1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. Acknowledgements
  6. References

Anticoagulants are the mainstay of therapy for thromboembolic diseases. In addition to the more traditional agents, new oral anticoagulants, including dabigatran etexilate, rivaroxaban, and apixaban, have been shown to be effective across several indications. Bleeding is a serious complication associated with any anticoagulant, but many of the traditional parenteral and new oral agents do not currently have specific antidotes. This review describes available and future options for the reversal of the effects of anticoagulants, in particular the new oral agents and discusses current management strategies for bleeding events in clinical practice. Am. J. Hematol. 2012. © 2012 Wiley Periodicals, Inc.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. Acknowledgements
  6. References

Anticoagulation is effective in reducing the risk of thromboembolism, and heparins and vitamin K antagonists (VKAs), such as warfarin, have been the anticoagulant mainstays for the past half century. Unfractionated heparin (UFH) is a parenterally administered anticoagulant that exerts its effect by binding to antithrombin, thereby catalyzing the inactivation of multiple factors in the coagulation cascade. It also binds to cells and other plasma proteins, leading to unpredictable pharmacokinetic and pharmacodynamic properties. Patients receiving therapeutic doses of UFH require routine laboratory monitoring and are also at risk of developing heparin-induced thrombocytopenia (HIT). In current practice, UFH has been replaced by the low molecular weight heparins (LMWHs) and fondaparinux for many indications, owing to their superior pharmacokinetic and pharmacodynamic properties [1]. However, LMWHs and fondaparinux also require parenteral administration, which can be inconvenient, especially in the outpatient setting. LMWHs may also be associated with a risk of HIT, although the risk is substantially lower than with UFH [1, 2]. Warfarin, which is administered orally, has a narrow therapeutic window, multiple food and drug interactions, unpredictable pharmacokinetics and pharmacodynamics, and a slow onset and offset of action. In addition, as a result of the considerable variability in warfarin sensitivity between patients, regular coagulation monitoring and periodic dose adjustments are required to maintain patients within the target international normalized ratio (INR) range [3]. This requirement for monitoring can be both costly and inconvenient.

The difficulties surrounding the management and practicalities of the established anticoagulants have led to the development of new agents that have been shown to be more effective and/or safer for some indications. The direct thrombin inhibitor dabigatran and the Factor Xa inhibitors rivaroxaban and apixaban are the agents that are at the most advanced stages of development. For indications such as prevention of venous thromboembolism after orthopedic surgery, treatment, and secondary prevention of venous thromboembolism (VTE) and stroke prevention in patients with atrial fibrillation (AF), the new agents have been studied without coagulation monitoring in adults with normal or mildly abnormal renal function. Although these new anticoagulants can address the unmet needs of physicians and patients, no anticoagulant has demonstrated efficacy without increasing bleeding risk [4]. Serious bleeding in patients receiving anticoagulant therapy presents a significant therapeutic challenge, requiring rapid restoration of hemostasis. This problem concerns those agents currently used in clinical practice in addition to the new anticoagulants that have reached the market in some countries or are at advanced stages of clinical development [4]. There are antidotes available to reverse the anticoagulant effects of warfarin and UFH; however, there are not yet specific agents available to completely neutralize the anticoagulant activity of LMWHs, despite their widespread use in clinical practice. Furthermore, although reversal agents for the new anticoagulants are under development, there have been none approved for clinical use.

With age being a major risk factor for VTE [5], and because elderly patients are more likely to suffer from conditions that require chronic anticoagulation, anticoagulant use is likely to increase given the growing elderly worldwide population. It will, therefore, be increasingly important to assess the risk of bleeding in individual patients prior to the initiation of the new agents and to be aware of the options available to reverse major bleeding events. This review will present the options currently available for the reversal of anticoagulation of well-established and new anticoagulants.

New oral anticoagulants

New oral anticoagulants have been developed for both the prevention and treatment of thromboembolism, in addition to the prevention of stroke and systemic embolism, where the small-molecule agents targeting thrombin or Factor Xa have shown clinical benefits. A number of reviews have described the results of clinical trials with these agents previously and so will not be discussed further here [6–8].

Direct thrombin inhibitors

Inhibition of thrombin is an attractive target in the development of anticoagulants, owing to the pivotal role of thrombin in hemostasis. Thrombin facilitates the conversion of fibrinogen to fibrin, activates Factors V, Factor VIII, and platelet-bound Factor XI, which generate more thrombin. Thrombin also activates platelets and Factor XIII, and thrombin bound to thrombomodulin activates protein C, an anticoagulant protein [9]. Dabigatran etexilate is administered orally and is the prodrug of the active compound dabigatran, a direct thrombin inhibitor. It is rapidly absorbed, has a bioavailability of 6.5%, a half-life of up to 11–14 hr in healthy, elderly subjects, and no reported food interactions (Table I) [10].

