noninferior to warfarin. Rivaroxaban is not superior to warfarin |
NR: not reported, TTR: time in therapeutic range.
EINSTEIN-PE was an open label, single-blinded (outcomes assessment only), randomized, multicenter, noninferiority trial enrolling 4833 patients with symptomatic PE with or without DVT [
EINSTEIN-Extension study was a placebo-controlled, double-blinded, superiority trial designed to determine if extended therapy would decrease the risk of recurrent VTE in patients with confirmed DVT or PE previously treated for 6–12 months with either VKA or rivaroxaban [
The EINSTEIN studies, although large and well executed, have the limitation of an open label design lending toward potential bias, although the investigators have indicated that this was unlikely to bias in favor of rivaroxaban. Additionally the use of a placebo control for the extension trial provides little assistance in comparing it against existing therapy, warfarin, and its relatively short duration of follow-up (mean of 9 months) does not clearly address safety and efficacy for years of chronic therapy. Although secondary analyses of the EINSTEIN studies did not identify any signals or concern for liver toxicity or evidence of unexpected thrombotic events, longer duration of therapy to address a chronic lifelong condition would have been valuable [
Based upon the data presented, dabigatran and apixaban cannot be recommended until further safety and efficacy data are available. Rivaroxaban however, represents the most promising of the NOACs for treatment of PE and is indicted in Canada and Europe for this indication and recently received FDA approval for this indication [
Based on the strength of available evidence, warfarin proves to be a monitorable, reversible, and effective agent in patients across the spectrum of concurrent disease and should remain the first-line option in conjunction with appropriate parenteral anticoagulation for the treatment of pulmonary embolism, especially in patients who require chronic therapy or therapy beyond 3 months. However, rivaroxaban represents a viable alternative in patients with less than optimal INR control (TTR < 60%) or in whom warfarin monitoring and management is not possible.
The authors report that they have no conflict of interests.
The authors would like to acknowledge Dr. Kimberly Zammit for her editorial assistance.