Author + information
- Received April 8, 2014
- Revision received May 20, 2014
- Accepted June 13, 2014
- Published online December 1, 2014.
- Jerrold H. Levy, MD∗∗ (, )
- Alex C. Spyropoulos, MD†,
- Charles M. Samama, MD, PhD‡ and
- James Douketis, MD§
- ∗Duke University, Durham, North Carolina
- †Hofstra North Shore/LIJ School of Medicine at Lenox Hill Hospital, Manhasset, New York
- ‡Cochin University Hospital, Paris, France
- §McMaster University, Hamilton, Ontario, Canada
- ↵∗Reprint requests and correspondence:
Dr. Jerrold H. Levy, Duke University Medical Center, 2301 Erwin Road, 5691H HAFS, Durham, North Carolina 27710.
Direct oral anticoagulants (DOACs) are approved for multiple thromboembolic disorders and provide advantages over existing agents. As with all anticoagulants, management protocols for the eventuality of bleeding are important. Randomized phase III studies generally show that DOACs have a similar risk of clinically relevant bleeding compared with standard anticoagulants, with reductions in major bleeding in some cases. This may be particularly important in patients with atrial fibrillation, for whom the rate of intracranial hemorrhage was approximately halved with DOACs compared with warfarin. Conversely, the risk of gastrointestinal bleeding may be increased. Specific patient characteristics, such as renal impairment, comedications, and particular aspects of each drug, including the proportion eliminated by the kidneys, must be taken into account when assessing the risk of bleeding. Although routine coagulation monitoring of DOACs is not required, it may be useful under some circumstances. Of the traditional clotting assays, a sensitive and calibrated prothrombin time may be useful for detecting the presence or absence of clinically relevant factor Xa inhibitor concentrations (rivaroxaban or apixaban), but specific anti–factor Xa assays can measure drug levels quantitatively. For dabigatran, the results of an activated partial thromboplastin time test may exclude a clinically relevant pharmacodynamic effect, but a calibrated dilute thrombin time assay can be used for quantification of drug levels. In the event of mild or moderate bleeding, normal hemostatic support measures are recommended. For life-threatening bleeding, use of nonspecific prohemostatic agents may be considered, although clinical evidence is scarce. Specific antidotes are in development.
Several direct oral anticoagulants (DOACs), namely, apixaban (1,2), rivaroxaban (3,4), and dabigatran etexilate (5,6), are currently licensed in Europe and the United States for various thromboembolic indications. A fourth DOAC, edoxaban, has also demonstrated efficacy and safety in venous thromboembolism (VTE) treatment and stroke prevention in patients with atrial fibrillation (AF) (7,8), but is not licensed in Europe or the United States.
The DOACs have a rapid onset (∼2 to 4 h) and offset (∼24 h) of action with normal renal function (9,10). They provide alternatives to low molecular weight heparin (LMWH) in a peri-operative setting for VTE prophylaxis and therapy and to vitamin K antagonists (VKAs) for longer term therapy. Because the DOACs have predictable pharmacokinetic/pharmacodynamic effects, routine coagulation monitoring for titration and maintenance is not required (11). However, if patients experience bleeding or need procedural interventions, laboratory monitoring can be performed. Although there are currently no specific reversal agents available to manage life-threatening bleeding with DOACs, most anticoagulants are not acutely reversible, except for unfractionated heparin with protamine (12). VKAs are acutely reversible with 4-component prothrombin complex concentrates (PCCs), including 1 recently approved in the United States (prothrombin complex concentrate), but there is no specific reversal agent for LMWHs, which may accumulate in patients with renal dysfunction (13). For all anticoagulants, management protocols for potential bleeding should be established. Clinical studies with the DOACs for current indications have provided extensive safety data. This review summarizes current and evolving data for the DOACs and management strategies for bleeding, when it occurs.
Therapeutic and Bleeding Profiles of DOACs in Clinical Studies
Apixaban, a direct factor Xa inhibitor, is widely approved for thromboprophylaxis in elective hip or knee replacement surgery (1,14) and for stroke prevention in patients with nonvalvular AF (Table 1) (1,2).
For VTE prophylaxis, apixaban, initiated 12 to 24 h postoperatively, was compared with enoxaparin for preventing VTE in elective hip/knee replacement surgery (ADVANCE studies) (15–17). In ADVANCE-1 (Apixaban Dosed Orally Versus Anti-coagulation with Injectable Enoxaparin to Prevent Venous Thromboembolism), apixaban did not demonstrate noninferiority for efficacy compared with enoxaparin 30 mg twice daily (bid) when given after knee replacement surgery (15). However, apixaban was superior to enoxaparin 40 mg once daily (qd) in ADVANCE-2 when given after knee replacement surgery (16), and in ADVANCE-3 (Apixaban Dosed Orally Versus Anticoagulation with Injectable Enoxaparin to Prevent Venous Thromboembolism-3) after hip replacement surgery (17). Major bleeding and clinically relevant bleeding occurred at a similar rate between the treatment groups in these studies (Table 2) (15–17). Apixaban has been compared with LMWH plus warfarin in a randomized phase III study for treating acute VTE (AMPLIFY [Apixaban for the Initial Management of Pulmonary Embolism and Deep-Vein Thrombosis as First-Line Therapy]) (Table 3) (18). Overall, there was a significantly lower incidence of major and nonmajor clinically relevant bleeding with apixaban (4.3% vs. 9.7%; p < 0.001) (18). A 12-month extension study (AMPLIFY-EXT [Apixaban after the Initial Management of Pulmonary Embolism and Deep Vein Thrombosis with First-Line Therapy–Extended Treatment]) compared apixaban 2.5 mg or 5 mg bid with placebo for the secondary prevention of recurrent VTE in patients who had already received 6 to 12 months of anticoagulation treatment (19). There was a similar incidence of major bleeding with both doses (Table 4). Rates of clinically relevant bleeding were numerically higher with active treatment (3.2% and 4.3% vs. 2.7%, respectively; p values nonsignificant for all comparisons), but rates of all-cause mortality were lower (0.8% and 0.5% vs. 1.7%, respectively) (19).