Table I. Comparison of the Pharmacokinetic Features of Oral Factor Xa and Thrombin Inhibitors [10–15]
ParameterApixabanRivaroxabanDabigatran etexilate
  1. CYP3A4 = cytochrome P450 3A4; NR = not reported; t½ = half-life; tmax = time to maximum plasma concentration.

Route of administrationOralOralOral
Molecular weight (Da)460436628
Drug–drug interactionsPotent inhibitors of CYP3A4Potent inhibitors and inducers of CYP3A4 or P-glycoproteinPotent inducers of P-glycoprotein
Oral bioavailability50% in humans (10 mg)80–100% in humans (10 mg)6.5% in humans (220 mg)
tmax (hours)1–32–40.5–2
t½ (hr)8–15 (healthy volunteers)5–9 (healthy volunteers)12–14 (healthy volunteers)
Effect of foodNRDelayed absorption with foodDelayed time to peak plasma concentrations (by 2 hr)
EliminationApproximately 25% is excreted via the kidneys and the remainder (about ∼ 55%), via the fecal routeOne-third (unchanged) in the urine, two-thirds undergoes metabolic degradation, of which half is excreted via the kidney and half via the hepatobiliary routeAfter intravenous dosing, eliminated primarily in the urine (85%); fecal excretion accounts for 6% of the administered dose

Direct factor Xa inhibitors

Factor Xa is also an attractive antithrombotic target as the rate-limiting factor in the generation of thrombin [16]. Oral direct Factor Xa inhibitors inhibit both free Factor Xa and Factor Xa associated with the prothrombinase complex [11], but they allow residual thrombin to carry out its other important functions [16]. Rivaroxaban has high oral bioavailability (∼ 80–100%) [14], a half-life of 5.7–9.2 hr [12], minimal risk of interactions with food, and few clinically relevant interactions with other drugs [14]. Apixaban has oral bioavailability of 66%, minimal potential for drug–drug interactions, and a half-life of 8–15 hr [17]. The pharmacokinetic characteristics of the drugs are presented in Table I.

Based on evidence from phase III clinical trials, rivaroxaban, dabigatran etexilate, and apixaban have been approved in a number of indications. Rivaroxaban has been approved in Europe, the US, and Canada for prophylaxis of deep vein thrombosis (DVT) in adults undergoing elective hip and knee replacement surgery. It has also been approved for the prevention of stroke and systemic embolism in adult patients with nonvalvular AF (with one or more risk factors, such as congestive heart failure, hypertension, age =75 years, diabetes mellitus, prior stroke or transient ischemic attack) in the US and Europe and for the treatment of DVT and prevention of recurrent DVT and pulmonary embolism (PE) following an acute DVT in adults in Europe. Dabigatran has also been approved in the US and Europe for the prevention of stroke and systemic embolism in patients with AF and one or more risk factors including previous stroke, transient ischemic attack, or systemic embolism, left ventricular ejection fraction <40 %, symptomatic heart failure, =New York Heart Association (NYHA) Class 2, age =75 years and age =65 years associated with one of the following: diabetes mellitus, coronary artery disease, or hypertension.

The risk of hemorrhage with anticoagulants

The major potential complication of anticoagulant therapy is an increased risk of bleeding, which can be severe and difficult to control [4]. Age is a significant risk factor for bleeding, and surgery or trauma can also increase this risk [4, 18]. Deliberate or accidental anticoagulant overdose may also lead to hemorrhagic complications. In addition, there are specific risk factors associated with the use of each anticoagulant. For patients receiving warfarin, the intensity of therapy is an independent predictor of the risk of bleeding, in addition to comorbid conditions such as hypertension, cerebrovascular disease, ischemic stroke, or diabetes [4]. Pharmacogenetic factors, such as certain polymorphisms of cytochrome P450 2C9 (the enzyme that metabolizes the S-racemer of warfarin) may also increase the risk of major bleeding [4, 19]. Bleeding risk may also be related to the intensity of anticoagulation with heparin, but there is limited data to prove this [4]. There is some evidence that the risk of bleeding with heparin is higher when it is administered by intermittent intravenous infusion compared with continuous intravenous infusion, and with concomitant use of acetylsalicylic acid or thrombolytic therapy [4]. Should a hemorrhagic complication occur during the period of anticoagulation, an antidote antagonizing the pharmacodynamic effect of the anticoagulant or an active prohemostatic agent may be required.

