WATCH: Warfarin and Antiplatelet Therapy in Chronic Heart Failure Trial

Walter Alexander, MS


April 16, 2004

Editorial Collaboration

Medscape &

Presenter: Barry M. Massie, MD, Veterans Affairs Medical Center (San Francisco, California)

The results of the Warfarin and Antiplatelet Therapy in Chronic Heart Failure Trial, the largest trial conducted to date of antithrombotic treatments in heart failure, failed to show significant differences between aspirin, warfarin, and clopidogrel in the primary composite endpoint of all-cause mortality, nonfatal myocardial infarction (MI), and nonfatal stroke. It did, however, reveal an "intriguing" lower rate of heart failure hospitalizations for warfarin vs aspirin.

Study Design and Methods

WATCH was designed to have an enrollment of 4500 patients in Canada, the United Kingdom, and the United States. However, an enrollment rate of only 40% of the anticipated level forced protocol modifications and, finally, early discontinuation of the study at enrollment of only 1587 patients. The study objective was to determine the optimal antithrombotic agent for chronic heart failure (CHF) patients with regard to clinical outcomes, safety, and cost.

The primary hypothesis of WATCH was that anticoagulation with warfarin is superior to antiplatelet therapy with aspirin in preventing vascular events in CHF patients. Dr. Massie noted that aspirin may have an adverse effect in these patients, possibly due to interference with the action of angiotensin-converting enzyme (ACE) inhibitors. Clopidogrel was included as an active control because it has a different mechanism of action. Patients were randomized in an open-label fashion to warfarin (international normalized ratio [INR] 2.5-3.0), aspirin (double-blind,182 mg/day) or clopidogrel (double-blind, 75 mg/day). Maximum follow-up was 5 years, and minimum was 2 years (mean follow-up, 23 months). The study's composite primary endpoint included all-cause mortality, nonfatal MI, and nonfatal stroke.

Mean INR in the warfarin group was 2.6, with 31% of measurements in the 2.5-3.0 INR target range, and 70% in the 2.0-3.5 range. Mean age was 63 years, and 85% of the subjects were male. Nearly three quarters of the study population had ischemic cardiomyopathy. Nearly all patients (~98%) were receiving either ACE inhibitors or angiotensin receptor blockers (ARBs), and 99% received loop diuretics, 70% beta-blockers, and approximately 30% spironolactone. More than one third of patients (35%) had been hospitalized within the last 6 months. Fifty-six percent were in either New York Heart Association (NYHA) class III or IV heart failure. Mean ejection fraction (EF) was 24%.


Analysis showed no differences between aspirin and warfarin with regard to the primary endpoint (Table). Among the endpoint components, there was a strong trend favoring warfarin in the treatment of nonfatal stroke.

Table. WATCH: Aspirin vs Warfarin
Endpoint Aspirin (n = 523) Warfarin (n = 540) P
N % N %
Death, MI, stroke 107 20.5 107 19.8 .20
Death 94 18.0 92 17.0 .58
Nonfatal MI 14 2.7 22 4.1 .15
Nonfatal stroke 11 2.1 4 0.7 .06
MI, myocardial infarction

For the prospectively defined additional endpoint of heart failure hospitalizations, there was a significant 27% reduction for warfarin over aspirin (16.1% vs 22.2%, P = .01). The primary endpoint rate for clopidogrel was nonsignificantly higher at 21.8% (P = .68 vs aspirin), with no remarkable differences among the individual components of the composite endpoint.

Major bleeding episodes were more frequent in the warfarin group than in the aspirin or clopidogrel groups (30 vs 19 vs 13; P = .012 for warfarin vs clopidogrel). Minor bleeding episodes occurred significantly more often in the warfarin group than in either the aspirin or clopidogrel groups (P < .01).


Dr. Massie assessed the likely effect of the trial's discontinuation on outcomes and concluded that, although it precluded definitive answers, major differences for most primary endpoints were "unlikely." He added that, although caution should be exercised with regard to the finding that warfarin is associated with a lower rate of heart failure hospitalization, this outcome is in line with results of the Warfarin-Aspirin Study in Heart Failure (WASH) pilot study, observational data, and studies showing a similar trend found for clopidogrel. Furthermore, he said, "It is consistent with the observation that prostaglandin inhibition can have an adverse effect in heart failure patients." The results, he concluded, affirm the need for additional studies comparing aspirin and other antithrombotic therapies.

Dr. Massie added, "I don't think aspirin should be used routinely in a heart failure population."