The trusted source for
healthcare information and
Abhi Mehrotra, MD, MBA, FACEP, Vice Chair, Strategic Initiatives & Operations, Medical Director, Hillsborough Campus ED, University of North Carolina Department of Emergency Medicine.
Timothy D. Buff, MD, Emergency Medicine, University of North Carolina, Chapel Hill.
Leah M. Hatfield, PharmD, BCPS, Lead Clinical Pharmacist Specialist, Emergency Medicine, University of North Carolina Medical Center, Chapel Hill.
Chad S. Kessler, MD, MHPE, Clinical Associate Professor, Emergency Medicine and Internal Medicine, Duke University, Durham, NC.
Charles V. Pollack, MD, FACEP, FAAEM, Professor and Senior Advisor for Interdisciplinary Research and Clinical Trials, Department of Emergency Medicine, Sidney Kimmel Medical College of Thomas Jefferson University, Philadelphia.
Statement of Financial Disclosure
To reveal any potential bias in this publication, and in accordance with Accreditation Council for Continuing Medical Education guidelines, we disclose that Dr. Farel (CME question reviewer) owns stock in Johnson & Johnson. Dr. Stapczynski (editor) owns stock in Pfizer, Johnson & Johnson, Walgreens Boots Alliance Inc., GlaxoSmithKline, Bristol Myers Squibb, and AxoGen. Dr. Hatfield (author) is a retained (nonpromotional) consultant for a product under development for Portola Pharmaceuticals. Dr. Mehrotra (author) is a stockholder and board member for MedScribes. Dr. Pollack (peer reviewer) has received grant/research support and is a retained consultant for: Boehringer Ingelheim, Daiichi Sankyo; retained consultant for Janssen Pharma, BMS/Pfizer Alliance. Dr. Schneider (editor), Ms. Fessler (nurse planner), Dr. Buff (author), Dr. Kessler (author), Ms. Mark (executive editor), Ms. Coplin (executive editor), and Mr. Landenberger (editorial and continuing education director) report no financial relationships with companies related to the field of study covered by this CME activity.
Case: A 73-year-old female presented to the emergency department (ED) after she tripped and fell, striking her head. On arrival, she appeared to be in no acute distress, was alert and oriented, and was neurologically intact with a Glasgow Coma Scale score (GCS) of 15. Vital signs: blood pressure 162/87 mmHg, heart rate 80/min and irregularly irregular, respiratory rate 16/min, and pulse oximetry 96% on room air. She did have a scalp hematoma but without overlying laceration or palpable bony abnormality. She was currently taking warfarin for atrial fibrillation and her international normalized ratio (INR) in the ED was 3.9. A head CT was obtained rapidly and was negative for acute intracranial bleed or skull fracture.
Five oral anticoagulants are available in the United States: warfarin (Coumadin®), dabigatran (Pradaxa®), rivaroxaban (Xarelto®), apixaban (Eliquis®), and edoxaban (Savaysa®). Warfarin has been in clinical use since 1954, and for more than 55 years, it was essentially the only oral anticoagulant available. In the past five years, new or novel oral anticoagulants (NOACs) have been released for use. Alternatively, these agents also are known as non-vitamin K antagonist oral anticoagulants, indicating that their mechanism of activity does not involve antagonism of the synthesis of the vitamin K-dependent coagulation factors, or direct oral anticoagulants (DOACs), reflecting that they directly inhibit one of the factors involved in the coagulation cascade. In the United States, four NOACs or DOACs currently are available: dabigatran, rivaroxaban, apixaban, and edoxaban.
