OBJECTIVE: To review the risk and pathogenesis of stroke associated with nonvalvular atrial fibrillation (AF) and the efficacies and risks of stroke prevention strategies. BACKGROUND: About 16% of ischemic strokes are associated with AF; AF is an independent risk factor for stroke. METHODS: Review of the literature, focusing on 13 randomized trials of antithrombotic therapy. RESULTS: The overall risk of stroke in AF patients averages about 5%/y, but with wide variation depending on the presence of coexistent thromboembolic risk factors. AF patients with low (about 1% per year), moderate (about 3% per year), and high (about 6% per year) stroke risks have been identified, but the generalizability of risk stratification schemes to clinical practice has not been fully assessed. AF patients with prior stroke or transient ischemic attack, even if remote, are at highest risk (about 12% per year). Adjusted-dose warfarin (target International Normalized Ratio [INR] 2-3) is highly efficacious for preventing stroke in AF patients (about 70% risk reduction) and is safe for selected patients, if carefully monitored. Aspirin has a modest effect on reducing stroke (about 20% risk reduction). The numbers of AF patients that would need to be treated with warfarin instead of aspirin for 1 year to prevent one ischemic stroke are about 200, 70, and 20 for those with low, moderate and high risk, respectively. CONCLUSIONS: Many patients with nonvalvular AF have substantial rates of ischemic stroke. Stratification of stroke risk identifies AF patients who benefit most and least from lifelong anticoagulation. Warfarin is recommended for high-risk AF patients who can safely receive it. Aspirin may be indicated for those with a low stroke risk and for those who cannot receive warfarin. For AF patients considered to have a moderate risk of stroke, individual bleeding risk during anticoagulation and patient preference should particularly influence the choice of antithrombotic prophylaxis.
OBJECTIVE: To review the risk and pathogenesis of stroke associated with nonvalvular atrial fibrillation (AF) and the efficacies and risks of stroke prevention strategies. BACKGROUND: About 16% of ischemic strokes are associated with AF; AF is an independent risk factor for stroke. METHODS: Review of the literature, focusing on 13 randomized trials of antithrombotic therapy. RESULTS: The overall risk of stroke in AFpatients averages about 5%/y, but with wide variation depending on the presence of coexistent thromboembolic risk factors. AFpatients with low (about 1% per year), moderate (about 3% per year), and high (about 6% per year) stroke risks have been identified, but the generalizability of risk stratification schemes to clinical practice has not been fully assessed. AFpatients with prior stroke or transient ischemic attack, even if remote, are at highest risk (about 12% per year). Adjusted-dose warfarin (target International Normalized Ratio [INR] 2-3) is highly efficacious for preventing stroke in AFpatients (about 70% risk reduction) and is safe for selected patients, if carefully monitored. Aspirin has a modest effect on reducing stroke (about 20% risk reduction). The numbers of AFpatients that would need to be treated with warfarin instead of aspirin for 1 year to prevent one ischemic stroke are about 200, 70, and 20 for those with low, moderate and high risk, respectively. CONCLUSIONS: Many patients with nonvalvular AF have substantial rates of ischemic stroke. Stratification of stroke risk identifies AFpatients who benefit most and least from lifelong anticoagulation. Warfarin is recommended for high-risk AFpatients who can safely receive it. Aspirin may be indicated for those with a low stroke risk and for those who cannot receive warfarin. For AFpatients considered to have a moderate risk of stroke, individual bleeding risk during anticoagulation and patient preference should particularly influence the choice of antithrombotic prophylaxis.