COPENHAGEN, DENMARK. Atrial fibrillation (AF) with underlying heart disease or other comorbidities is associated with an increased risk of ischemic stroke. Thus, treatment with aspirin or vitamin K antagonists (warfarin) is often recommended for AF patients. The justification for prescribing aspirin or warfarin is based on the results of closely controlled clinical trials, which may or may not reflect the "real world". A study reported in 2003 found that warfarin therapy had no net benefit in AF patients with no risk factors for ischemic stroke, but was of significant benefit to those who had previously suffered an ischemic stroke[1].

A team of Danish and British researchers now reported on a major study aimed at determining the benefits of aspirin and warfarin therapy in a "real world" setting. The study included 132,372 AF patients discharged from hospital with a diagnosis of non-valvular AF (AF without a previous diagnosis of mitral or aortic valve disease, and absence of mitral or aortic valve surgery). As Danish citizens, all study participants had a unique person registration number which allowed precise linking of databases regarding hospital admissions, drug prescriptions, vital statistics, comorbid conditions, and causes of death. Of the 132,372 AF patients discharged between 1997 and 2008, 44.5% (58,883 patients) were not prescribed aspirin or anticoagulants (no treatment), 28.3% (37,425 patients) were prescribed an anticoagulant (vitamin K antagonist – most likely warfarin), 18.9% (24,984 patients) were prescribed aspirin, and the remaining 8.4% (11,080 patients) were prescribed both aspirin and anticoagulant at discharge.

Access to the comprehensive databases made it possible to construct risk scores for ischemic stroke and bleeding for each of the 132,372 patients. The following risk scores were used:

• CHADS2 – This scoring system assigns 1 point each to the presence of congestive heart failure, hypertension, diabetes, age of 75 years or older, and 2 points for a history of stroke or transient ischemic attack (TIA).
• CHA2DS2-VASc – This score assigns 1 point each to the presence of congestive heart failure, hypertension, diabetes, vascular disease, age 65 to 74 years, female gender, and 2 points for a history of thromboembolism and age 75 years or older.
• HAS-BLED – This scoring system assigns 1 point each for the presence of hypertension, abnormal liver/kidney function, history of thromboembolism, history of bleeding, alcohol or drug abuse, and age above 65 years. One point is also assigned to warfarin-treated patients whose INR values fluctuate excessively (not used in this study).

Primary study outcomes were hospitalizations or deaths from thromboembolism (ischemic stroke, TIA and peripheral artery embolism) or bleeding (gastrointestinal bleeding, intracranial bleeding including hemorrhagic stroke, and bleeding from the urinary tract). The researchers also calculated a net clinical benefit defined as:

Net clinical benefit = (ischemic stroke rate with no treatment – ischemic stroke rate with treatment) – 1.5 x (intracranial hemorrhage rate with treatment – intracranial hemorrhage with no treatment)

NOTE: The 1.5 multiplication factor for hemorrhagic stroke reflects the fact that hemorrhagic strokes are usually far more serious than are ischemic strokes.

During the first year following discharge from hospital, 5298 thromboembolic events were recorded among patients who had remained on their prescribed treatment for the entire year. Distribution of these events (%/year) according to CHADS2 and CHA2DS2-VASc scores was as follows:

The incidence of bleeding events during the first year (%/year) for patients remaining on treatment was:

The researchers also provided data for a 12-year follow-up period; however, the value of this data is questionable as changes in treatment were not allowed for. Distribution of events over the 12-year period (%/year) according to CHADS2 score was as follows:

The incidence of bleeding during the 12-year follow-up was as follows:

Net clinical benefit over the 12-year follow-up for a HAS-BLED score of 2 or less was:

It is clear that aspirin therapy is not beneficial and that no treatment at all is the best option for AF patients with a CHADS2 score of 0. The researchers conclude that warfarin, but not aspirin is effective in reducing the risk of thromboembolism and that the net clinical benefit of warfarin is clearly positive in AF patients with an increased risk of stroke/thromboembolism (CHADS2 score of 1 or higher and CHA2DS2-VASc score of 2 or higher).
[1] Go, AS, et al. Anticoagulation therapy for stroke prevention in atrial fibrillation. JAMA, Vol. 290, November 26, 2003, pp. 2685-92
Olesen, JB, et al. Risks of thromboembolism and bleeding with thromboprophylaxis in patients with atrial fibrillation: A net clinical benefit analysis using a 'real world' nationwide cohort study. Thrombosis and Haemostasis, Vol. 106, No. 4, October 2011, pp. 739-49

Editor's comment: This study confirms my long-held beliefs that aspirin is of no benefit in stroke prevention for AF patients Aspirin: Friend or Foe? and that no antiplatelet or anticoagulant treatment is the best option for lone afibbers with a CHADS2 score of 0 or a CHA2DS2-VASc score of less than 2. I would also suggest that the use of natural stroke prevention agents such as nattokinase, vitamin C, garlic and resveratrol would provide adequate and safe stroke protection for lone afibbers with a CHADS2 score of 1 (one risk factor for stroke).