The primary indication for emergency reversal of oral anticoagulation is on-going major or life-threatening bleeding. A secondary indication is when an emergent surgical or invasive procedure is required and reversal is indicated to prevent periprocedural bleeding. For both warfarin and the DOACs, a primary reversal strategy for emergent reversal is replacement of the deficient or inhibited coagulation factor(s).1 This is most intuitive in cases of warfarin-related bleeding, in which multiple coagulation factors are deficient. DOAC-treated patients have only one factor inhibited, and replacement strategies involve administration of enough factor to overcome inhibition induced by the DOAC present in the circulation. Many institutions have developed standardized anticoagulation reversal protocols to streamline patient care and minimize potential for errors.2
In many circumstances, the challenge is not the method of reversal, but rather the decision of whether to initiate reversal. Temporary withholding of anticoagulation is almost always generally advisable. It should be inherently obvious that any life-threatening bleed should be reversed promptly with the best available agent. This is typified by a patient on anticoagulation who presents with non-compressible bleeding (e.g., intracranial hemorrhage, severe gastrointestinal bleed, multi-system trauma), or hemodynamic instability secondary to exsanguination. The debate arises when a patient has the propensity to decompensate but is currently hemodynamically stable or presents with non-life-threatening bleeding. (See Table 1.) When possible, the provider can have a discussion with the patient or caregiver regarding the risks and benefits of immediate anticoagulation reversal. The major risk of anticoagulation reversal is primarily in patients who are on therapeutic anticoagulation to prevent recurrent thromboembolism; urgent reversal exposes the patient to occurrence of such an event. Physicians should always consider the alternatives, such as manual, balloon, or tampon compression; topical pro-hemostatic therapy (tranexamic acid, oxymetazoline); and/or supplementing with phytonadione (warfarin-induced bleeding only).3-6
As noted, warfarin was the first oral anticoagulant developed. It was the agent used almost exclusively in studies that identified the benefits and risks of treating thromboembolic conditions, as well as studies that investigated the duration of treatment necessary to reduce the risk of thromboembolic recurrence. Warfarin is the agent with which the DOACs were compared in Phase III studies evaluating their efficacy and safety.
Warfarin is readily absorbed and metabolized, with elimination primarily by the liver. (See Table 2.) Warfarin is an antagonist for the vitamin K-dependent carboxylation of several of the proteins involved in the coagulation system: four thrombotic and two antithrombotic.7 (See Table 3.)
Upon initiation of warfarin therapy, the serum levels of the prothrombotic factors drop according to their half-lives: initially factor VII (FVII), followed by factor IX (FIX), factor X (FX), and then prothrombin (FII). The levels of protein C also decrease, and because its half-life is shorter than three of the prothrombotic coagulation factors, there is an initial increase in thrombotic activity at the start of warfarin therapy. This transient hypercoagulable state usually is counteracted by the use of bridging therapy with a non-vitamin K-dependent parenteral anticoagulant, such an unfractionated heparin (UFH) or low molecular-weight heparin (LMWH), initiated concurrently with warfarin and continued until the INR is within the therapeutic range: 2.0 to 3.0 for stroke prevention in atrial fibrillation and treatment of venous thromboembolism (VTE).
Warfarin has several drug and dietary interactions that can either potentiate or reduce its anticoagulant effects. (See Table 4.) In clinical practice, routine monitoring is required to establish the daily dose necessary to achieve the therapeutic goal. Some patients have a difficult time achieving a stable level of anticoagulation, perhaps due to differences in the time of day when the drug is ingested, proximity to meals, variations in diet, and interaction with drugs that are either initiated, adjusted, or discontinued.8
A benefit of using warfarin is having the ready availability of agents that can correct the anticoagulant effect by replenishing the deficient coagulation factors, a mechanism properly termed “repletion” rather than “reversal.” Phytonadione helps naturally replenish factors II, VII, IX, X, proteins C and S by supporting epoxide reductase in the liver. On average, oral phytonadione takes 6-24 hours and intravenous (IV) phytonadione takes two to six hours for demonstrable action, with maximal effect at 12-24 hours to decrease the risk of bleeding.3,6,9 Because of the prolonged onset, both oral and IV dosing are more helpful in correcting an asymptomatic supratherapeutic INR and are less relevant for rapid reversal, but are used as an adjunct medication in conjunction with emergent repletion agents. The anticoagulant effect of warfarin can be more readily reduced by using agents that replace the reduced coagulation cascade factors: fresh frozen plasma (FFP), recombinant factor VIIa, and prothrombin complex concentrate (PCC) products.
