• Anticoagulation therapy for patients with atrial fibrillation [Letter]

Anticoagulation reduces stroke for all patients with atrial fibrillation,² but the magnitude of benefit (that is, the absolute risk reduction) is small for many patients with atrial fibrillation who have relatively low inherent risks of stroke. Many younger patients with atrial fibrillation have low (less than 2% per year) or moderate (3–5% per year) rates of stroke, and the number-needed-to-treat with warfarin for 1 year to prevent 1 stroke is between 30 and 100 for such patients; the number-needed-to-treat figures are doubled for prevention of strokes leaving even minimal residual disability.³ Patients over age 75 with atrial fibrillation are more likely to be at high risk⁴ but less likely to receive anticoagulation.^5,6 Ironically, younger patients with fewer comorbidities are more attractive candidates for anticoagulation, yet, on average, accrue less benefit when the absolute risk reduction is considered.

Those decrying underuse of warfarin often imply that anticoagulation therapy is underused because physicians lack the knowledge or commitment to prevent stroke, yet it is often the patients themselves who choose not to receive anticoagulation.⁷ Patient-perceived thresholds of benefit for choosing anticoagulation vary widely; often those with stroke risks in the moderate range elect not to receive anticoagulation after the benefits and risks are explained to them.⁷ Further study of the preferences of informed patients and of the influence of different educational methods is sorely needed.⁸

I contend that the 50% frequency of coagulation use among patients with atrial fibrillation reported in recent studies does not represent gross underuse for many populations of patients with atrial fibrillation⁹ (I acknowledge that patients at high risk may make up a larger proportion of the patients in clinical practice than of the participants in clinical trials [Evidence]¹⁰). Rather, anticoagulation is too often given to those who benefit least rather than most. Additional studies of the the reliability of risk stratification schemes when applied in clinical practice¹¹ and of patient perceptions of minimal thresholds of benefit are needed to foster the optimal use of this highly efficacious therapy to prevent stroke.