New Approaches for Stroke Prevention in Atrial Fibrillation

Mark J. Alberts, MD


December 30, 2009


Are there any alternative therapies to warfarin in the prevention of stroke for patients with atrial fibrillation?

Response from Mark Alberts, MD
Professor of Neurology, Department of Neurology, Northwestern University Feinberg School of Medicine, Chicago, Illinois; Director, Stroke Program, Northwestern Memorial Hospital, Chicago, Illinois

Atrial fibrillation is a common cause of cardioembolic stroke, particularly among the elderly. While warfarin is an effective therapy for the prevention of stroke in atrial fibrillation, the need for frequent monitoring of the INR, frequent medication adjustments, food and drug interactions, risk of bleeding, and patient/physician resistance has generally led to an under-utilization of warfarin in patients who might benefit from this therapy.

Alternative agents to warfarin have recently been studied. The ACTIVE A trial showed that in patients unable or unwilling to take warfarin, the combination of aspirin plus clopidogrel compared with aspirin alone reduced the risk for stroke by 28%, although it also increased the risk for major bleeding events. In my opinion, there was a net benefit in favor of dual antiplatelet therapy, because in general bleeding events are easier to treat than a massive ischemic stroke, which is a frequent scenario in strokes caused by atrial fibrillation.

Another recently published trial was the Randomized Evaluation of Long-Term Anticoagulant Therapy (RE-LY), which studied a direct-acting thrombin inhibitor (dabigatran in 2 doses) vs standard warfarin therapy. RE-LY was a large study with more than 18,000 patients randomized in a 3-arm design (dabigatran 110 mg twice daily, dabigatran 150 mg twice daily, warfarin with target INR 2.0 to 3.0). Although the dabigatran was administered in a blinded manner, open-label warfarin was used as the comparator. The primary endpoint was all strokes and systemic emboli. RE-LY was designed as a noninferiority study with a median follow-up of 2 years.

RE-LY was a positive study, with the low-dose dabigatran group proving to be noninferior to warfarin, while the high dose dabigatran group showed superiority to warfarin for the primary endpoint. Much of this benefit came in the form of a reduction in fatal or disabling strokes with high-dose dabigatran compared with warfarin (0.66%/yr vs 1.00%/yr, respectively, P = .005 for noninferiority). To the best of my knowledge, this is the first time that any agent has shown superiority to warfarin for stroke prevention in atrial fibrillation. Rates of hemorrhagic stroke were also lower in either dabigatran group compared with warfarin (0.10%-0.12%/yr vs 0.38%/yr, respectively, P < .001).

The benefits of dabigatran were somewhat tempered by a trend for more major bleeding with the high dose (3.11%/yr) compared with the low dose (2.71%/yr). The rate of major bleeding with high-dose dabigatran approached that of warfarin (3.36%/yr). Dabigatran was associated with a slight increase in the rate of myocardial infarction when compared with warfarin, with an absolute increase of about 0.2%/yr for either dose (P = .05-.07). A calculation of net clinical benefit, defined as the combined rate of major vascular events, major bleeding, and death, showed a significant benefit for high-dose dabigatran vs warfarin (6.91%/yr vs 7.64%/yr, respectively, P = .04). There was no evidence of liver toxicity with dabigatran, as had been seen with other direct thrombin inhibitors.

RE-LY did have some limitations. The overall event rates were quite low for the primary endpoint, perhaps reflecting a bias in patient selection toward relatively lower risk patients. The use of open-label warfarin may have introduced some biases in terms of bleeding events.

In routine clinical practice, dabigatran would have some obvious advantages to warfarin, including (1) fixed dosing, (2) no need for routine blood monitoring, (3) no apparent food and drug interactions, and (4) apparent enhanced efficacy and net clinical benefit. Ongoing studies of other agents such as the anti-Xa compounds for stroke prevention in atrial fibrillation will add more data for use to ponder. Of course, it will be up to the US Food and Drug Administration to decide which of these agents physicians will be able to use, and the manufacturers will decide the cost for these new medications. But I am hopeful that the future will bring more options that will be safe, effective, and easier for healthcare professionals and our patients to use for stroke prevention in atrial fibrillation.