Warfarin reversal can be achieved in a number of ways. (See Table 5.) The simplest is to stop the daily dose. (See Table 6.) Since warfarin is a competitive vitamin K antagonist, pharmacologic doses of phytonadione or vitamin K1 will overcome the anticoagulant effect of excess warfarin, but it takes time to synthesize new factors. Historically, FFP has been the preferred agent for rapid reversal, but the rapidity of reversal is slower than desired and the completeness of reversal is variable.10 A more rapid and thorough response is seen with 3- and 4-factor PCC.9,11,12 Recombinant FVIIa also may be effective,13-15 although it is not FDA approved for warfarin reversal.16 For life-threating hemorrhage due to warfarin-induced coagulopathy, a combination approach is recommended. (See Table 7.)
Fresh Frozen Plasma. For many years, FFP was the standard therapy for coagulation factor repletion because it provides multiple factors (although not in concentrated amounts) and is widely available. It does, however, require cross-matching, thawing time, and consideration of the patient’s volume status.6,12,18 Each unit of FFP is roughly 200-250 mL, and 10-15 mL/kg is recommended in an adult patient to reverse the effects of warfarin anticoagulation in cases of life-threatening bleeding. This can be a significant intravenous volume, leading to side effects such as transfusion associated circulatory overload (TACO) and immunogenic complications of transfusion-related acute lung injury (TRALI).18,19
Coagulation Factor VIIa (recombinant). Coagulation factor VIIa (recombinant) or rFVIIa administration initially gained traction in the early 2000s because of its targeted approach. Factor VII has the shortest biological half-life of all coagulation factors, so it is often the rate-limiting reagent in patients with multiple coagulation factor deficiencies. Significant reduction in both INR and bleeding time occurs after rFVIIa administration to patients taking warfarin who present with life-threatening bleeding.13,14,20 Studies comparing rFVIIa to PCC found no significant difference in the short-term anticoagulation reversal as assayed by laboratory analysis.14,20
In addition, rFVIIa does have some drawbacks. It received an FDA black-box warning because of an increase in arterial thromboembolism when given to patients with hemophilia.16,20 Evidence regarding the thrombosis risk of rFVIIa in anticoagulation reversal is mixed; some studies show no difference in thrombosis risk,15 while others show an increased risk.14,21 This risk is thought to be secondary to rFVIIa’s short half-life, potentially creating a rebound increase in coagulability.13 The current consensus is that rFVIIa is a second- or third-line agent for emergent warfarin or dabigatran reversal. This is because rFVIIa is only one of the depressed factors in warfarin therapy, it is not the inhibited factor in dabigatran therapy, better reversal agents are available, and it has the associated risk of thrombosis.
Prothrombin Complex Concentrates. Prothrombin complex concentrates (PCCs) are freeze-dried preparations of the vitamin K-dependent coagulation factors involved in assembly of the prothrombin complex: factors X, IX, VII, and thrombin. These products are commercially available in three variations: 3-factor, 4-factor, and activated 4-factor versions.18,22-24 (See Table 8.) Three-factor versions contain small amounts of FVII, whereas 4-factor versions contain clinically relevant amounts of FVII. The activated 4-factor version contains thrombin, FIX, and FX in primarily inactivated form, and FVII mainly in the active form. Four-factor PCCs contain variable amounts of protein C, protein S, and antithrombin. Since PCCs are derived from human plasma, there is variation in the amounts of each factor in individual vials. The vials are assayed for FIX activity and dosing usually is according to the degree of FIX activity desired.
PCC products carry a small risk of inducing thrombosis, estimated to be between 1% and 2%.25 Because of variation in the clinical conditions for which these agents are used, it is not possible to precisely compare risk among the different PCC products.