Despite an increased risk of bleeding associated with these agents, rates of bleeding remain generally low in clinical trials. In a clinical trial for the initial treatment of DVT, the rates of major bleeding for rivaroxaban compared with enoxaparin/VKA were 0.8 and 1.2%, respectively [20]. When dabigatran was compared with warfarin in this indication, rates of major bleeding were 1.6 and 1.9%, respectively [21]. Rates of major bleeding were also similar when these agents were compared with warfarin in patients with stroke and systemic embolism (rivaroxaban vs. warfarin, 3.6% vs. 3.4%; dabigatran (150 mg bid) vs. warfarin, 3.1% vs. 3.4%; apixaban vs. warfarin, 2.1% vs. 3.1%, respectively) [22–24]. Although all anticoagulants are associated with an increased risk of bleeding, the actual risks of bleeding in clinical practice (in contrast to carefully controlled clinical trials) cannot be defined accurately until postmarketing surveillance and experience with long-term exposure becomes available.

Coagulation monitoring

When major bleeding occurs, the ability to assess or measure levels of the anticoagulant is essential to establish whether bleeding is the consequence of high drug levels or arises for another reason, such as bleeding from a discrete anatomic site, or some other etiology (e.g., coagulopathies resulting from liver disease or disseminated intravascular coagulation). Although there is a lack of standardized commercial assays for the new anticoagulants, tests are available to measure the anticoagulant effects or drug levels of commonly used anticoagulants. In the case of warfarin, the prothrombin time (PT) test is used to calculate the INR [3]. In the case of LMWHs and fondaparinux, anti-Factor Xa assays can be used to measure drug levels, and the activated partial thromboplastin time (aPTT) can be used to monitor the effects of UFH [1].

Although the new oral anticoagulants have being developed without the requirement for coagulation monitoring and is not routinely recommended, coagulation assays may be required in certain situations. From the studies conducted with dabigatran, assays using the ecarin clotting time seem to be the most sensitive and accurate assay [25, 26]; however, this test is not available routinely in clinical practice. Thrombin time determinations are more widely available but are generally too sensitive for the clinically relevant plasma concentration range [25]. Dabigatran prolongs the aPTT, but the effects are not dose dependent. Therefore, although a prolonged aPTT may indicate the presence of dabigatran, it will not provide an exact level of anticoagulant activity [25]. For apixaban, the dilute PT test and HepTest have been shown to be sensitive [26]. Several commercially available assays have been evaluated as potential methods to measure levels of rivaroxaban. Of the assays evaluated, PT assays, when used with rivaroxaban calibrators and controls (expressed as ng ml−1), or a sensitive reagent such as Neoplastin (expressed as seconds) can be useful [26, 27]. It is important to note that conventional INR measurements used for monitoring VKA therapy should not be used for assessment of rivaroxaban anticoagulation. Anti Factor Xa chromogenic assays are more sensitive and specific for measurement of rivaroxaban plasma concentrations (ng/ml) when used with a rivaroxaban calibration curve [28]. The results should be interpreted in relation to timing of dose administration and pharmacokinetics. Rivaroxaban calibrators and controls are now commercially available for clinical use (Technoclone®, Diagnostica Stago).

Currently available options for the reversal of traditional anticoagulants

Warfarin therapy inhibits vitamin K epoxide reductase, an enzyme that recycles vitamin K from the oxidized to the reduced form, which then catalyzes the gamma-carboxylation of glutamic acid residues on procoagulant clotting factors (prothrombin and Factors VII, IX, and X) and anticoagulant clotting factors (proteins C, S, and Z) [29]. Normal hemostasis is attained when the levels of these factors are within the normal range.

There are several options available for the management of patients receiving traditional anticoagulants who present with an acute bleeding episode. In emergency situations, bleeding episodes are typically managed using tamponade and aggressive volume and blood product replacement until hemostasis is restored [30]. In addition, rapid reversal of anticoagulant effects can be achieved using specific reversal agents.

