The Sin of Omission: A Systematic Review of Antithrombotic Therapy to Prevent Stroke in Atrial Fibrillation

RATIONALE: To characterize the comparative efficacy and safety of antithrombotic therapy for the prevention of stroke in patients with atrial fibrillation. BACKGROUND: Atrial fibrillation is a major risk factor for stroke and is increasingly prevalent with age, occurring in over 2 million people in the United States, 1 7% to 14% of whom are age 65 and older. 2–5 The most common causes of atrial fibrillation are hypertensive heart disease and ischemic heart disease. DATA SOURCES: Randomized trials testing long‐term (>3 months) use of antithrombotic agents in patients with atrial fibrillation were sought by a search of OVID/MEDLINE databases (from 1966 to 1999, not restricted by language) and by inquiries to the Cochrane Collaboration Stroke Review Group and Antithrombotic Trialists Collaboration. STUDY SELECTION CRITERIA: Double‐blind and non‐blinded trials were included, and sensitivity analysis was used to compare pooled results. Trials reporting results for subgroups of participants with atrial fibrillation, among other participants without atrial fibrillation, were included, except for one trial in which the participants with atrial fibrillation were not reported separately. Studies of atrial fibrillation associated with prosthetic cardiac valves or mitral stenosis were not considered. DATA EXTRACTION: Data were extracted from 16 randomized trials published between 1989 and 1999, which included 9874 participants with nonvalvular atrial fibrillation. Two reviewers independently extracted data from published sources on the number of patients treated, total follow‐up exposure, and the occurrence of five outcomes by intention‐to‐treat analysis: all stroke (hemorrhagic and ischemic), ischemic stroke, intracranial hemorrhage, all‐cause mortality, and major extra cranial bleeding. The criteria used for each was that used in each trial. All stroke was the primary outcome as the evaluation of each event was not consistent. MAIN RESULTS: Warfarin was the antithrombotic agent used exclusively in 10 trials; 2 others used other coumadin derivatives, for a total of 2239 participants. In 6 trials, the efficacy of adjusted‐dose warfarin or coumadin derivatives in preventing stroke was compared with placebo. Mean age was 69 years at entry; 20% were over 75. Twenty‐nine percent were women, 45% had hypertension, and 20% had prior stroke or TIA. Although the target range for the International Normalized Ratio (INR) varied, the mean range achieved was 2.0 to 2.6 in the five primary prevention trials and 2.9 in one secondary prevention trial. Mean duration of follow‐up ranged from 1.2 years to 2.3 years; 4 trials were stopped early due to treatment efficacy at an interim analysis. The placebo group had an average stroke rate of 4.6% for primary prevention and 12.3% for secondary prevention. CONCLUSIONS: This meta‐analysis demonstrates the efficacy of antithrombotic agents in the prevention of all stroke in patients with atrial fibrillation, which was not offset by an increased risk for major hemorrhage. Adjusted‐dose warfarin reduced the risk of stroke by about 60% compared with a 20% reduction achieved with aspirin therapy (about 40% by warfarin versus aspirin). This data was compiled from six trials, four of which were concluded early, which lead to an overestimate of treatment effect. The largest stroke reductions were found in two trials with estimated target INRs between 1.4 and 2.8. Aspirin had less efficacy than warfarin, and appeared to be most efficacious in the prevention of nondisabling stroke, which may not be cardioembolic in nature.