Head-to-head trials have shown 4-factor PCC to have equal anticoagulation reversal to 3-factor but overall lower thrombosis risk.26,27 Experimental literature indicates that 4-factor PCC can reverse the anticoagulant effects of warfarin, dabigatran, and rivaroxaban.9 The benefits of either PCC formulation compared to FFP are small volume, no need for cross-matching, and quick therapeutic response. The reversal duration with warfarin-induced hemorrhage often is temporary because the half-life of infused factors can be shorter than the elimination half-life of warfarin, but coadministration of IV vitamin K can help replenish intrinsic production of coagulation factors to extend anticoagulation reversal.1,11,12
For reversal of warfarin-induced coagulopathy, the current strategy is a PCC dose based on the estimated depletion of coagulation factors as determined by the pretreatment INR and body weight; with greater depletion as noted by a higher INR and increasing body weight, a larger dose is used. The manufacturer’s prescribing information for 4-factor PCC has dosing recommendations based on reversal of the coagulopathy to an INR < 1.3.23 (See Table 9.) Normalization of the coagulopathy to this degree may not be necessary in many cases, and it is primarily reserved for life-threatening hemorrhage. In addition, a systematic review of PCC protocols using 8 to 50 units FIX/kg to reverse warfarin-induced coagulopathy found no superior dosing regimen, suggesting that smaller doses may be just as effective as those currently recommended.28
Dabigatran initially was approved in the United States in 2010 for preventing atrial thromboembolism (e.g., embolic stroke) in patients with non-valvular atrial fibrillation. With time, approved uses of dabigatran have expanded to include deep venous thrombosis (DVT) and pulmonary embolism (PE).29,30
Dabigatran is a direct-acting inhibitor of thrombin, activated FII or FIIa. (See Table 10.) This drug is administered as dabigatran etexilate mesylate, which is absorbed as dabigatran etexilatre ester and subsequently hydrolyzed to form the active agent dabigatran. The purported advantages of dabigatran compared with warfarin are reduced dietary and drug interactions, the lack of a need for routine monitoring tests, and the reduced incidence of serious bleeding (especially intracranial).31-35 These advantages also have been demonstrated in real-world experience.36-41 The one adjustment with dabigatran dosing is for patients with impaired renal function as determined by a calculated creatinine clearance (CrCL) below 30 mL/min.
Dabigatran has some drug interactions of concern. The most important of these interactions are with verapamil and amiodarone, both of which increase the serum concentration of dabigatran. This effect can be minimized by a two-hour gap between ingestion of dabigatran and the other agents.
Because dabigatran is predominately excreted unchanged by the kidneys, maintaining a robust urine flow or initiating hemodialysis will hasten removal. Idarucizumab is a specific antibody Fab fragment that binds dabigatran and renders it inactive. For life-threatening hemorrhage due to dabigatran-induced coagulopathy, idarucizumab is the most effective treatment. (See Table 11.) Clinical reports describe apparent reversal of the dabigatran-induced coagulopathy with both rFVIIa and aPCC, whereas FFP has no demonstrable effect.6,18,20,42-45
Idarucizumab. Idarucizumab is a humanized monoclonal antibody fragment (MW 47.8 kDa) that binds dabigatran with a much higher affinity, about 350-fold, than dabigatran’s binding to thrombin.46 (See Table 12.) Therefore, idarucizumab will bind and neutralize both free and thrombin-bound dabigatran. Idarucizumab has no demonstrable procoagulant or anticoagulant effects, no endogenous targets, and no Fc receptor binding, all of which result in a low risk of adverse effects.47 Studies in healthy young volunteers, older volunteers, and volunteers with renal insufficiency found that IV idarucizumab has a rapid onset of action, with reversal of the anticoagulant effect of therapeutic doses of dabigatran in a few minutes after completion of the IV infusion and an elimination half-life of about 47 minutes.47 Like other small proteins, idarucizumab is filtered into the urine with a portion that is excreted unchanged and a portion undergoing catabolism in the renal tubules.