To reverse the anticoagulant effect of warfarin in patients with serious or life-threatening bleeding, administration of vitamin K along with either fresh frozen plasma (FFP) or prothrombin complex concentrates (PCCs) is recommended by the American College of Chest Physicians (ACCP) [3]. Recombinant Factor VIIa (rFVIIa) has also been shown to be effective in the reversal of INR values, but this does not restore normal hemostasis [31] and is not currently approved for this indication [3]. The use of one or more of these agents depends on the urgency of the situation and the magnitude of the INR elevation. However, each of these reversal options is associated with certain drawbacks. The full effect of vitamin K administration on the reversal of the INR has been shown to take up to 24 hr; hence, for immediate reversal, it must be used together with faster-acting agents such as FFP or PCC [32]. Vitamin K administration must be repeated to maintain the reversal achieved by FFP or PCC to keep the INR within a safe range [3]. A recent study by Crowther et al. demonstrated that although vitamin K was capable of correcting high INRs (4.0–10.0), it had little effect on the frequency of overall or major bleeding events compared with placebo [33]. FFP can achieve rapid reversal but relatively rapid, large-volume infusions are required [30], which may lead to volume overload [34]. It also carries the rare risks of allergic [32] or anaphylactic reactions and septicemia or blood-borne infection [32, 34]. In addition, there may be time delays owing to preparation time required for thawing, delivery and administration, which has obvious implications in patients requiring immediate therapy [34].

Consequently, PCCs are now recommended over FFP in many national guidelines for the reversal of warfarin activity [35]. PCCs contain varying amounts of prothrombin and Factors VII, IX, and X, as well as proteins C and S [36], which may lead to variability in dosing recommendations, depending on the specific PCC used. PCCs containing Factor VII are known as four-factor PCCs, whereas products without Factor VII are known as three-factor PCCs [37]. In Europe, four-factor PCCs are commercially available (Octaplex®, Octapharma, Vienna, Austria; Beriplex® P/N, CSL Behring, Germany); however, only three-factor PCCs are currently licensed for use in the US (Bebulin® VH, Baxter, CA; Profilnine® SD, Grifols Biologicals, CA). PCCs provide more rapid and complete factor replacement, require lower-volume infusions (with only a single dose required to completely and rapidly reverse the effect of warfarin in most patients) and are not thought to carry infectious risk, owing to viral inactivation [35]. Beriplex® P/N was developed specifically for the rapid reversal of warfarin anticoagulation in patients requiring immediate hemostatic control [38], and its efficacy and long-term safety have been shown in clinical settings [39]. Antiinhibitor coagulation complex concentrates, such as Factor VIII inhibitor bypassing activity (FEIBA), are PCCs that undergo in vitro activation during manufacturing, resulting in an increased amount of activated vitamin K-dependent clotting factors [36].

The current ACCP guidelines recommend PCCs to reverse the effects of warfarin in patients with serious or life-threatening bleeding with an elevated INR [3]. PCCs have been associated with a risk of thrombotic complications and disseminated intravascular coagulation [32, 40], and there is some uncertainty regarding an effective yet safe dose [35]. In hemophilia A patients, a comparative study has shown that the rate of thrombotic adverse events over a 3-year period was significantly lower with FEIBA compared with rFVIIa [41]. In addition, the rate of thrombotic complications in patients who received FEIBA for emergency warfarin reversal was similar to that reported in patients who received nonactivated PCCs [42]. Treatment-related adverse events have been shown to be low or absent with both three- and four-factor PCC use in emergency situations [38, 43]; however a direct comparison of their efficacy and safety has not yet been performed. In some cases, use of three-factor PCCs alone may not be advised; a study showed that PCC use alone was not as effective in lowering the supratherapeutic INR of patients compared with a PCC–fresh frozen plasma combination [44]. rFVIIa has proved safe and effective in hemophilia patients with inhibitors, but its use in other indications is off-label, with limited clinical trial evidence available. A systematic review of clinical studies involving off-label use of rFVIIa showed there was no significant change in mortality rates, but the risk of thromboembolism increased with adult cardiac surgery and intracranial hemorrhage [45]. rFVIIa is a relatively expensive agent that may require the administration of multiple doses owing to its short half-life [30].

Protamine sulfate can rapidly and completely reverse the anticoagulant effects of UFH [1]. Although there is no proven method for completely reversing the effect of LMWH, protamine sulfate is recommended if reversal is required [1]. However, its use has been associated with an increased risk of allergic reactions, including anaphylaxis in patients who have previously received protamine sulfate-containing insulin [46, 47]. There is also an increased risk of respiratory problems and severe cardiovascular adverse reactions, such as hypotension or bradycardia, although these adverse reactions can be minimized by administering protamine at a reduced rate [1, 34].

Assessing the effectiveness of reversal agents will be necessary in the course of treatment. In vitro tests, such as PT, have variable sensitivity to the levels of each coagulation factor: they are most sensitive to Factor VII and less sensitive to prothrombin. Therefore, normalization or near-normalization of PT may not adequately reflect the levels of reversal achieved in vivo by vitamin K, FFP, or PCCs.

