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


Warfarin and aspirin did not differ for preventing embolic complications of atrial fibrillation

ACP J Club. 1994 Sept-Oct;121:39. doi:10.7326/ACPJC-1994-121-2-039

Source Citation

Stroke Prevention in Atrial Fibrillation Investigators. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study. Lancet. 1994 Mar 19;343:687-91.



To compare warfarin with aspirin for preventing ischemic stroke and systemic embolism in adults with atrial fibrillation (AF).


2 randomized controlled trials.


16 clinical centers.


715 adults (76% men) aged ≤ 75 years and 385 adults (59% men) aged > 75 years with nonrheumatic AF in the preceding year. Exclusion criteria were need for or contraindications to aspirin or warfarin and age < 60 years without overt cardiovascular disease other than AF. Follow-up was > 99%.


188 older and 357 younger patients were randomly allocated to receive enteric-coated aspirin, 325 mg daily. 197 older and 358 younger adults were randomly allocated to warfarin (prothrombin time ratios were from 1.3 to 1.8 for the first half of the study and the international normalized ratio (INR) was between 2.0 and 4.5 subsequently).

Main outcome measures

Ischemic stroke and systemic embolism. Secondary end points were all strokes and primary end point plus vascular death.

Main results

The younger patients were followed for a mean of 3.1 years and older patients for 2 years. For younger patients, the groups did not differ for annual rates for primary end points (1.9% for aspirin vs 1.3% for warfarin, 95% CI for the 0.6% per year difference -0.4% to 1.7%), all strokes with residual deficit (1.8% vs 1.4%), myocardial infarction (1.3% vs 0.9%), total mortality (3.7% vs 3.2%), or primary events plus vascular death (3.7% vs 2.9%). For older patients, the groups did not differ for annual rates for primary end points (4.8% for aspirin vs 3.6% for warfarin, CI for the 1.2% per year difference -1.7% to 4.1%), all strokes with residual deficit (4.3% vs 4.6%), myocardial infarction (1.3% for both), total mortality (6.1% vs 6.4%), or primary events plus vascular death (7.2% vs 7.4%). Patients with other risk factors for thromboembolism assigned to aspirin had a higher rate of primary end points compared with patients without risk factors (P < 0.001). Older, but not younger, patients had a lower rate of major hemorrhage on aspirin compared with warfarin (1.6% vs 4.2%, P = 0.04).


No conclusive differences existed between warfarin and aspirin for preventing thromboembolism and ischemic stroke in adults with atrial fibrillation.

Sources of funding: National Institute of Neurological Disorders and Stroke; Du Pont Pharmaceuticals (Coumadin [warfarin]); Smith Consumer Brands (Ecotrin [aspirin]).

For article reprint: Ms. R. McBride, Statistics and Epidemiology Research Corporation, 1107 NE 45th Street, Suite 520, Seattle, WA 98105, USA. FAX 206-547-2829.


Since the Stroke Prevention in Atrial Fibrillation II study (SPAF-II), more evidence on stroke prevention in non-valvular AF has accumulated. In 1994, the gold standard for the management of patients with AF remained anticoagulation with adjusted-dose warfarin and a target INR of 2.0 to 3.0. However, the high rate of bleeding complications stimulated the search for effective interventions with fewer adverse events.

Although less effective for stroke prevention than warfarin, aspirin caused less severe adverse events, and the more recent studies explored the combination of fixed low dose of warfarin and aspirin. This approach failed: both (SPAF III [1]—investigating this scheme in high risk patients—and the Second Copenhagen Atrial Fibrillation, Aspirin and Anticoagulation Study [AFASAK-2] [2]—recruiting patients indiscriminate of their baseline risk) were terminated early because of substantially fewer thromboembolic events in the conventional adjusted-dose warfarin groups (SPAF-III 7.9% vs 1.9% per year, AFASAK-2 7.2% vs 2.8% per year). Patients with AF vary in their individual risk for stroke. Although the risk for bleeding is acceptable in high-risk patients (e.g., > 8% risk for stroke per year), it might outweigh the benefit of anticoagulation in patients with moderate to low risk, in whom the absolute risk reduction declines considerably. Validation of a retrospectively derived low-risk stratification scheme (absence of congestive heart failure or left ventricular fractional shortening < 25%, previous thromboembolism, systolic blood pressure > 160 mm Hg, women > 75 y) successfully identified patients with low rates of systemic emboli or ischemic stroke (2.2% per year), which was even further reduced to 1.1% per year—the risk in the general population—in the subgroup of normotensive patients [4].

However, because all patients were taking aspirin, the influence of aspirin on the low event rate remains unclear. This validated risk stratification enables physician and patient to better quantify individual risks and benefits and integrate patients' preferences in the decision, particularly in those with only marginal benefit of anticoagulation.

Regina Kunz, MD, MSc
CharitBerlin, Germany


1. Stroke Prevention and Atrial Fibrillation Investigators. Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial. Lancet. 1996;348:633-8.

2. Gulløv AL, Koefoed BC, Petersen P, et al. Fixed minidose warfarin and aspirin alone and in combination vs adjusted-dose warfarin for stroke prevention in atrial fibrillation. Arch Intern Med. 1998;158:1513-21.

3. The SPAF III Writing Committee for the Stroke Prevention in Atrial Fibrillation Investigators. Patients with nonvalvular atrial fibrillation at low risk of stroke during treatment with aspirin: Stroke Prevention in Atrial Fibrillation III Study. JAMA. 1998;279:1273-7.