A Phase III study found that idarucizumab 5 g IV completely reversed the anticoagulant effect of dabigatran in patients with serious bleeding or who required an urgent procedure.48 The majority of patients had laboratory evidence of an anticoagulated state, and the reversal effect from idarucizumab was evident within minutes and was sustained for up to 72 hours. In those patients in the group with serious bleeding who could be evaluated for continued hemorrhage, cessation of bleeding was noted after a mean of 11.4 hours.
Following publication of this study, idarucizumab underwent expedited FDA review and was approved on Oct. 16, 2015. Based on its ability to bind to dabigatran (Pradaxa), the registered trademark name for idarucizumab is Praxbind®.
Rivaroxaban and apixaban initially were released for use in the United States in 2011 and 2012, respectively. In January 2015, the FDA approved an additional drug in this class, edoxaban, for treatment of DVT and PE, and stroke prevention in patients with non-valvular atrial fibrillation.7 A difference with edoxaban compared to rivaroxaban and apixaban is its minimal metabolism and elimination primarily unchanged in the urine. In a trial comparing edoxaban with warfarin, an increased rate of ischemic stroke was seen in patients with a creatinine clearance > 95 mL/min receiving edoxaban. The obvious but unverified assumption is that patients with robust renal function excrete the drug more rapidly, producing a diminished anticoagulant effect. Thus, edoxaban is not recommended for patients with a creatinine clearance > 95 mL/min.7
Rivaroxaban, apixaban, and edoxaban are direct inhibitors of activated factor X (FXa). (See Table 13.) They inhibit not only free FXa, but also FXa already incorporated into the prothrombinase complex, composed of FXa, FVa, plasma membrane phospholipids, and calcium ion.
These FXa inhibitors have drug interactions so there should be at least a two-hour gap between administration of these agents and ingestion of verapamil, amiodarone, HIV protease inhibitors, and azone antifungals.
The factor Xa inhibitors have effects on routine coagulation tests that vary according agent, test, and assay reagent used.50-54 There is no value using the standard PT, aPTT, or TT to monitor therapeutic effect of oral FXa-inhibitors. With careful choice of reagents and analyzers using the PT, it is possible to determine if an anticoagulative effect is present with rivaroxaban and apixaban, but such testing modifications need to be requested specifically and are not routinely available.54 Since rivaroxaban, apixaban, and edoxaban competitively inhibit FXa, the chromogenic anti-FXa activity assay can be used to measure their anticoagulant effect, and with calibration for the specific drug, the serum level.50 Such a measurement may be useful to assess the degree of drug-induced anticoagulation before either invasive procedures or initiation of fibrinolytic therapy in order to avoid excessive hemorrhage.
The specific reversal agent for the FXa inhibitors is andexanet alfa, currently under FDA review. On Aug. 18, 2016, the FDA sent a Complete Response Letter to Portola Pharmaceuticals, the developer of andexanet alfa, requesting additional information related to the manufacturing process and additional data supporting inclusion of edoxaban and enoxaparin in the prescribing information. Because andexanet alfa is an FDA-designated Breakthrough Therapy, approval for use in the United States is anticipated within the next few months.
Until then, there is a paucity of data regarding which agent consistently reverses rivaroxaban, apixaban, or edoxaban.55 Limited studies suggest that aPCC would be expected to have greater effectiveness than 4-factor PCC or rFVIIa at reversing the hemorrhagic state producing serious bleeding associated with rivaroxaban, apixaban, or edoxaban.55,56 (See Table 14.)
Andexanet alfa. Andexanet alfa is a truncated version of FXa produced by recombinant protein technology that is enzymatically inactive because it lacks a 34-amino acid fragment required for assemblage into the prothrombinase-complex. Andexanet alfa acts as a decoy, binding and competitively sequestering oral and parenteral FXa inhibitors away from active native FXa. In Phase II studies with healthy volunteers taking therapeutic range doses of rivaroxaban or apixaban, andexanet alfa reduced anti-FXa activity by more than 90% and restored thrombin generation within two to five minutes.57 Because the half-life of andexanet alfa is about an hour, the drug is administered as a bolus plus an infusion for two additional hours.