Long-term apixaban 5 mg bid was compared with acetylsalicylic acid (ASA) and warfarin in 2 separate trials for the prevention of stroke and systemic embolism in patients with nonvalvular AF (AVERROES [Apixaban Versus Acetylsalicylic Acid to Prevent Stroke in Atrial Fibrillation Patients Who Have Failed or Are Unsuitable for Vitamin K Antagonist Treatment] and ARISTOTLE [Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation], respectively) (20,21). Apixaban was superior to warfarin in terms of major and nonmajor clinically relevant bleeding (p < 0.001) and all-cause mortality (p = 0.047) (Table 5). Rates of major gastrointestinal bleeding were similar with apixaban and both comparators, and there was significantly less intracranial bleeding with apixaban compared with warfarin (0.3%/year vs. 0.8%/year; p < 0.001) (20,21).
Apixaban in combination with standard antiplatelet therapy was compared with antiplatelet therapy alone in patients with recent acute coronary syndrome (ACS) (APPRAISE-2 [Apixaban for Prevention of Acute Ischemic Events-2]) (22). However, the risk of bleeding outweighed the clinical benefit of anticoagulation in these patients, and the trial was stopped early.
Rivaroxaban is a direct factor Xa inhibitor and is licensed in the European Union and North America for: 1) the treatment of deep venous thrombosis (DVT) and pulmonary embolism (PE); 2) the prevention of recurrent DVT and PE in adults; 3) thromboprophylaxis in adults undergoing elective hip or knee replacement surgery; and 4) the prevention of stroke and systemic embolism in adults with nonvalvular AF (3,4). In the European Union, rivaroxaban has been approved (at a dose of 2.5 mg bid), in combination with ASA alone or ASA plus clopidogrel or ticlopidine to prevent atherothrombotic events in patients with ACS and elevated cardiac biomarkers (Table 1) (3).
The phase III RECORD (REgulation of Coagulation in ORthopaedic surgery to prevent Deep vein thrombosis and pulmonary embolism) program evaluated the use of rivaroxaban for VTE prophylaxis in patients undergoing elective total hip or knee replacement surgery and consisted of 4 trials of rivaroxaban 10 mg qd (started 6 to 8 h after surgery) compared with 2 standard subcutaneous enoxaparin regimens (30 mg bid initiated after surgery and 40 mg qd initiated before surgery) (23–26). Rivaroxaban was superior to enoxaparin 30 mg bid and 40 mg qd for VTE prevention, with a similar incidence of major bleeding (Table 2) (23–28). However, bleeding at the surgical site was not classified as major bleeding but was included as part of a composite of major and nonmajor clinically relevant bleeding. In a pooled analysis of the 4 trials, major and nonmajor clinically relevant bleeding occurred more frequently with rivaroxaban than with enoxaparin over the total treatment duration (3.2% vs. 2.6%; p = 0.04) but not during the 12 ± 2 days of active treatment (2.9% vs. 2.5%; p = 0.19) (29).
Three phase III randomized studies of rivaroxaban in the VTE treatment setting were conducted (30,31). In the EINSTEIN DVT and EINSTEIN PE trials, rivaroxaban was noninferior to standard enoxaparin/VKA treatment in patients who had acute DVT (without PE) (30) and PE (with or without DVT) (31), respectively. In the EINSTEIN EXT, extended rivaroxaban treatment was superior to placebo for the prevention of recurrent VTE in patients already successfully treated for an initial VTE and for whom the benefit-risk balance of continuing or stopping treatment was unclear (30). In the EINSTEIN DVT and EINSTEIN EXT, there was no significant difference in major bleeding between rivaroxaban and the comparator regimen (Tables 3 and 4); however, in the EINSTEIN PE, rivaroxaban treatment led to a significant 51% relative risk reduction in major bleeding compared with enoxaparin/VKA (Table 3) (31). In both acute treatment studies, major bleeding in a critical site, associated with a decrease in hemoglobin of ≥2 g/dl and/or transfusion of ≥2 units of blood, or leading to death, occurred with an incidence of <1% in the rivaroxaban arms (30,31). In the EINSTEIN PE, there were fewer cases of major bleeding at a critical site, especially intracranial and retroperitoneal bleeding, with rivaroxaban than with enoxaparin/VKA (31).