New potential agents for the reversal of anticoagulation

There are no specific antidotes for some of the older anticoagulants on the market, in addition to the new oral agents dabigatran etexilate, rivaroxaban, and apixaban. However, several recent studies have shown there are potentially effective agents for the reversal of one or more of these anticoagulants. Following a serious bleeding event, the requirement for a reversal agent may depend upon the half-life of the anticoagulant agent being used. Anticoagulants with short to intermediate half-lives, such as the new oral agents, have a potential safety advantage over longer-acting drugs, such as warfarin; therefore, an antidote may not be required for most patients with normal renal function. However, even for drugs with short half-lives, an agent that is capable of reversing the effects of the anticoagulant would be beneficial in the event of an overdose or if patients present with an acute bleeding episode, require emergency surgery or have or acutely develop renal dysfunction (which prolongs the half-lives of the new oral agents).

rFVIIa (NovoSeven®; NovoNordisk, Princeton, NJ) and activated PCCs, such as FEIBA, have received attention as potential prohemostatic agents for the reversal of targeted anticoagulants. The rationale for investigating these compounds as reversal agents is their potential to enhance thrombin generation and ultimately fibrin formation. Recently, modified plasma-derived or recombinant Factor Xa have also been investigated, which could act as universal antidotes for the reversal of anticoagulation of all current Factor Xa inhibitors, including antithrombin-dependent agents [48, 49].

Recombinant activated factor VII

rFVIIa was originally developed as a Factor VIII or IX “bypassing” agent for the treatment of bleeding episodes in hemophiliac patients who have developed inhibitors to Factors VIII and IX [50, 51]. rFVIIa is a vitamin-K dependent glycoprotein, structurally similar to human Factor VIIa [52]. It is approved in the US for patients with inhibitors to Factor VIII and IX and also gained approval for the treatment of hereditary Factor VII deficiency, albeit using lower doses.

The ability of rFVIIa to enhance hemostasis has led to increased attention as a potential treatment for profuse or life-threatening hemorrhage in settings other than hemophilia. Based on the anecdotal evidence available, it is being used on an off-label basis for patients with traumatic, surgical, and coagulopathic bleeding [53, 54], and has reportedly helped achieve hemostasis when massive transfusions of blood products proved ineffective [30]. It has also demonstrated effectiveness in reversing the anticoagulant effects of fondaparinux in ex vivo studies on human whole blood [55], although complete reversal of the inhibition of thrombin generation was not achieved in vitro [56, 57]. In healthy volunteers, rFVIIa normalized coagulation times and thrombin generation during fondaparinux therapy [58], but there is limited evidence of its use in clinical practice [59, 60]. A recent study in eight patients receiving fondaparinux who were bleeding or experiencing cardiovascular collapse showed that rFVIIa managed to successfully control bleeding in four patients [61]. Few randomized trial results are available to guide the administration or dosing of rFVIIa [54, 62], and there is particular concern related to its potential thrombogenicity in high-risk patients requiring anticoagulant therapy [63]. Furthermore, there have been few large-scale clinical trials validating the safety and efficacy of rFVIIa [54] except in patients with inhibitor-associated hemophilia [64, 65]. In addition, there is no definitive satisfactory laboratory test to monitor the safety or efficacy of rFVIIa [66]. The US product label has been updated recently to include a black box warning regarding rFVIIa use outside the approved indications [52].

The use of rFVIIa for the reversal of the new oral anticoagulants

In vitro studies showed that rFVIIa did not reverse the effects of dabigatran, as measured by lag time, time to peak thrombin generation (tmax), peak thrombin generation, and endogenous thrombin potential [67]. Similar results were obtained in vivo in a model of intracerebral hemorrhage, in which rFVIIa had no effect on the prevention of excess hematoma expansion induced by dabigatran [68]. However, in a more recent study using blood from healthy volunteers, rFVIIa successfully reversed the actions of dabigatran by reducing clot initiation time to baseline levels [69]. There have been several studies involving rFVIIa and direct Factor Xa inhibitors. In vitro, rFVIIa has been shown to partially reverse the effects of rivaroxaban on thrombin generation, leading to an increase in the concentration of rivaroxaban required to double the lag time and tmax, and an increase in the half maximal inhibitory concentration of rivaroxaban for the endogenous thrombin potential. Peak thrombin generation was affected less by rFVIIa [67]. Reversal of the hemostatic effects of rivaroxaban by rFVIIa was also assessed in baboons receiving high-dose rivaroxaban (higher than normal clinical practice) [70]. After infusion of rFVIIa, the prolongation of bleeding time was reversed rapidly, and the PT was also shortened. rFVIIa significantly decreased the ear bleeding time in a rabbit model of bleeding and thrombosis following rivaroxaban treatment in addition to decreasing the aPTT and thromboelastographic clotting times compared with animals who received rivaroxaban only [71].

