Share this article on:

InFocus: Dabigatran's Potential to Replace Warfarin

Roberts, James R. MD

doi: 10.1097/01.EEM.0000398955.71161.eb

Each month brings a slew of new drugs to the market, making it very difficult for even the most diligent CME-enthralled clinician to keep up with the latest innovation. A recent addition to our hospital's pharmacy is a medication called dabigatran (Pradaxa). This drug holds promise to replace warfarin (Coumadin) as the oral anticoagulant of choice for numerous indications. It has been primarily studied for a venous thrombolism and stroke prevention in nonvalvular atrial fibrillation. This is a nifty drug that appears to be physician- and patient-friendly. If it turns out to be clinically effective and safe, real progress will have been made.

Dabigatran is a direct thrombin inhibitor. This is a new class of anticoagulants that bind directly to thrombin to block interactions with substrates. One might recall hearing cardiologists talk about argatroban, another thrombin inhibitor. Argatroban is useful as a replacement for heparin in patients who develop heparin-induced thrombocytopenia (HIT). I came across dabigatran through my activities on the pharmacy committee of our hospital, and I have seen a few patients who have had their Coumadin replaced with this rather interesting agent.

Two mega-trials introduced dabigatran to the market as a potential replacement for warfarin for treating acute thromboembolism (DVT/PE) and for reducing embolic stroke in patients with atrial fibrillation.

Dabigatran versus Warfarin in Patients with Atrial Fibrillation

Connolly S, et al

New Engl J Med


This article summarizes the RE-LY trial (Randomized Evaluation of Long Term Anticoagulation Therapy) comparing two fixed doses of dabigatran with warfarin in patients who had atrial fibrillation and were at risk for stroke. The primary outcome was either stroke or systemic embolization. This pharmaceutical-sponsored study was designed to be a noninferiority trial. Patients with extant chronic atrial fibrillation were randomized to either 110 mg or 150 mg of dabigatran twice a day via a blinded protocol or open-label INR-adjusted dose warfarin. The follow-up was two years. This rather gargantuan study recruited about 18,000 patients from almost 1,000 clinical centers in 44 countries. Warfarin was given in doses required to maintain a therapeutic INR (measured monthly) at 2.0-3.0. Concomitant low-dose aspirin (less than 100 mg/day) was permitted with all regimens, and about 40 percent took this antiplatelet medication.

Stroke or systemic embolization, the primary outcome, occurred in 1.69 percent each year in the warfarin group compared with 1.53 percent each year for low-dose dabigatran (110 mg BID) and 1.11 percent each year for high-dose dabigatran (150 mg BID).

The rate of a major bleeding was 3.36 percent each year with warfarin compared with 2.7 percent each year (P=0.003) with low-dose dabigatran and 3.11 percent per year (P=0.031) with high-dose dabigatran. Compared with warfarin, bleeding was statistically less with both doses of dabigatran. The rate of hemorrhagic stroke was higher in the warfarin group versus low-dose and high-dose dabigatran, 0.38 percent per year versus 0.12 percent/0.1 percent per year (P<0.001), respectively. The mortality rate was higher in the warfarin group compared with the low-dose and high-dose dabigatran groups: 4.13 percent per year versus 3.75 percent/3.64 percent per year (P=0.13/0.051) respectively.

When all the statistics were evaluated, patients with atrial fibrillation given low-dose dabigatran experienced similar rates of stroke and systemic embolization (noninferior) compared with warfarin, but experienced a lower rate of major hemorrhage. When the higher dose was used, dabigatran was also associated with a lower rate of stroke and systemic embolization, but caused an incidence of major hemorrhage similar to warfarin. It was noted, however, that the rate of bleeding with the warfarin in this study was higher than reported in prior studies.

While the results are not overwhelming, they are quite encouraging, and other benefits of dabigatran are extremely attractive. Warfarin, the quintessential vitamin K antagonist, is cumbersome to use, has multiple interactions with foods and other drugs, and requires frequent INR monitoring. Such vagaries contribute to a high rate of discontinuation, inadequate anticoagulation, and significant time and effort for clinicians and patients alike.

For similar stroke-related prevention in AF, alternatives to warfarin are clopidrogrel plus aspirin. This regimen is more effective than aspirin alone but less effective than warfarin. Subcutaneous heparins are effective for stroke prevention but associated with a higher risk of bleeding.

While the significance was uncertain, patients given low- and high-dose dabigatran had a slightly higher rate of myocardial infarction than warfarin-treated individuals (0.72%/0.74% per year vs. 0.53% per year, respectively).

