BOSTON, MASSACHUSETTS. Clinical trials carried out in 1994 concluded that the use of warfarin in atrial fibrillation (AF) patients was relatively safe with an annual rate of major hemorrhage of 1.3%. Major hemorrhage is defined as a fatal bleeding incident, a bleeding incident requiring hospitalization with transfusion of 2 or more units of packed red blood cells, or a bleeding incident involving a critical site (intracranial, intraspinal, pericardial, intraocular, etc). The average annual reduction in ischemic stroke rate in the five 1994 trials was 1.8% for patients over the age of 75 years with no risk factors for stroke, and 6.9% for those with one or more risk factors. Thus, it was concluded that treating older patients with warfarin had a favourable benefit/risk ratio.

Elaine Hylek and colleagues at the Boston University School of Medicine now question this conclusion. Their clinical trial involved 472 AF patients with an average age of 77 years (32% were 80 years or older). Forty-seven percent of the patients were women and 91% had one or more risk factors for ischemic stroke (75% had hypertension and 35% had coronary artery disease). After being admitted with a first AF episode (59%), a recurrent episode (35%), or permanent AF (6%) all study participants were prescribed warfarin with an INR target of 2.0 – 3.0. Management of warfarin dosage was carried out by the hospital's own anti-coagulation clinic. More than 10,000 INR measurements were made during the 1-year follow-up period. The time spent within the prescribed INR range (2.0 – 3.0) was only 58% with 29% being spent below 2.0 and 13% above 3.0.

The overall incidence of major hemorrhage was 7.2% and that of intracranial hemorrhage (hemorrhagic stroke) was 2.5%. A third of the hemorrhagic strokes were fatal and 89% of them occurred in patients 75 years or older. The incidence of major hemorrhage was particularly high (13.1%) among patients 80 years or older. Age and an INR greater than 4 were strong risk factors and 58% of the major hemorrhages occurred within the first 90 days after initiation of warfarin therapy. Concomitant use of aspirin was also a significant risk factor for major bleeding and there was no indication that taking 81 mg/day was any safer than taking the standard 325 mg/day.

During the study 26% of participants aged 80 years or older were taken off warfarin – 81% because of safety concerns and 19% because they regained normal sinus rhythm. The Boston researchers conclude that the risk of major bleeding among older AF patients on warfarin has been significantly underestimated in previous trials. They also point out that the rate of bleeding observed in their closely controlled clinical trial would likely be significantly lower than that experienced in the "real world".

In an accompanying editorial Dr. George Wyse of the Health Sciences Center in Calgary, Canada states, "there is reason to be sceptical about net benefit when warfarin is used in some elderly patients with AF." Dr. Wyse also points out that warfarin therapy would appear to be over-utilized in patients with low to moderate risk of ischemic stroke. A recent European study found that 50% of AF patients with no risk factors for stroke were being treated with warfarin or similar anticoagulants.

Hylek, EM, et al. Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation. Circulation, Vol. 115, May 29, 2007, pp. 2689-96
Wyse, DG. Bleeding while starting anticoagulation for thromboembolism prophylaxis in elderly patients with atrial fibrillation. Circulation, Vol. 115, May 29, 2007, pp. 2684-86

Editor's comment: This study adds to the growing evidence that warfarin therapy is far from ideal for AF patients. It would appear to be over-prescribed for patients who don't need it and of no overall benefit for older patients with one or more risk factors for ischemic stroke.