Warfarin: an inconvenient truth.

See related article, pages 235–240. In 1989, the Copenhagen AFASAK study1 reported results from the first randomized, placebo-controlled trial evaluating adjusted-dose warfarin and aspirin in the prevention of stroke for nonvalvular atrial fibrillation (NVAF). Like the 5 randomized, placebo-controlled trials that came after it,2 the Copenhagen AFASAK study demonstrated a reduction in embolic events in patients taking warfarin. Today, the weight of evidence from 18 trials comparing warfarin with either placebo or antiplatelet drugs overwhelmingly favors the use of warfarin for stroke prevention in NVAF.2,3 In 20 years since AFASAK and 10 years since the first meta-analysis of trials testing the effect of warfarin on stroke risk,4 anticoagulation has remained underused and opportunities for preventing fatal and disabling stroke have been frequently missed.5–12 In this issue of Stroke, Gladstone and colleagues5 report a prospective practice audit of 920 patients presenting with ischemic stroke and a history of atrial fibrillation to 12 hospitals in Ontario, Canada. All patients had a high risk of stroke. Approximately 60% of the 597 patients with a first-ever stroke were not receiving warfarin at the time of admission. Of those receiving warfarin, approximately three fourths had a subtherapeutic international normalized ratio (INR) recorded on admission. A further 323 patients with a history of stroke or transient ischemic attack arguably had the most to gain from warfarin. Approximately 40% of this group was not on warfarin at the time of admission, whereas 70% of those using warfarin had subtherapeutic INR levels. These findings are disturbing because all patients included in the audit were judged to be “ideal candidates” for warfarin, living independently without known contraindications to anticoagulation. Gladstone et al5 highlight the severity of stroke in patients with NVAF. Eighty percent of the patients with a first-ever stroke either died or were disabled at discharge (modified Rankin score of 2 or more). Gladstone et al’s5 study paints a depressing picture that is replicated elsewhere.6–12 Taken together, existing research indicates that the suboptimal uptake of warfarin and subtherapeutic anticoagulation is the norm. The implication is clear. Wider uptake of warfarin would reduce death and disability in people with NVAF. Meanwhile, aging populations have an increasing prevalence of NVAF and a growing burden of preventable stroke. Gladstone et al5 recognize wider implementation of warfarin as a major challenge. Three main objections are often raised as barriers to the use of warfarin: that it may not be a safe treatment for elderly patients who are presumed to be at greater risk of a major hemorrhage, that the proven benefits of warfarin derived from trials of highly selected patients may not be realized in routine clinical practice, and that therapeutic control is difficult. The Birmingham Atrial Fibrillation Treatment of the Aged Study (BAFTA)13 has now resolved the first 2 objections. BAFTA is one of the largest studies of stroke risk reduction in atrial fibrillation in the elderly. BAFTA randomized patients older than age 75 years (average age, 81.5 years). Patients received adjusted-dose warfarin or 75 mg per day of aspirin. Significantly, patients were recruited by primary care physicians and received usual anticoagulation management. Results confirmed the superiority of warfarin over aspirin and demonstrated equivalent risks of major bleeding, even in those older than 85 years of age. A meta-analysis of 8 previous comparisons between warfarin and aspirin demonstrated a relative risk reduction of 38% (95% CI, 18% to 52%) for all strokes and subdural hematoma in favor of warfarin.3 In comparison, BAFTA demonstrated a relative risk reduction of 47% (95% CI, 19% to 66%).3 Previous studies evaluating warfarin had restrictive selection criteria.2,4 In BAFTA, only 6.1% of patients did not meet inclusion criteria. According to the uncertainty principle, physicians excluded 54% of the screened sample of patients because one treatment was preferred over the other. Importantly, 79% of these were excluded because warfarin was selected as the drug of choice by the treating physician. BAFTA provides compelling evidence of the effectiveness of warfarin in elderly people. The study demonstrates the safety of warfarin in an elderly cohort and that warfarin’s benefits are generalizable to patients managed in routine clinical practice. Achieving therapeutic control with warfarin is considered difficult. As a vitamin K antagonist, warfarin has diet and some drug sensitivity.14 INR levels can be maintained within a therapeutic range2,3 approximately 65% to 70% of the time13,15–17 and suband supratherapeutic INR levels still confer some protection from stroke.18 Although better control The opinions in this editorial are not necessarily those of the editors or of the American Heart Association. From School of Public Health and Community Medicine (M.G., N.Z.), The University of New South Wales, Australia; the Centre for Research, Evidence Management and Surveillance (M.G.), Division of Population Health, Sydney South West Area Health Service, NSW, Australia; The Northern Beaches Stroke Service (J.W.), Sydney, NSW, Australia; Sydney South West Clinical School, The University of New South Wales and Liverpool Health Service, Liverpool, NSW, Australia. Correspondence to John Worthington, Senior Staff Specialist, Department of Neurophysiology, Liverpool Health Service, Locked Bag 7017, Liverpool NSW 1871 Australia. E-mail John.Worthington@sswahs. nsw.gov.au (Stroke. 2009;40:5-7.) © 2008 American Heart Association, Inc.