Further data on the long-term use of rivaroxaban 20 mg qd were provided by the ROCKET AF (Rivaroxaban Once daily, Oral, Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation) study, in which rivaroxaban was noninferior to warfarin for the prevention of stroke or systemic embolism in patients with nonvalvular AF, and rivaroxaban did not increase the rate of clinically relevant bleeding (14.9%/year vs. 14.5%/year; p = 0.44) (Table 5) (32). Rivaroxaban was associated with significant reductions in the annual rates of intracranial hemorrhage (ICH) (0.5% vs. 0.7%; p = 0.02), critical site bleeding (0.8% vs. 1.2%; p = 0.007), and fatal bleeding (0.2% vs. 0.5%; p = 0.003) compared with warfarin, set against an increase in gastrointestinal bleeding (3.2% vs. 2.2%; p < 0.001), major bleeding associated with a ≥2 g/dl decrease in hemoglobin (2.8% vs. 2.3%; p = 0.02), and major bleeding requiring blood transfusion (1.6% vs. 1.3%; p = 0.04) (Table 5) (32).
In the ATLAS ACS-2 TIMI 51 (Anti-Xa Therapy to Lower cardiovascular events in Addition to aspirin with/without thienopyridine therapy in Subjects with Acute Coronary Syndrome 2-Thrombolysis In Myocardial Infarction 51) study, rivaroxaban 2.5 mg or 5 mg bid in combination with standard antiplatelet therapy (ASA with or without a thienopyridine) was compared with antiplatelet therapy alone in patients with recent ACS (33). Rivaroxaban significantly reduced the incidence of death of cardiovascular causes, myocardial infarction, or stroke (p = 0.008 across both doses compared with antiplatelet therapy alone), but also led to a significant increase in major bleeding not related to coronary artery bypass grafting (2.1% vs. 0.6%, respectively; p < 0.001) and in ICH (0.6% vs. 0.2%, respectively; p = 0.009). However, fatal bleeding was not significantly increased (0.3% vs. 0.2%, respectively; p = 0.66). Overall, the 2.5-mg bid rivaroxaban dose was associated with a lower risk of bleeding compared with the higher (5 mg bid) dose (0.1% vs. 0.4%; p = 0.04). The U.S. Food and Drug Administration has not approved rivaroxaban use in patients with ACS.
Edoxaban is a direct factor Xa inhibitor that is not currently licensed in Europe or the United States (Table 1). Edoxaban was compared with LMWH/warfarin for the treatment of VTE in the randomized phase III Hokusai-VTE study (8). Patients in both treatment arms received heparin induction at the start of treatment. Edoxaban was noninferior to warfarin for the prevention of recurrent symptomatic VTE and led to a significantly lower incidence of major plus nonmajor clinically relevant bleeding (p = 0.004) (Table 3) (8). There was a similar incidence of major bleeding in both treatment arms (1.4% vs. 1.6%; p = 0.35), and fatal bleeding occurred in 2 patients in the edoxaban arm compared with 10 in the warfarin arm. There were no fatal intracranial or retroperitoneal bleeding events with edoxaban, and fewer nonfatal bleeding episodes in a critical site compared with warfarin (0.3% vs. 0.6%, including 5 vs. 12 nonfatal ICHs) (8).
The efficacy and safety of edoxaban for the prevention of stroke in patients with nonvalvular AF was evaluated in the Engage AF-TIMI 48 (Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis In Myocardial Infarction 48) study (Table 5) (7). Edoxaban was noninferior to warfarin for the incidence of stroke and systemic embolism. Major bleeding occurred with a significantly lower incidence with both edoxaban doses compared with warfarin (1.6% and 2.8%/year, respectively, vs. 3.4%/year; p < 0.001 for both doses) (Table 5) (7). The endpoint of death or ICH also occurred in significantly fewer patients receiving edoxaban than warfarin (4.0% and 4.3%/year, respectively, vs. 4.9%/year; p < 0.001 and p = 0.004, respectively). Of note, fatal bleeding (0.1% and 0.2%/year vs. 0.4%/year; p < 0.001 and p = 0.006, respectively) and life-threatening bleeding (0.3% and 0.4%/year vs. 0.8%/year; p < 0.001 for both doses) were significantly less frequent with edoxaban, as was gastrointestinal bleeding with the lower dose (0.8% vs. 1.2%/year; p < 0.001). In contrast, the higher edoxaban dose led to more gastrointestinal bleeding than warfarin (1.5% vs. 1.2%/year; p = 0.03) (7).
Dabigatran is a direct factor IIa (thrombin) inhibitor and is approved in Europe for thromboprophylaxis in patients undergoing total hip and knee replacement, in the United States for VTE treatment, and in Europe and North America for the prevention of stroke and systemic embolism in patients with nonvalvular AF (Table 1) (5,6).
RE-NOVATE (Dabigatran Etexilate in Extended Venous Thromboembolism [VTE] Prevention After Hip Replacement Surgery) and RE-NOVATE II (Dabigatran Etexilate Compared With Enoxaparin in Prevention of Venous Thromboembolism [VTE] Following Total Hip Arthroplasty) were noninferiority studies comparing dabigatran 150 mg or 220 mg qd (starting with a half-dose 1 to 4 h after surgery) with enoxaparin 40 mg qd (initiated before surgery) for VTE prophylaxis in patients undergoing total hip replacement (34,35). The same doses were also studied after knee replacement surgery in the RE-MODEL (Dabigatran Etexilate 150 mg or 220 mg Once Daily (o.d.) Versus (v.s.) Enoxaparin 40 mg o.d. for Prevention of Thrombosis After Knee Surgery) (vs. enoxaparin 40 mg qd) and the RE-MOBILIZE (vs. enoxaparin 30 mg bid) (36,37). In these studies, the rates of major bleeding were similar (Table 2) (34–37).
