Warfarin and Antiplatelet Therapy in Chronic Heart Failure

Primary Hypothesis: The hypothesis to be tested is whether aspirin, warfarin, and clopidogrel are equally effective in the treatment of patients with symptomatic CHF and reduced ejection fraction.

Secondary Hypothesis: If one therapy proves to be superior with regard to outcomes, what is the cost of this benefit? Can subsets of patients be identified who benefit more from a specific approach to antithrombotic therapy?

Intervention: The three treatment regimens are:

1. Open-label Warfarin titrated to an INR of 2.5-3.0;

2. Double blind aspirin 162 mg once daily;

3. Double blind clopidogrel 75 mg once daily.

Primary Outcomes: Any death (all causes), non-fatal stroke, non-fatal MI.

Study Abstract: Whether patients with chronic heart failure (CHF) should be anticoagulated is one of the oldest unresolved questions in cardiovascular therapeutics. Some authorities do not recommend anticoagulation for CHF patients in sinus rhythm, others recommend anticoagulation in patients with primary cardiomyopathy, and still others consider it more appropriate in patients with coronary artery disease (CAD). This absence of consensus reflects the lack of evidence in this area and different outlooks on the objectives of such therapy (e.g., prevention of arterial emboli or reduction in vascular events).

The original target sample size was 4,500 over a 3 year enrollment period with a 2 year follow-up. This sample size yielded 90% power to detect a relative difference of 20% between treatment groups. The sample size was later amended to 1,500 over a 30 month enrollment period with a 12 month follow-up. The reduced sample size yielded 85% to detect a between treatment difference of 30%. This change became effective in March 2002.

This clinical trial enrolled 1,587 patients in 142 medical centers; VA and non-VA centers in the U.S., and medical centers in the United Kingdom and Canada. Patients were randomly and equally allocated to the 3 treatments: warfarin (administered open-label), aspirin and clopidogrel (the latter two administered double-blind). The study was conducted over a 3.5 year period, with a 2.5 year enrollment phase.

Patients with NYHA class II, III, or IV and left ventricular ejection fractions less than or equal to 35% on an ACE inhibitor (unless not tolerated) and a diuretic were entered. The primary end point is the composite of death from any cause, non-fatal MI, and non-fatal stroke. All-cause mortality is the secondary end point.

The WATCH design paper has been published in the Journal of Cardiac Failure. Preliminary results were presented at the meeting of the American College of Cardiology in New Orleans on March 9, 2004. There were no significant differences between the treatment groups for the primary endpoints. The paper with final results is being prepared.