Patients with a creatinine clearance less than 30 ml/min and those with liver disease were excluded. The drug is renally excreted, and many older patients will have liver or kidney issues so further clarification prior to widespread use is needed. Dabigatran had about a 12 percent incidence of dysphagia, but there were no significant issues with liver toxicity as were seen with similar antithrombin drugs. The authors conclude that the rate of intracranial hemorrhage with either dose of dabigatran was less than a third the rate of brain bleed with warfarin, without a reduction in the efficacy against ischemic stroke. This is an important advantage for dabigatran, if it is confirmed. The twice-daily dosing of dabigatran may reduce some variability in the anticoagulation effect seen with daily warfarin.

Comment: While the modest statistical advantage of dabigatran over warfarin for reducing stroke and systemic embolization in atrial fibrillation may not be gargantuan, warfarin is clearly a difficult and problematic drug. Dabigatran, however, seems to be quite simple to prescribe and take. The vitamin K content of food varies widely, making dietary configurations difficult for warfarin-treated patients to understand or follow. Many drugs, such as simple antibiotics, have an interaction with warfarin that makes the drug significantly more potent and can increase bleeding. Only a few days of trimethoprim/sulfa (Bactrim), for example, can cause significant elevation in the INR leading to bleeding. (Arch Intern Med 2010;170[7]:617.) The variability and difficulty in maintaining anticoagulation with warfarin results in complications that every physician knows, including bleeding, thrombosis, and increased risk of death. One downside of dabigatran is that it likely must be dosed based on renal function, but this issue requires more investigation. Likely, further refinements in dosing will be coming in the future.

In an accompanying editorial, “Can We Rely on RE-LY?” (catchy title), use of dabigatran was endorsed. (New Engl J Med 2009;361[12]:1200.) The bottom line was that higher dose dabigatran prevented more AF-related strokes than warfarin, and the low dose caused fewer hemorrhages. The editorial notes that warfarin prevents about 65 percent of AF-related strokes, but it is only prescribed to about two-thirds of appropriate candidates. With warfarin, the drug and dietary interactions, the inconvenience of INR monitoring, and risk of hemorrhage are not miniscule, and a new oral anticoagulant with fewer downsides is welcome. A slight increased risk in myocardial infarction with dabigatran was commented upon, but not enough information is available to draw strict conclusions. Statistically, the number needed to treat with dabigatran to experience one myocardial infarction was 500.

Overall, dabigatran prevented more strokes than warfarin, with fewer hemorrhages at the lower dose. Although the higher dose was more effective, the lower dose appears somewhat safer. Some patients taking amiodarone will experience an elevated level of dabigatran, which might potentially cause increased bleeding.

Thankfully, most emergency physicians don't have to deal with prescribing warfarin or following the INR. We do see the complications of over- and underuse, and they are not insignificant. GI or renal bleeds due to warfarin over-anticoagulation are omnipresent. Spontaneous brain bleeds are usually seen with INR in the therapeutic range, and are not only related to marked perturbations in clotting. If you haven't seen a spontaneous intracranial hemorrhage in a patient on warfarin, you just haven't worked enough shifts. It's common to have a patient sent from the office with an astronomically high INR, so EPs are well aware of the problems in regulating this anticoagulant. I can't imagine having my INR checked on a regular basis, or having to worry if I eat too many green leafy vegetables that might contain vitamin K and inactivate my warfarin.

The specific mechanism of direct thrombin inhibition is daunting and extremely complex. See Di Nisio, et al, for an erudite explanation. (New Engl J Med 2005;353[10]:1028.) This executive summary says direct thrombin inhibitors suppress thrombus growth by inhibiting the action of fibrin-bound and free thrombin. Thrombin converts fibrinogen to fibrin, so no thrombin, no fibrin clot.

Most clinicians will support some sort of anticoagulation in AF to reduce embolic strokes. A perusal of the dabigatran statistical analysis reveals that one would have to use dabigatran (150 mg BID) rather than warfarin in 357 patients to prevent one non-hemorrhagic stroke. The number of patients who would need to be treated with dabigatran vs. warfarin to prevent one hemorrhagic stroke is 370. Given the cost of the new drug, one wonders what will happen. If I were marketing this drug, I would stress compliance and patient and physician convenience. While this drug seems to be a bit too good to be true, its effect on liver function has not been well defined.

The best monitoring therapy has not been clearly established either. Essentially, clinicians are not able to monitor thrombin inhibition at all, and dabigatran levels are not available so there is no easy way for clinicians to evaluate the anticoagulant effect of dabigatran. Having to add these tests routinely could cancel out the benefit of eschewing INR checks. Importantly, there is no antidote to reverse rapidly the anticoagulant effects of direct thrombin inhibitors. One can administer vitamin K and fresh frozen plasma to those over-anticoagulated with warfarin, although they may require 12 to 24 hours for this intervention to be effective. Fortunately, the major anticoagulant effects of direct thrombin inhibitors should disappear in a similar timeframe.