Dabigatran was studied for acute treatment of VTE in the RE-COVER (Efficacy and Safety of Dabigatran Compared to Warfarin for 6 Month Treatment of Acute Symptomatic Venous Thromboembolism) and the RE-COVER II (Phase III Study Testing Efficacy & Safety of Oral Dabigatran Etexilate vs Warfarin for 6 m Treatment for Acute Symp Venous Thromboembolism [VTE]) (38,39). All patients received initial parenteral anticoagulation. In both trials, dabigatran was noninferior to standard care, and there was no significant difference in the incidence of major bleeding (Table 3) (38,39). In the RE-COVER, there were no cases of ICH with dabigatran, but approximately one-fourth of all bleeding events with dabigatran were gastrointestinal. Two further studies considered the potential role of dabigatran as a long-term therapy for the prevention of recurrent VTE after patients had received initial treatment for a primary event. Dabigatran was noninferior to warfarin (RE-MEDY [Secondary Prevention of Venous Thrombo Embolism (VTE)]) and superior to placebo (RE-SONATE [Twice-daily Oral Direct Thrombin Inhibitor Dabigatran Etexilate in the Long Term Prevention of Recurrent Symptomatic VTE]) for the prevention of recurrent VTE (40). Only 2 major bleeding events occurred with dabigatran in the RE-SONATE (Table 4), and there were numerically fewer incidences of major bleeding with dabigatran than with warfarin in the RE-MEDY, including major bleeding in a critical organ, causing a decrease in hemoglobin, or requiring a blood transfusion. However, there was a greater incidence of ACS in patients taking dabigatran than in those receiving warfarin (0.9% vs. 0.2%; p = 0.02).
The profile of long-term dabigatran therapy has been further defined by the RE-LY (Randomized Evaluation of Long term anticoagulant therapY) study (Table 5), in which 110-mg and 150-mg bid doses were compared with standard warfarin therapy for the prevention of stroke and systemic embolism in patients with nonvalvular AF (41). The lower dabigatran dose was noninferior for efficacy to warfarin in this trial, and the higher dose was superior. The 110-mg dose of dabigatran conferred a significantly lower rate of major bleeding, and the 150-mg dose had a similar rate of major bleeding compared with warfarin (2.7% vs. 3.1% vs. 3.4%/year; p = 0.003 and p = 0.31, respectively). Both doses significantly reduced intracranial and life-threatening bleeding, but the higher dabigatran dose was associated with a higher rate of gastrointestinal bleeding (Table 5) and a slight increase in the rate of myocardial infarction compared with warfarin (41). In the RELY-ABLE (Long Term Multi-Center Extension of Dabigatran Treatment in Patients with Atrial Fibrillation) extension study, patients randomized to dabigatran in the RE-LY who had not permanently discontinued treatment continued to receive dabigatran. Rates of major bleeding remained similar to those in the RE-LY, with the lower dose associated with a significantly lower risk than the higher dose (3.7% vs. 3.0%/year, respectively; hazard ratio: 1.26; 95% confidence interval: 1.04 to 1.53). There was no significant difference between the doses in the risk of life-threatening, fatal, gastrointestinal, or intracranial bleeding (Table 5) (42).
Bleeding Risk in Patients Treated with DOACs
Based on data from phase III studies, DOACs can be expected to have a risk of clinically relevant bleeding similar to that with standard anticoagulants. The rate of major bleeding is also generally similar; however, in clinical trials of apixaban for VTE treatment and rivaroxaban for PE treatment, significant (69% and 51%) relative risk reductions in major bleeding compared with standard therapy have been demonstrated (18,31). When used for extended periods for the prevention of stroke, the DOACs were also associated with clinically important reductions in major bleeding compared with warfarin, including life-threatening bleeding types (7,21,32,41). An ∼50% reduction in ICH, a major complication associated with long-term warfarin use (43), is notable. This may be related to lower suppression of thrombin generation with DOACs compared with warfarin (44) and possibly tissue factor–dependent mechanisms. However, there may also be an increase in other types of bleeding compared with warfarin, such as gastrointestinal hemorrhage (7,32,38,41).
Certain patient groups are at increased risk of bleeding and therefore require careful assessment of the benefit-risk balance of anticoagulant treatment, particularly when continued long term. When bleeding occurs in patients treated with a DOAC, knowledge of the pharmacokinetic and pharmacodynamic characteristics of the agent concerned is important to inform clinical management. Apixaban, rivaroxaban, edoxaban, and dabigatran all reach maximal concentrations between 1 and 4 h after intake and have relatively short half-lives, ranging from 5 to 17 h in healthy subjects (1–6,45) (Table 1), which contrasts with the long half-life of warfarin (∼40 h) (46). However, drug elimination may be prolonged owing to specific factors, the most important of which are the renal clearance profiles of the patient and the drug. Dabigatran is mostly removed by the kidneys (∼80% of a dose is recoverable as unchanged drug in the urine) (47) and may therefore accumulate in patients with renal insufficiency, whereas rivaroxaban (48,49) and apixaban (50) are less affected to a clinically relevant degree by moderate renal impairment (creatinine clearance [CrCl] 30 to 49 ml/min): ∼33% of rivaroxaban is cleared as active drug by renal mechanisms (3,4); 25% to 28% of apixaban is cleared by renal elimination (Table 1) (1,2). Severe renal impairment (CrCl, 15 to 29 ml/min) leads to a doubling of the half-life of dabigatran (51). Edoxaban has an intermediate profile, with 50% of the dose undergoing renal elimination (52).
