The Question

Given the recent U.S. Food and Drug Administration (FDA) approval of edoxaban, the fourth direct oral anticoagulant (DOAC) to come to the U.S. market since 2010, what should I know about this class of medications, and what is their role in patient care? Are there patients for whom these medications are not (or possibly not) appropriate?

My Response

This is an exciting time for clinicians caring for patients with (or at risk for) thrombosis, but the amount of clinical trial data, not to mention the dosing peculiarities, relevant to the DOACs, can be difficult to digest. Here, I will try both to provide take-home points about safety and efficacy relative to other therapies as well as to address some of the practical management questions I am frequently asked about DOACs.

Stroke Prevention in Atrial Fibrillation ✓ ✓ ✓ ✓ 
Venous Thromboembolism TreatmentAcuteExtended ✓*✓ ✓✓ ✓✓ ✓*- 
Knee Arthroplasty ✓ ✓ 
Hip Arthroplasty ✓ ✓ 
Stroke Prevention in Atrial Fibrillation ✓ ✓ ✓ ✓ 
Venous Thromboembolism TreatmentAcuteExtended ✓*✓ ✓✓ ✓✓ ✓*- 
Knee Arthroplasty ✓ ✓ 
Hip Arthroplasty ✓ ✓ 

*Edoxaban and dabigatran are approved for the acute treatment of venous thromboembolism only after an initial 5-day course of treatment with a parenteral anticoagulant (e.g., low-molecular-weight heparin, fondaparinux, heparin).

One or more of the DOACs has now been approved for three main indications: 1) the prevention of stroke and systemic embolism in patients with atrial fibrillation (AF), 2) the acute and extended treatment of venous thromboembolism, and 3) the prevention of venous thromboembolism (VTE) after total knee or hip arthroplasty (Table).

These medications are small molecules that inhibit their serine protease target by binding to its active site (thrombin in the case of dabigatran and activated factor X [FXa] in the case of rivaroxaban, apixaban, and edoxaban). There are some very important pharmacologic differences with warfarin: All these drugs rely, to varying degrees, on elimination by the kidneys. The prescriber must consult the package insert of a particular agent to determine whether his or her patient’s renal function would be a contraindication to treatment or mandate a dose adjustment. There are no important dietary interactions for the DOACs. A possible exception is that the bioavailability of rivaroxaban seems to be increased if it is taken with a meal; however, the clinical relevance of this is uncertain. Although the DOACs can interact with other medications, the number of potentially clinically important interactions is small compared to the corresponding number with warfarin.

My patient taking a DOAC needs to undergo an elective procedure. How far in advance of the procedure should I stop the medication, and do I need to provide “bridge” therapy?

Unlike vitamin K antagonists (VKAs), all of the DOACs have both a rapid onset of action and a relatively short half-life. DOACs induce maximal anticoagulant effect within two to three hours of the first dose and, unless renal function is impaired, will essentially disappear from the plasma 24 to 48 hours after the last dose. Therefore, periprocedural anticoagulation (“bridging”) with a parenteral anticoagulant is not necessary for a patient taking a DOAC unless the patient undergoes a surgery that will preclude swallowing or gastrointestinal absorption. Patients with renal impairment will need to interrupt the DOAC longer (see individual prescribing information for more details), but others can safely undergo most interventions as early as 24 to 48 hours after their last DOAC dose.

My patient has moderate renal impairment. Does that mean I cannot prescribe a DOAC?

Although all DOACs depend on renal clearance more than warfarin does, renal insufficiency is not a contraindication in all cases. Indeed, an analysis of the subgroup of patients with moderate renal impairment (creatinine clearance 30 to 50 mL/min) in the phase III clinical trials indicates that the DOACs (when dosed according to the manufacturers’ recommendations) are at least as safe as warfarin for such individuals.The thresholds at which dose-adjustment is required vary depending on the DOAC; most providers would avoid DOACs altogether in patients with severe renal impairment (creatinine clearance < 30 mL/min). That being said, the U.S. FDA has approved a dose adjustment for apixaban in patients on hemodialysis. Apixaban is the DOAC that depends least on renal clearance. On the other end of the spectrum of renal function, the U.S. FDA has recommended that edoxaban not be used in AF patients with excellent renal function (creatinine clearance > 95 mL/min) because of a subgroup analysis suggesting it may be less effective than warfarin to prevent stroke or systemic embolism in this subgroup.