Andexanet alfa doses used in clinical trials have varied and a recommended effective dose awaits further analysis and FDA approval. It is anticipated that the recommended dose of andexanet alfa will vary according to the specific FXa inhibitor agent being reversed because of the differences in therapeutic doses for these drugs. Andexanet alfa is currently undergoing a Phase III trial to determine efficacy and safety in bleeding patients.58 This study does not include an evaluation of FXa-inhibitor-treated patients who need reversal to undergo a procedure.
Andexanet alfa also reverses the effect of the anti-FXa parenteral anticoagulants heparin, LMWH, and fondaparinux.59,60 Dose recommendations, and understanding of efficacy and safety, await additional studies.
A stepwise approach should include:
In life-threatening bleeding, such as intracranial hemorrhage, administer the emergency reversal agent without waiting for coagulation test results. In non-life-threatening bleeding, obtain the appropriate test to determine if there is a clinically relevant anticoagulant effect from the reported oral anticoagulant and to guide therapy.50,56,61 (See Table 15.)
Many of the studies supporting anticoagulation reversal are based on ability to correct abnormal coagulation tests. Although the different coagulation tests are a good measurement of coagulability, more research needs to be directed toward patient-centered outcomes, such as duration of hospitalization and mortality. Intracranial bleeding is the most devastating complication from oral anticoagulants, and research needs to target neurological outcomes, both short- and long-term, of patients who present with a bleed while on anticoagulation and receive rapid reversal.
There is a targeted therapy for DOAC reversal under study that has the potential to reverse multiple agents. Ciraparantag or aripazine is a nonspecific binder of multiple anticoagulants, including dabigatran and the factor Xa inhibitors as well as unfractioned heparin and low-molecular weight heparin.64
Warfarin long has been the primary agent for outpatient anticoagulation. The DOACs have several FDA-approved indications, but their use has been somewhat impeded by the lack of a rapid reversal agent.3,5 Over the past five years, 4-factor PCC has been found useful for rapid reversal of both supratherapeutic INR in warfarin use and laboratory hemostasis in use of DOACs,1,12,49 but more research needs to be done on patient-centered outcomes.
Case: There are no established guidelines on whether the provider should rapidly reverse the warfarin-induced anticoagulation in this patient. This is different from hemophilia, for which there are guidelines that “all significant head trauma, with or without hematoma, must be treated promptly with the major dose of factor replacement before any diagnostic tests.”65 This patient does not have indications for a life-threatening bleed at the moment, but her chances of deterioration are high given her age and supratherapeutic INR. In this case, she was observed and started to deteriorate and she became obtunded. A repeat head CT showed an intracranial bleed three hours after arrival. Four-factor PCC and IV vitamin K1 were given with rapid correction in her INR. There was no further neurologic deterioration, and subsequent CT scans showed stabilization in the size of the hematoma. The patient made a slow recovery over two weeks and was discharged to a rehabilitation facility.
Financial Disclosure: To reveal any potential bias in this publication, and in accordance with Accreditation Council for Continuing Medical Education guidelines, we disclose that Dr. Farel (CME question reviewer) owns stock in Johnson & Johnson. Dr. Stapczynski (editor) owns stock in Pfizer, Johnson & Johnson, Walgreens Boots Alliance Inc., GlaxoSmithKline, Bristol Myers Squibb, and AxoGen. Dr. Hatfield (author) is a retained (nonpromotional) consultant for a product under development for Portola Pharmaceuticals. Dr. Mehrotra (author) is a stockholder and board member for MedScribes. Dr. Pollack (peer reviewer) has received grant/research support and is a retained consultant for: Boehringer Ingelheim, Daiichi Sankyo; retained consultant for Janssen Pharma, BMS/Pfizer Alliance. Dr. Schneider (editor), Ms. Fessler (nurse planner), Dr. Buff (author), Dr. Kessler (author), Ms. Mark (executive editor), Ms. Coplin (executive editor), and Mr. Landenberger (editorial and continuing education director) report no financial relationships with companies related to the field of study covered by this CME activity.