Prothrombin complex concentrates

As previously described, PCCs are recommended for the reversal of warfarin activity. Studies involving the use of PCCs for reversal of the new anticoagulants are discussed below.

The use of prothrombin complex concentrates for the reversal of the new oral anticoagulants

A study investigated whether PCC could neutralize the anticoagulant effects of high-dose rivaroxaban in rats. The prolongation of bleeding time induced by high-dose rivaroxaban was almost completely reversed by PCC. PT prolongations were partially reversed by coadministration of PCC [72]. PCC significantly reversed the anticoagulant effects of rivaroxaban, but did not completely reverse bleeding, in a rabbit model of bleeding and arterial thrombosis [71]. Importantly, in a randomized study of healthy male volunteers conducted by Eerenberg et al., a single bolus of PCC (50 IU kg−1) significantly reversed the anticoagulant effects of rivaroxaban (20 mg twice daily), with PT restored to baseline levels and endogenous thrombin potential normalized. The effects were sustained for 24 hr and suggest that PCC is a potential agent for the reversal of rivaroxaban activity [73]. In the same study, PCC failed to reverse the effects of dabigatran [73]. PCC however was successful in reversing the effects of dabigatran in a model of intracerebral hemorrhage, where excess hematoma expansion was prevented, and tail vein bleeding time was reversed, both of which were dose-dependent following increasing doses of dabigatran (4.5–9 mg kg−1) [68]. It is possible that the dose of PCC required for optimal reversal of dabigatran may be higher than that for rivaroxaban. Overall, PCC has been shown to be an effective reversal agent of direct Factor Xa inhibitors in healthy volunteers; however, the potential for thrombogenicity remains. The efficacy of PCC in patients on a new oral anticoagulant who are actively bleeding or at high risk for bleeding will need to be evaluated in future studies.

Factor VIII inhibitor bypassing activity

FEIBA has been used successfully for more than 30 years in controlling bleeding in hemophiliac patients who have developed inhibitory antibodies against Factor VIII or Factor IX [74]. FEIBA increases thrombin generation by targeting different sites of the coagulation cascade, of which a major target is the prothrombinase complex (Factor Xa–Factor Va plus Ca2+ on a phospholipid surface), in the common pathway of the blood coagulation cascade [74]. Low doses of FEIBA have been shown to reverse the anticoagulant effect of fondaparinux in vitro [57], but further evidence in the clinical setting will be required. Experience in hemophilia patients suggests that FEIBA is generally well-tolerated; infusion-related adverse events, thromboembolic complications, and disseminated intravascular coagulation are rare [75], although there is an increased risk of thrombosis and disseminated intravascular coagulation in patients with cirrhosis.

The use of factor VIII inhibitor bypassing activity in the reversal of the new oral anticoagulants

FEIBA has been shown to partially neutralize the effect of high-dose rivaroxaban in studies in rats and baboons (2 and 0.6 mg kg−1 rivaroxaban, respectively) [70, 76]. In the rat study, the administration of FEIBA significantly reduced bleeding time prolongation induced by high-dose rivaroxaban [76]. In the baboon study, bleeding time was doubled following administration of high-dose rivaroxaban but returned to baseline after infusion of FEIBA. The prolongation of PT was also shortened after the infusion of FEIBA [70]. Rivaroxaban-induced PT prolongation was significantly reversed by coadministration of FEIBA [76]. In a study using whole blood from healthy volunteers, FEIBA-VHR reduced clot initiation time to one-third of baseline following incubation with dabigatran in vitro, which was suggestive of a potent effect [69].

Plasma-derived and recombinant factor Xa

Recently, modified plasma-derived and recombinant Factor Xa (pd-Factor Xa and r-Factor Xa, respectively) were investigated as potential antidotes for Factor Xa inhibitors. These proteins have the potential to be universal antidotes for both small-molecule direct inhibitors and antithrombin-dependent indirect Factor Xa inhibitors [48]. Both pd-Factor Xa and r-Factor Xa antidotes were modified to lack catalytic and membrane-binding activities but still maintain high affinity for Factor Xa inhibitors [48, 49].

