Warfarin Therapy for Atrial Fibrillation
ABSTRACT & COMMENTARY
Synopsis: Low fixed-dose warfarin plus aspirin is inferior to warfarin adjusted to maintain an INR of 2.0-3.0 for prevention of stroke or systemic embolism.
Source: Stroke Prevention in Atrial Fibrillation Investigators. Lancet 1996;348:633-638.
This study is the third in a series examining the value of various anticoagulation strategies in patients with nonvalvular atrial fibrillation. The hypothesis for the study was that a combination of low-intensity, fixed-dose warfarin plus aspirin would be as efficacious as adjusted-dose warfarin for prevention of stroke in high-risk atrial fibrillation patients. The study enrolled 1044 patients with nonvalvular atrial fibrillation and one of the following thromboembolic risk factors: congestive heart failure, left ventricular dysfunction, previous thromboembolism, hypertension, or female gender with age over 75. Patients randomly assigned to low-intensity fixed-dose warfarin received a dose of warfarin between 0.3 and 3.0 mg/d with a target INR of 1.2-1.5. These patients also received aspirin at a dose of 325 mg/d. Once the fixed warfarin dose was set, it was not varied unless subsequent INR monitoring showed a value greater than 3. The other group received adjusted-dose warfarin. The target INR was 2.0-3.0, with necessary adjustments made weekly at the beginning of the study and later monthly. The major end points for the study were ischemic stroke and systemic embolism.
The study enrolled a total of 1044 high-risk patients from 20 clinical centers. The group had a mean age of 82 years; 84% had continuous atrial fibrillation for more than one year, and 61% were male. No patients were lost to follow-up. Withdrawal from assigned therapy occurred at a rate of 8.2% per year in the combined therapy group compared to a rate of 5.6% per year in the adjusted dose warfarin group. The study was stopped early when a higher primary event rate was noted in the combined aspirin warfarin group. The combined therapy group demonstrated a rate of ischemic stroke or systemic embolism of 7.9% per year. This was significantly higher than the event rate of 1.9% per year noted in the adjusted-dose warfarin group. The risk factors of previous thromboembolism, systolic hypertension, and female gender plus age over 75 were associated with the highest event rates. Rates of major hemorrhage did not differ between the groups, being 2.4% per year with combination therapy and 2.1% per year with adjusted-dose warfarin. There was a slight, but not significant, increase in total mortality in the combined therapy group (7.2% vs 5.9%). The authors conclude that low fixed dose warfarin plus aspirin is inferior to warfarin adjusted to maintain an INR of 2.0 to 3.0 for prevention of stroke or systemic embolism.
COMMENT BY JOHN P. DiMARCO, MD, PhD
This is an important study that extends earlier observations made in previous trials. In the 1980s and early 1990s, a series of trials were performed that clearly demonstrated the value of warfarin anticoagulation in patients with nonvalvular atrial fibrillation. However, the lower threshold at which the benefit of anticoagulation would appear was not clearly defined. Since risk of hemorrhage and the cost of inconvenience of prothrombin-time monitoring are the major factors discouraging the use of warfarin, it was hoped that a low-intensity fixed-dose warfarin regimen with the addition of aspirin would also prove effective.
This study convincingly contradicts the hypothesis. The magnitude of the treatment difference was so great that the study was stopped before its planned completion. Although this does limit the study’s ability to analyze truly long-term effects, the graphs showing event rates appear to be roughly linear throughout the course of the study. This suggests that a longer study period would not have greatly influenced the outcome.
These results are disappointing to those who had hoped for an easier anticoagulation regimen that would retain effectiveness. The lower threshold for warfarin benefit is now more clearly defined, and the INR range of 2.0-3.0 should become the accepted standard for anticoagulation of patients with nonvalvular atrial fibrillation. (Dr. DiMarco is Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville.)
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content