Patients with moderate renal impairment (CrCl, 30 to 49 ml/min) who are receiving rivaroxaban for VTE treatment do not require dose adjustment, although in Europe, a 15-mg qd dose after the initial 3 weeks of 15-mg bid dosing may be considered based on clinical evaluation of the risk of thrombosis and bleeding (3). In contrast, patients with AF and moderate renal impairment who receive rivaroxaban for stroke prevention should always receive a 15-mg qd dose (Table 1). In Europe, caution is recommended in all patients who have severe renal insufficiency (CrCl, 15 to 29 ml/min); in the United States, rivaroxaban is not recommended in these patients (3,4). Apixaban is given at a reduced dose for the prevention of stroke in some patients with AF (Table 1) (1,2). Dose reduction with dabigatran for patients with AF should also be considered in patients with renal impairment and those receiving comedications with interaction potential (Table 1). Dabigatran is contraindicated in patients with CrCl 15 to 29 ml/min in Europe but may be used with caution in these patients in the United States at a reduced dose (5,6). No DOAC should be used in patients with CrCl <15 ml/min. Recommendations for edoxaban, if and when approved in North America or Europe, remain to be determined, but a dose reduction was mandated in the Hokusai-VTE and Engage AF-TIMI 48 studies for certain patients (Table 1) (7,8).
Hepatic impairment also increases the risk of bleeding. Moderate hepatic impairment (Child-Pugh B) affects the pharmacokinetics of rivaroxaban and apixaban (but not dabigatran) to a clinically relevant degree (1,2,53,54), and severe hepatic impairment would be expected to lead to a substantial increase in bleeding risk with any anticoagulant. Rivaroxaban is contraindicated in patients with hepatic disease associated with coagulopathy and clinically relevant bleeding risk, including cirrhotic patients with Child-Pugh B or C (3,4). Apixaban can be used with caution in patients with Child-Pugh B (1,2), whereas any liver impairment expected to affect survival is a contraindication to dabigatran (5,6). In Japan, caution is advised when using edoxaban in patients with severe hepatic impairment (55).
Interactions with concomitant drugs that share the elimination pathways of an anticoagulant may also serve to increase exposure and thus trigger a bleeding episode. The DOACs have a considerably lower potential for drug-drug interactions than VKAs (9), but there are relevant interactions (Figure 1). Apixaban and rivaroxaban are metabolized mainly via cytochrome P450 (CYP) 3A4-dependent and P-glycoprotein (P-gp)–dependent pathways (1,2,49), and bleeding may be caused by the use of comedications that interact strongly with both these pathways. This is of greatest clinical relevance with strong inhibitors of both CYP3A4 and P-gp, such as azole-antimycotics (e.g., ketoconazole) and human immunodeficiency virus protease inhibitors (e.g., ritonavir) (49), and apixaban (1,2) and rivaroxaban (3,4) should not be coadministered with these drugs (Table 1). Strong inhibitors of 1 pathway or moderate inhibitors of both had a lesser effect that was not considered clinically relevant (1,2,49), but concomitant use in patients with renal impairment could still lead to relevant pharmacodynamic effects. Strong CYP3A4 inducers should also be used with caution or avoided with rivaroxaban and apixaban. In contrast, neither dabigatran nor its prodrug, dabigatran etexilate, is metabolized by CYP-dependent mechanisms (5,6). However, both are P-gp substrates (5,6), and the effect of strong P-gp inhibitors on the bioavailability of dabigatran could be greater than with rivaroxaban and apixaban. Less than 4% of an edoxaban dose is subject to CYP3A4-dependent clearance, which may allow use in patients taking concomitant medications that would preclude use of rivaroxaban or apixaban (52). Unlike with VKAs, food interactions with DOACs are minimal and not likely to cause overexposure. Rivaroxaban doses of 15 mg and 20 mg should be taken with food (Table 1) (56,57). There was a modest effect on the pharmacokinetic parameters of edoxaban when taken with food, but this is not expected to be of clinical relevance (58).
In patients with AF who are receiving long-term anticoagulation therapy for stroke prevention, ACS or VTE may develop, the latter perhaps owing to poor warfarin control. For the former, unless the event is immediately life-threatening (e.g., massive PE requiring thrombolysis or embolectomy), such patients can be transitioned to rivaroxaban (as the only DOAC approved for VTE treatment in the European Union and the United States) at the initial 15-mg bid dose (3,4). During the initial 3-week 15-mg bid dosing period with rivaroxaban, patients should be monitored closely for signs of bleeding, although in the EINSTEIN DVT and EINSTEIN PE trials, there was no increase in major bleeding compared with enoxaparin/VKA during this phase (30,31). If rivaroxaban is used in patients taking antiplatelet agents for AF, an increase in bleeding risk is likely; this may be particularly important because patients with AF are generally elderly and may have renal impairment or other comorbidities and/or be taking medications that interact with rivaroxaban to increase exposure. If the benefit-risk profile is favorable, rivaroxaban may be combined with doses of ASA not exceeding 100 mg/day (59), but dual-antiplatelet therapy should not be combined with rivaroxaban in patients with AF. In contrast, a low dose of rivaroxaban may now be combined with single- or dual-antiplatelet therapy in Europe for patients with recent ACS, if they have elevated cardiac biomarkers indicating a likely secondary event (3). The approved dose of rivaroxaban (2.5 mg bid) in ACS is much lower than that in other indications. In patients without elevated biomarkers, the addition of anticoagulation to antiplatelet treatment cannot be justified because of the significant increase in risk of major bleeding, as observed in the APPRAISE-2 (22) and ATLAS ACS 2 TIMI 51 (33) trials. Rivaroxaban is not approved for patients with both AF and ACS.