The use of plasma-derived and recombinant factor Xa for reversal of new anticoagulants

Preliminary studies with pd-Factor Xa antidote showed a dose-dependent reversal of the anticoagulant effect of rivaroxaban and apixaban [48]. pd-Factor Xa antidote also demonstrated reversal of the in vitro anticoagulant activity of enoxaparin [48]. Kinetic measurements showed that the antidote retained low nanomolar affinity (Kd = 0.7–3.0 nM) for rivaroxaban and apixaban [49]. With both pd-Factor Xa and r-Factor Xa antidotes, there was no effect on Factor Xa activity in the absence of inhibitors, demonstrating that both proteins are capable of restoring hemostasis through reversal of Factor Xa inhibitor-mediated anticoagulation [48, 49].

Other antidotes in development

An antithrombin variant (AT-N135Q-Pro394), which combines two mutations, has been developed as a potential antidote for heparin derivatives [77]. Preliminary results suggest that this antithrombin variant could be used routinely in the future as an antidote for heparin derivatives [77]. Results of a recent study have shown that an engineered monoclonal antibody (clone 22) was a highly potent and specific inhibitor of dabigatran activity both in vitro and in vivo [78].

A summary of the reversal agents currently in development and their action on the new oral anticoagulants is shown in Table II.

Table II. Summary of Reversal Agents Currently in Development, and Their Action on New Oral Anticoagulants
Reversal agentApixabanRivaroxabanDabigatran etexilate
  1. aPTT = activated partial thromboplastin time; EC50 = half maximal reversal of inhibitory activity; FEIBA = Factor VIII inhibitor bypassing activity; N/A = not applicable; PCC = prothrombin complex concentrate; PT = prothrombin time; rFVIIa = recombinant Factor VII.

Plasma-derived Factor Xa antidoteEC50 attained at a concentration of 122 nM in vitro [48]EC50 attained at a concentration of 49 nM in vitro [48]N/A
Recombinant Factor Xa antidoteEC50 attained at a concentration of 45 nM in vitro [48]EC50 attained at a concentration of 17 nM in vitro; normalized PT [48]N/A
PCCNo published studies to date• Partially reversed PT prolongation; reversed inhibition of thrombin formation; shortened bleeding time in vivo [72]; Reversed rivaroxaban activity in vivo; however, bleeding not reversed [71]; Reversed PT prolongation and normalized endogenous thrombin potential in healthy volunteers [73]• Prevented hematoma expansion and reversed tail bleeding time in vivo [68]; No effect on reversal of dabigatran activity in healthy volunteers [73]
rFVIIaNo published studies to date• Partial reversal of thrombin generation in vitro [67]; Reduced clotting times, aPTT and bleeding time in vivo [71]; Reversed PT prolongation and shortened bleeding time in vivo [70]• Ineffective in vitro and in vivo [67,68]; Reduced clot initiation times to baseline in vitro [69]
FEIBANo published studies to date• Reduced clot initiation time to one-third of baseline in vitro [69]; Partial neutralization of activity in vivo [70,76]• No published studies to date
Monoclonal antibodiesNo published studies to dateNo published studies to dateClone 22 completely inhibited dabigatran activity in vitro and in vivo [78]

Managing bleeding in clinical practice

The advantages of the new anticoagulants, including oral administration and no requirement for routine coagulation monitoring, are leading to their uptake in clinical practice as alternatives to heparins and VKAs. However, the new agents do not yet have specific antidotes to reverse their anticoagulant effect in the event of a bleeding emergency, urgent surgery, or severe overdose.

In a study of the requirement for the reversal of heparin and warfarin in the initial treatment of VTE (1,877 patients), 9.6% of patients experienced a clinically relevant bleeding episode, of which 14.4% received some type of antidote. However, only 2.5% experienced a major bleeding event, of which 41% of received an antidote. Vitamin K was given to 23 (1.2%) patients, one (0.05%) patient received protamine sulfate and seven (0.4%) patients received fresh frozen plasma. Though this suggests that reversal agents will be required infrequently, the availability of specific antidotes for the new oral agents would nevertheless be desirable and beneficial [79].

In the case of emergency surgery or an invasive procedure, anticoagulation should be stopped before surgery, depending on the risk of bleeding for the patient and their level of renal impairment. For anticoagulants such as rivaroxaban and dabigatran, it is recommended to withhold treatment until hemostasis is restored [10, 14]. It is advised that dabigatran treatment should be stopped 1–5 days before surgery [80], and for those receiving rivaroxaban, it should be stopped at least 24 hr before the procedure [14]. For apixaban, the use of an assay, such as the Rotachrom® anti-Factor Xa assay, may be useful in exceptional situations where knowledge of apixaban exposure may help to inform clinical decisions [15]. Strategies such as delaying the next drug dose or discontinuation (as required in all cases of major bleeding), activated charcoal (in case of overdose or very recent administration), mechanical compression, surgical intervention (e.g., for refractory exsanguinating gastrointestinal bleeding or later after the bleeding site has been identified), or blood product transfusion should control bleeding in most cases. If these strategies fail, a reversal agent antagonizing the pharmacodynamic effect of the anticoagulant may be helpful.

