Current issues of ACP Journal Club are published in Annals of Internal Medicine


Anticoagulation prevented recurrent vascular events

ACP J Club. 1994 May-June;120:74. doi:10.7326/ACPJC-1994-120-3-074

Source Citation

European Atrial Fibrillation Trial Study Group. Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke. Lancet. 1993 Nov 20;342:1255-62.



To determine the effectiveness of anticoagulation or aspirin in preventing vascular events in patients with recent transient ischemic attack (TIA) or minor ischemic stroke and non-rheumatic atrial fibrillation.


3-year randomized, multicenter, placebo-controlled trial.


108 hospitals in Europe and 1 in Israel.


1007 patients (mean age 73 y, 56% men) with nonrheumatic atrial fibrillation who had had a TIA or minor ischemic stroke in the previous 3 months. Exclusion criteria were secondary atrial fibrillation (e.g., hyperthyroidism); contraindication to or absolute indication for aspirin; treatment with nonsteroidal anti-inflammatory, antiplatelet-aggregating, or oral anticoagulant drugs; other sources of cardiac emboli; coagulation disorder; or scheduling of carotid endarterectomy or coronary surgery in the next 3 months. Follow-up was 99%.


Patients eligible for anticoagulation (group 1) were assigned to receive anticoagulants (open label n = 225), double-blind treatment with aspirin (n = 230), or placebo (n = 214). Patients ineligible for anticoagulation (group 2) because of unwillingness of patients or their physicians, alcoholism, hemorrhagic retinopathy, previous intracranial hemorrhage, or expected poor compliance were assigned to receive aspirin (n = 174) or placebo (n = 164). Compliance was assessed by patient interviews, pill counts, and international normalized ratios (INRs).

Main outcome measure

Combined rate of death from vascular disease; and nonfatal stroke, myocardial infarction, or systemic embolism.

Main results

In an intention-to-treat analysis, the rate of primary outcome events was reduced in patients in group 1 who received anticoagulants compared with patients who received placebo (8%/y vs 17%/y, hazard ratio [HR] 0.53, 95% CI 0.36 to 0.79, P = 0.001) (Table) or aspirin (HR 0.60, CI 0.41 to 0.87, P = 0.008). The intended aspirin comparison combined aspirin and placebo from groups 1 and 2. The primary outcome event rates between patients who received aspirin or placebo did not differ (15%/y and 19%/y, respectively; HR 0.83; CI 0.65 to 1.05) (Table). Major bleeding rates were 2.8%/y, 0.9%/y, and 0.7%/y in patients assigned to receive anticoagulation, aspirin, and placebo, respectively.


Anticoagulation was more effective than aspirin or placebo in reducing the rate of vascular death, nonfatal stroke, myocardial infarction, and system embolism in patients with nonrheumatic atrial fibrillation who had a recent transient ischemic attack or minor ischemic stroke.

Sources of funding: Netherlands Heart Foundation; Bayer Germany; UK Stroke Association; University Hospitals of Utrecht and Rotterdam.

For article reprint: Dr. P.J. Koudstaal, Department of Neurology, University Hospital Rotterdam, Dr Molewaterplein 40, 3015 GD Rotterdam, the Netherlands. FAX 31-10-436-7501

Table. Anticoagulants vs placebo and aspirin vs placebo in recent transient ischemic attack*

Outcome at mean follow-up of 2 years Anticoagulants Placebo RRR (95% CI) NNT (CI)
Primary outcome event† 19% 31% 39% (15 to 56) 8 (5 to 24)
Primary outcome event† 32% 36% 11% (-9 to 26) Not significant

*Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article.
†Primary outcome events were death from vascular disease; and nonfatal stroke, myocardial infarction, and systemic embolism.


The European Atrial Fibrillation Trial (EAFT) results represent the first instance of trial-proved efficacy of warfarin in any setting of secondary stroke prevention. The rate of the composite end point for patients in the placebo group is remarkably high at 17%/y. Most events were stroke recurrences at 12%/y. For perspective, the placebo rate of initial stroke in the primary prevention trials in atrial fibrillation was about 5%/y (1). Considering the results of the primary prevention trials, it was not surprising that warfarin was effective in secondary prevention, although, interestingly, overall mortality was not improved. The greater issue was whether it could be used safely because these patients have risks for general and intracranial bleeding, including atrial fibrillation itself, previous stroke, and advanced age (2). Moreover, the EAFT anticoagulation target (INR 2.0 to 4.0) exceeded that of 3 of the 5 primary prevention trials as well as current American College of Chest Physicians recommendations (International Normalized Ratio [INR] 2.0 to 3.0) (1). At 2.8%/y (about twice that of the primary prevention trials), the EAFT rate of major hemorrhages was acceptable because of the substantial benefit of warfarin. It is important, although surprising, that no intracranial hemorrhages were documented.

Aspirin seems to have little role in secondary prevention, except as a default when warfarin is contraindicated. Subgroups with low intrinsic risk or special aspirin responsiveness were not reported. The timing of the initiation of anticoagulation is not unanswered, balancing concern about inducing hemorrhage into fresh infarction. Temporal analysis of end points shows that risk continued as long as patients were studied in the EAFT, indicating that short-term warfarin therapy will not suffice.

David C. Anderson, MD
Hennepin County Medical CenterMinneapolis, Minnesota, USA


1. Laupacis A, Albers G, Dunn M, Feinberg W. Antithrombotic therapy in atrial fibrillation. Chest. 1992;102(4 Suppl.):426S-33S.

2. Landefeld CS, Beyth RJ. Anticoagulant-related bleeding: clinical epidemiology, prediction, and prevention. Am J Med. 1993;95:315-28.