Monitoring Anticoagulation with the DOACs
Routine coagulation monitoring is not required with DOACs but is recommended in patients with renal impairment, acute bleeding, overdoses, or emergency surgery (10). The interval between the last dose and sampling must be considered when interpreting the test results. Rivaroxaban prolongs the prothrombin time (PT), with substantial interassay variability (60). The PT provides a qualitative indication of the anticoagulant effect but does not measure drug levels. The international normalized ratio (INR) should not be used for rivaroxaban (60) or for other direct factor Xa inhibitors (61). Specific anti–factor Xa assays, distinct from LMWH testing, are recommended for quantitative measurements of rivaroxaban, apixaban, and likely for edoxaban (Table 6) (60,61).
Dabigatran prolongs most coagulation assays except PT (62). A normal thrombin time assay can be used to exclude a clinically relevant dabigatran effect and is better for this purpose than the activated partial thromboplastin time, although the dilute thrombin time assay HEMOCLOT (Aniara, West Chester, Ohio) better correlates with plasma concentrations and is more sensitive for dabigatran (Table 6) (63). The ecarin clotting time test provides a dose-dependent correlation with dabigatran (64) but is not widely available.
DOACs have a faster onset/offset of action than VKAs and can theoretically be stopped closer to the time of surgery (10). Rivaroxaban can be stopped up to 24 h before based on European and U.S. prescribing recommendations (3,4). A general principle is that pre-procedural DOAC discontinuation should be based on the specific pharmacokinetics, renal function, and procedural bleeding risk; post-procedural DOAC resumption should be based on bleeding risk and whether adequate hemostasis has been achieved (65).
Recommendations suggest stopping DOACs ∼24 h (2 to 3 half-lives) before a procedure that carries a low bleeding risk, but 5 days before with a medium or high bleeding risk, dependent on the DOAC and the patient’s renal function (66,67). The European Heart Rhythm Association suggests stopping DOACs ≥24 h before surgery for low-risk procedures and ≥48 h before high-risk surgery, but longer for patients with CrCl <80 ml/min for dabigatran and CrCl 15 to 30 ml/min for rivaroxaban or apixaban (68). Other expert consensus documents recommend 24- to 48-h discontinuation windows (65). Using such a scheme in the RE-LY trial yielded similar rates of peri-operative bleeding/thromboembolism in warfarin- and dabigatran-treated patients (69). Additional studies are ongoing (65).
If the patient’s risk of thrombosis warrants resumption of anticoagulation after periprocedural cessation, DOAC administration can be resumed 12 to 24 h after procedures associated with rapid and complete restoration of hemostasis. In general, DOACs may be resumed within 24 h for a procedure with a low risk of bleeding, and within 48 to 72 h for a procedure with a high risk of bleeding (65). For procedures associated with an inability to take oral medications (e.g., post-operative intestinal ileus), bridging with either unfractionated heparin or reduced-dose LMWH may be considered before transitioning to a DOAC 48 to 72 h post-surgery (68). Bridging therapy with a DOAC should otherwise be avoided, except in patients with very high thrombotic risk (65).
Interruption of DOACs and Switching Between Anticoagulants
In the ROCKET AF study, thromboembolic events increased when patients discontinued rivaroxaban; however, temporary interruption led to low rates of stroke and major bleeding similar to those with warfarin (32,70). Prolonged inadequate anticoagulation should be avoided if a DOAC is discontinued for reasons other than bleeding, and transitioning to another anticoagulant should be considered. If switching to warfarin/VKA, advice differs between Europe and the United States and between the factor Xa inhibitors and dabigatran. For apixaban and rivaroxaban in Europe, concurrent administration of the DOAC and VKA is recommended for at least 2 days and thereafter until the INR is ≥2.0 (tested at the trough DOAC concentration to minimize interference), after which the DOAC can be discontinued (1,3). For dabigatran, a similar approach is recommended, but with at least 2 days of concurrent DOAC and VKA administration for patients with CrCl of 30 to 49 ml/min, and at least 3 days for those with CrCl ≥50 ml/min (to account for the dependence of dabigatran on renal clearance) (5). The U.S. prescribing information suggests a different approach for apixaban and rivaroxaban of discontinuing the DOAC and starting the VKA plus a parenteral anticoagulant as bridging therapy until the INR reaches the therapeutic range (2,4). For dabigatran, the U.S. advice is similar to that given in Europe, but with 3 days of concurrent administration of DOAC and VKA suggested for patients with CrCl ≥50 ml/min, 2 days for those with CrCl of 30 to 50 ml/min, and 1 day in the case of CrCl of 15 to 30 ml/min (6). For transition to a parenteral anticoagulant (e.g., LMWH in the case of a patient with cancer), the advice is simpler and more uniform: start the parenteral agent and discontinue the DOAC when the next dose of DOAC is scheduled (1–6). However, for dabigatran, it may be necessary to wait 24 h before initiating the new anticoagulant in patients with CrCl <30 ml/min (5,6).