One note of caution: When patients are started on warfarin for atrial fibrillation, it takes some time for oral anticoagulation to be effective. Full anticoagulation with dabigatran occurs with the first dose. Because warfarin first inactivates proteins S and C, a stint of hypercoagulability may ensue. Heparin is used as a bridging maneuver to attain rapid anticoagulation and ameliorate the potential for warfarin to increase thrombosis during the first few days of use. Using heparin and dabigatran together is not a good idea, but old habits die hard and the reflex of ordering bridging heparin might be ingrained enough to be problematic for the neophyte (AKA resident). Bottom line: Don't let the housestaff write for heparin when anticipating the use of dabigatran in the hospital.

Dabigatran versus Warfarin in the Treatment of Acute Venous Thromboembolism

Schulman S, et al

New Engl J Med


This is the RE-COVER study, a double blind randomized trial comparing six months of treatment for acute venous thromboembolism with dabigatran versus dose-adjusted warfarin. This was also an industry-sponsored noninferiority trial.

Pathology included deep vein thrombosis of the legs or pulmonary embolism. Only one dose of dabigatran was studied: 150 BID. Initially, and prior to randomization, patients were treated with a parenteral anticoagulant, either unfractionated heparin IV or low-molecular-weight heparin subcutaneously. The period of parenteral anticoagulation was quite long — a mean of nine days — before oral therapy using dabigatran or warfarin. Because parenteral therapy was used first, the value of dabigatran as monotherapy for acute thromboembolism could not be proven. Importantly, the heparins were stopped after the institution of dabigatran. The primary outcome was the six-month incidence of recurrent, symptomatic, or objectively proven venous thromboembolism or death. The safety endpoints included bleeding, acute coronary syndrome, or liver dysfunction.

As seen with the atrial fibrillation RE-LY study, a fixed dose of dabigatran was as effective as warfarin for acute DVT/PE, and had a safety profile that was similar to warfarin without the need for laboratory monitoring. The rate of recurrent thromboembolism was similar in the dabigatran and warfarin groups (2.4% and 2.1%, respectively). Major bleeding was similar in both groups, as was the incidence of death, acute coronary syndrome, and liver dysfunction.

Comment: Score another one for dabigatran with regard to venous thromboembolism. The drug seems to work as well as warfarin to treat DVT/PE, with similar side effects but with an impressive ease of administration. No INRs to check, no food to avoid, and very few known drug-drug interactions. Importantly, these patients were all anticoagulated with a heparin prior to the administration of warfarin or dabigatran, so all patients were in the hospital.

Optimism over the findings could result in a misguided attempt to mandate outpatient oral anticoagulant as initial therapy for DVT/PE. Currently self-administered fractionated heparins by patients discharged from the ED are attractive and seemingly possible in some centers, but it is nearly impossible, not economically feasible, and intellectually difficult for many patients in my practice. We can only hope that the day doesn't come when we have to discharge patients with acute DVT/PE simply because we have a pill they can take, and they don't have to go to the lab to get their INR or PTT checked.

Dabigatran is very expensive, many times the cost of warfarin (about $250-$350 per month). If one factors in physician and patient convenience and compliance, office visits, and lab testing, the cost could be acceptable. The drug is currently FDA-approved (2010) only to prevent stroke and systemic embolism in patients with nonvalvular AF, but has Canadian labeling for postoperative thromboprophylaxis (such as hip and knee replacement). There also are reports of using it before and after elective AF cardioversion. (Circulation 2011;123[2]:131.)

A few critically important questions remain. I am sure that drug-drug interactions will be found. Remember that amiodarone increases bleeding when used with dabigatran. How does one monitor and evaluate an overdose? What does one do about a nonreversible anticoagulant in emergency surgery and trauma patients? And what can you do for those with active bleeding, especially in the brain? Dabigatran is currently experiencing a pharmaceutical honeymoon. One wonders if honeymoon cystitis is not far behind.

Comments about this article? Write to EMN at

Dr. Roberts is the chairman of emergency medicine and the director of the division of toxicology at Mercy Catholic Medical Center, and a professor of emergency medicine and toxicology at the Drexel University College of Medicine, both in Philadelphia.

Back to Top | Article Outline

Reader Feedback:

Readers are invited to ask questions and offer experiences on InFocus topics. Please send comments to

© 2011 Lippincott Williams & Wilkins, Inc.