When will antidotes for some or all of the DOACs be available? Should I avoid prescribing DOACs until antidotes reach the market?

The lack of experience managing DOAC-associated bleeding is an understandable concern. However, the absence of a dedicated antidote should probably not be a major consideration in the decision to use (or not use) this class of medications. Outcome data from randomized phase III clinical trials that included more than 100,000 patients show that, although no antidote was available, the likelihood of fatal bleeding was lower among the patients randomized to a DOAC than among the patients randomized to warfarin. Furthermore, taking a DOAC (as opposed to warfarin) was not associated with a higher likelihood of death among patients who experienced a major bleed. These seemingly counterintuitive observations are probably explained by several factors: 1) Despite the availability of vitamin K and plasma products, warfarin-associated major bleeding has a very high associated mortality; 2) compared with warfarin, the DOACs cause 50 percent fewer intracranial bleeds; and 3) the short half-life of the DOACs means that in many cases of major bleeding, no antidote or “reversal agent” is needed. These observations notwithstanding, antidotes and reversal agents in late-phase clinical development are likely to reach the U.S. market within the next two to three years. The DOAC antidotes that are furthest in the development process are andexanet (a factor Xa “decoy” molecule that is intended for patients on a FXa inhibitor) and idarucizumab (an antibody fragment targeting the direct thrombin inhibitor, dabigatran).

What should I do if consulted about a patient with DOAC-associated bleeding?

Many patients with DOAC-associated bleeding will be best managed with supportive care such as red blood cell transfusions and volume support; excellent renal perfusion will maximize the rate at which the anticoagulant effect dissipates. Unless a patient has compromised renal function, the plasma DOAC concentration (and corresponding drug effect) will have decreased by more than 50 percent six to 12 hours after the last dose. In the small minority of patients with DOAC-associated bleeding who require more aggressive efforts to counteract the anticoagulant effect immediately, one could consider prothrombin complex concentrates or recombinant activated factor VII.However, the thrombosis risk associated with these interventions is not trivial, and the rationale for their use is based entirely on animal and other preclinical models.

In which patients should DOACs be avoided?

In general, DOACs should be used only for approved indications. Because patients with mechanical prosthetic heart valves were not included in the phase III studies, and because dabigatran was inferior to warfarin in the only head-to-head comparison that has been done in this population,the DOACs should not be prescribed to patients with mechanical prosthetic heart valves. There are also subgroups of patients with (or at risk for) VTE where routine DOAC use should also be avoided pending further evidence. Because of their highly prothrombotic tendencies, patients with active cancer, heparin-induced thrombocytopenia, and bona fide antiphospholipid syndrome are other groups for whom I do not think DOACs should be a first option. Instead, we should work hard to enroll such patients in clinical trials that will provide evidence on which we can base future practice.


In summary, the DOACs offer hematologists and well-selected patients four oral alternatives to warfarin and other VKAs.4,5 Important ongoing trials will tell us more about how best to manage serious DOAC-related bleeding, whether DOACs can be safely used in patients with cancer-associated VTE, and what is the safest way to combine DOACs with antiplatelet agents (e.g., in patients with previous acute coronary syndromes). In the meantime, many patients for whom these medications are indicated (and for whom cost is not a barrier) will likely choose them over warfarin because of their convenience and impressive safety profile.


Harel Z, Sholzberg M, Shah PS, et al.
Comparisons between novel oral anticoagulants and vitamin K antagonists in patients with CKD.
J Am Soc Nephrol.
Siegal DM, Garcia DA, Crowther MA.
How I treat target-specific oral anticoagulant-associated bleeding.
Eikelboom JW, Connolly SJ, Brueckmann M, et al.
Dabigatran versus warfarin in patients with mechanical heart valves.
N Engl J Med.
Yeh CH, Gross PL, Weitz JI.
Evolving use of new oral anticoagulants for treatment of venous thromboembolism.
Dzeshka MS, Lip GY.
Non-vitamin K oral anticoagulants in atrial fibrillation: Where are we now?
Trends Cardiovasc Med.

Competing Interests

Dr. Garcia indicated no relevant conflicts of interest.