Recommended Management Strategies for Bleeding Associated with DOACs
For moderate or severe bleeding, standard hemodynamic support measures, such as fluid replacement and blood transfusion, can be applied to patients receiving DOACs as with other anticoagulants (Figure 2). These include mechanical compression (e.g., severe epistaxis), surgical hemostasis with bleeding control procedures, fluid replacement and hemodynamic support, use of blood products (packed red cells, fresh frozen plasma, or platelets), and, depending on laboratory testing and other factors, cryoprecipitate or fibrinogen concentrates (1–6,52). Rivaroxaban, apixaban, and, it is anticipated, edoxaban, have high protein binding; therefore, they are not dialyzable (1,2,52,71), whereas dabigatran can be partially removed by dialysis (51,72). The use of activated charcoal can be considered in the event of an overdose, provided this is within ∼6 h of ingestion. If bleeding occurs and cannot be controlled with these measures, interventions may be required. DOACs should be discontinued before a planned intervention, as discussed (3,4), although renal function is important (10), especially for patients at risk of bleeding (66,67). In emergencies, immediate surgery may be required, and clinical judgment must be exercised. Rivaroxaban, although approved for PE therapy, should not be given to patients with hemodynamically unstable PE (3,4).
Management of Life-Threatening Bleeding
If bleeding is life-threatening, the off-label therapeutic use of PCC or activated PCC may be considered to attempt to reverse the anticoagulant effect of the DOACs (1–6). However, experience with these therapeutic approaches is limited to preclinical studies, which have shown variable results (73–78), and reversal of anticoagulation in healthy volunteers (79–83), as well as some case reports in patients. One recent study in healthy volunteers found that 3-factor PCC reversed rivaroxaban-induced changes in thrombin generation more than 4-factor PCC (84). With ICH or serious bleeding, recommendations suggest PCC administration at 50 U/kg or activated PCC (anti-inhibitor coagulant complex) at 30 to 50 U/kg (85), and readministered once if required (85). Hemodialysis guided by measured drug concentrations should be considered for dabigatran.
A specific reversal agent for factor Xa inhibitors (andexanet alfa) is in development. The molecule is a recombinant protein analog of factor Xa that binds to direct factor Xa inhibitors and antithrombin but does not itself have any catalytic activity. In ex vivo and animal studies, andexanet alfa was able to dose-dependently reverse factor Xa inhibition (86). In addition, a potent monoclonal antibody directed against dabigatran, idarucizumab, is now under clinical investigation; the RE-VERSE AD (RE-VERSal Effects of idarucizumab on Active Dabigatran) study (NCT02104947) will evaluates idarucizumab for dabigatran reversal in patients with uncontrolled bleeding or who require emergency surgery (87).
Ongoing Clinical Studies
Ongoing trials to investigate the use of DOACs in patients with various cardiovascular conditions and undergoing cardiac interventions are summarized in Table 7. Additionally, clinical registries, such as GARFIELD (http://www.tri-london.ac.uk/garfield), will provide further information on the real-world use of DOACs.
DOACs provide important advantages in the short-term prophylaxis of VTE in patients undergoing hip or knee replacement surgery and in the longer term treatment of VTE and prevention of stroke in patients with AF compared with traditional agents, including reductions in dangerous bleeding types. However, they also have different bleeding profiles that require individualized management approaches. Further study and increasing use of apixaban, rivaroxaban, dabigatran, and edoxaban in real-world practice will help to familiarize physicians with best practice in this area. Development of specific measurement techniques and reversal agents will also provide further tools for the management of bleeding.
The authors acknowledge Stephen Purver, who provided editorial assistance with funding from Bayer HealthCare Pharmaceuticals and Janssen Scientific Affairs, LLC.
Funding for additional editorial support was provided by Bayer HealthCare Pharmaceuticals and Janssen Scientific Affairs, LLC. Dr. Levy serves on research steering committees for CSL Behring, Boehringer-Ingelheim, Grifols, and Jansen; and is a consultant to Medco, Portola, and Roche. Dr. Spyropoulos serves on the steering committee of Bayer; and has served as a consultant for Johnson & Johnson, Bristol-Myers Squibb, Astellas, Boehringer Ingelheim, Jansen, Daiichi Sankyo, and Sanofi-Aventis. Dr. Samama has received grants from NovoNordisk, CSL Behring, and Laboratoire français du fractionnement et des biotechnologies; has received speaker honoraria from Abbott, Bayer, Bristol-Myers Squibb, Boehringer Ingelheim, CSL Behring, Daiichi Sankyo, GlaxoSmithKline, Laboratoire français du fractionnement et des biotechnologies, Octapharma, Pfizer, Rovi, and Sanofi-Aventis; has served on the advisory boards of Bayer, Bristol-Myers Squibb, Boehringer Ingelheim, Daiichi Sankyo, GlaxoSmithKline, Pfizer, Roche, and Sanofi-Aventis; is a primary investigator for Bayer, Bristol-Myers Squibb, Boehringer Ingelheim, Laboratoire français du fractionnement et des biotechnologies, GlaxoSmithKline, and Sanofi-Aventis; and serves on the steering committees of Boehringer Ingelheim, Bayer, Bristol-Myers Squibb, Pfizer, Daiichi Sankyo, CSL Behring, and Laboratoire français du fractionnement et des biotechnologies. Dr. Douketis has reported that he has no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute coronary syndrome
- atrial fibrillation
- acetylsalicylic acid
- twice daily
- creatinine clearance
- cytochrome P450
- direct oral anticoagulant
- deep venous thrombosis
- intracranial hemorrhage
- international normalized ratio
- low molecular weight heparin
- prothrombin complex concentrate
- pulmonary embolism
- prothrombin time
- once daily
- vitamin K antagonist
- venous thromboembolism
- Received April 8, 2014.