Measuring the reversal of the anticoagulant effect is complicated by the fact that there are no specific, accurate, and readily available assays for the new oral anticoagulants. In addition, surrogate markers (such as INR) are not always indicative of the risk of bleeding in anticoagulated patients [33, 81]. Agents such as rFVIIa and FEIBA have the potential to overcome, or at least partially ameliorate, the hemorrhagic effects of high doses of the new direct Factor Xa or thrombin inhibitors. To our knowledge, there have been no reports to date of accidental or intentional overdose; however, these agents could potentially be of use in such unlikely events, or if patients receiving the anticoagulant at therapeutic levels present with a critical bleeding episode, i.e., gastrointestinal/intracranial hemorrhage. Despite a favorable safety profile in the setting of hemophilia, where these agents were originally developed, there is little clinical data for rFVIIa and FEIBA use in patients with hemorrhage in association with the new anticoagulants. Off-label use of rFVIIa has expanded considerably in recent years, but the evidence supporting its effectiveness for bleeding episodes remains limited and anecdotal [30]. The safety of rFVIIa during off-label use needs to be addressed in specific patient populations before an evidence-based conclusion can be made regarding its safety profile. In addition, there is limited evidence available to guide rFVIIa dosing [54, 62], and there is no definitive laboratory test to monitor its effectiveness [66]. Concerns have been raised regarding the complications associated with the use of rFVIIa and FEIBA, such as potential thrombogenicity and the risk of adverse events, such as disseminated intravascular coagulation [63, 75]. Serious complications, such as myocardial infarction, are rare [63, 82] and are often associated with overdose [82]. Another important issue for clinical use may be cost: rFVIIa, for example, is a relatively expensive agent—the total cost of rFVIIa as a first-line therapy in hemophilia patients is estimated at $36,000, according to a literature-based decision model [83].

PCCs and plasma-derived or recombinant proteins have the potential to reverse the effect of anticoagulation with Factor Xa inhibitors, and preliminary clinical data appears promising, but, again, further studies are needed to establish evidence-based conclusions [48, 72, 73]. Blood transfusions and adequate supportive care should be the preferred options. FFP infusions should only be administered to patients with coagulopathies, because only patients whose coagulation factors are abnormally low would benefit from this treatment. For other patients, FFP is likely to be no more effective than the usual means of hemodynamic management, such as volume replacement. There has been considerable experience with emergency warfarin reversal in clinical practice, where vitamin K administration has been shown to be effective, safe, and convenient in noncritical situations [84]. Vitamin K may also be used in elderly patients (with INR >5.0), which corresponds to the ACCP guidelines [85]. PCC is also effective for warfarin reversal in patients who are bleeding or are in need of emergency surgery, but may be associated with a low risk of thromboembolism [86]. In life-threatening situations, such as warfarin-induced intracranial hemorrhage, immediate administration of a nonactivated PCC, ideally a four-factor PCC, should be administered in addition to vitamin K; FFP should also be given if only a three-factor PCC is available. Although agents such as rFVIIa, PCC, and FEIBA show potential as reversal agents for new anticoagulants, until further retrospective efficacy and safety data are available, the use of these agents should be restricted to those rare, life-threatening situations in patients for whom the drug offers significantly more benefit than risk and for who other options have failed.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. Acknowledgements
  6. References

There is no universal antidote for the reversal of anticoagulation. Even when specific antidotes are available, such as for the VKAs, they may not be optimal for emergency use. LMWHs have been used routinely and successfully in some patient populations for periods of months to years (i.e., cancer-associated and pregnancy-associated VTE, VTE patients refractory to warfarin), despite not having a specific reversal agent. If the new oral anticoagulants are prescribed appropriately, the lack of a specific antidote may not be a great drawback in most bleeding situations because there will be relatively few circumstances in which a reversal agent will be required. Until specific antidotes for the new oral agents become available, some reversal strategies such as PCC have shown promising results and could provide useful options for the management of severe bleeding episodes in clinical practice.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. Acknowledgements
  6. References

The author would like to acknowledge Kelly Farrell who provided editorial assistance with funding from Bayer HealthCare Pharmaceuticals and Janssen Research & Development, LLC.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Conclusion
  5. Acknowledgements
  6. References