- Revision received May 20, 2014.
- Accepted June 13, 2014.
- American College of Cardiology Foundation
- ↵Bristol-Myers Squibb, Pfizer EEIG. Eliquis (apixaban) Summary of Product Characteristics. 2014. Available at: http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/002148/WC500107728.pdf. Accessed November 17, 2014.
- ↵Bristol-Myers Squibb Company, Pfizer Inc. Eliquis (apixaban) Prescribing Information. 2014. Available at: http://packageinserts.bms.com/pi/pi_eliquis.pdf. Accessed November 17, 2014.
- ↵Bayer Pharma AG. Xarelto (rivaroxaban) Summary of Product Characteristics. 2014. Available at: http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000944/WC500057108.pdf. Accessed November 17, 2014.
- Janssen Pharmaceuticals Inc. Xarelto (rivaroxaban) Prescribing Information. 2014. Available at: http://www.xareltohcp.com/sites/default/files/pdf/xarelto_0.pdf. Accessed November 17, 2014.
- ↵Boehringer Ingelheim International GmbH. Pradaxa (dabigatran etexilate) Summary of Product Characteristics. 2014. Available at: http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000829/WC500041059.pdf. Accessed November 17, 2014.
- ↵Boehringer Ingelheim Pharmaceuticals Inc. Pradaxa (dabigatran etexilate) Prescribing Information. 2014. Available at: http://bidocs.boehringer-ingelheim.com/BIWebAccess/ViewServlet.ser?docBase=renetnt&folderPath=/Prescribing Information/PIs/Pradaxa/Pradaxa.pdf. Accessed November 17, 2014.
- Eikelboom J.W.,
- Weitz J.I.
- Bristol-Myers Squibb Canada Inc., Pfizer Canada Inc. Eliquis (apixaban) Product Monograph. 2014. Available at: http://www.bmscanada.ca/static/products/en/pm_pdf/ELIQUIS_EN_PM.pdf. Accessed November 17, 2014.
- Lassen M.R.,
- Gent M.,
- Kakkar A.K.,
- et al.
- Schulman S.,
- Kakkar A.K.,
- Goldhaber S.Z.,
- et al.
- Connolly S.J.,
- Wallentin L.,
- Ezekowitz M.D.,
- et al.
- Schulman S.,
- Beyth R.J.,
- Kearon C.,
- Levine M.N.,
- American College of Chest Physicians
- ↵Bristol-Myers Squibb. Coumadin (warfarin sodium) Prescribing Information. 2011. Available at: http://packageinserts.bms.com/pi/pi_coumadin.pdf. Accessed May 15, 2014.
- Troconiz I.F.,
- Tillmann C.,
- Liesenfeld K.H.,
- Schafer H.G.,
- Stangier J.
- Weinz C.,
- Schwarz T.,
- Kubitza D.,
- Mueck W.,
- Lang D.
- Zhang D.,
- He K.,
- Raghavan N.,
- et al.
- Heidbuchel H.,
- Verhamme P.,
- Alings M.,
- et al.
- Spyropoulos A.C.,
- Douketis J.D.
- Sié P.,
- Samama C.M.,
- Godier A.,
- et al.
- Heidbuchel H.,
- Verhamme P.,
- Alings M.,
- et al.
- Healey J.S.,
- Eikelboom J.,
- Douketis J.,
- et al.
- Sherwood M.W.,
- Douketis J.D.,
- Patel M.R.,
- et al.
- Zhou W.,
- Zorn M.,
- Nawroth P.,
- et al.
- Eerenberg E.S.,
- Kamphuisen P.W.,
- Sijpkens M.K.,
- Meijers J.C.,
- Buller H.R.,
- Levi M.
- Körber M.K.,
- Langer E.,
- Ziemer S.,
- Perzborn E.,
- Gericke C.,
- von Heymann C.
- Pernod G.,
- Albaladejo P.,
- Godier A.,
- et al.
- Schiele F.,
- van Ryn J.,
- Canada K.,
- et al.
- Therapeutic and Bleeding Profiles of DOACs in Clinical Studies
- Bleeding Risk in Patients Treated with DOACs
- Monitoring Anticoagulation with the DOACs
- Peri-Procedural Management
- Interruption of DOACs and Switching Between Anticoagulants
- Recommended Management Strategies for Bleeding Associated with DOACs
- Management of Life-Threatening Bleeding
- Ongoing Clinical Studies