Warfarin and Aspirin in Patients With Heart Failure and Sinus Rhythm

Study Questions:

Is warfarin superior to aspirin alone for patients with heart failure (HF) who are in sinus rhythm?


WARCEF (Warfarin versus Aspirin in Reduced Cardiac Ejection Fraction) was a double-blind, multicenter, randomized trial of warfarin (goal international normalized ratio [INR], 2-3.5) versus aspirin 325 mg daily in patients (n = 2,305) with a left ventricular ejection fraction (LVEF) ≤35%. Patients could not have had other reasons for anticoagulation. However, they did not exclude patients with prior history of transient ischemic attack/cerebrovascular accident (TIA/CVA). The primary outcome of interest was time to first event, which was a composite of ischemic stroke, intracerebral hemorrhage, or death. Hazard ratios (HRs) [95% confidence intervals] are provided.


The mean LVEF of the cohort was 24.7 ± 7.5%, 13% had had a prior TIA/CVA, and the mean age was 61 years. There were 1,163 patients in the aspirin arm and 1,142 in the warfarin arm. Groups were similar in age, sex, comorbidities, New York Heart Association class, and blood pressure. In the warfarin group, INRs were at goal 63% of the total treatment time, <2.0 for 27%, and >3.5 for 10% of treatment time. There were 622 events, of which 85% were deaths, 13.5% were ischemic strokes, and 1.1% were intracerebral bleeds. The rate of primary outcome in the warfarin group was 7.47 per 100 patient-years, which was similar to that of the aspirin group at 7.93 per 100 patient-years (HR, 0.93 [0.70-1.10]). Warfarin use consistently reduced the risk of ischemic stroke (HR, 0.52 [0.33-0.82]), but the risk of major hemorrhage was nearly twofold higher with warfarin use (HR, 2.1 [1.4-3.1]) at a rate of 1.78 vs. 0.87 events per 100 patient-years with warfarin versus aspirin, respectively.


The authors concluded that in patients with systolic HF, warfarin reduces ischemic strokes at a cost of increased major hemorrhage.


Cardioembolic strokes tend to be very disabling. In this analysis, the use of warfarin cut the risk of ischemic stroke nearly in half. As with most medications, there is a risk:benefit ratio that needs to be considered prior to prescription. There was no significant difference between aspirin and warfarin in the composite endpoint of death, ischemic stroke, intracerebral hemorrhage, myocardial infarction, or HF hospitalization. However, it is important to look at each endpoint component individually. There was no difference in mortality, which would be expected unless warfarin use led to fatal bleeding events. Oddly, warfarin patients had a trend toward more HF hospitalizations (HR, 1.21 [0.998-1.47]). There is no clear clinical explanation for why this would be the case, but it impacted the overall composite hazard ratio. Likewise, when major bleeding events were considered, warfarin increased risks of bleeding twofold. The types of bleeding encountered were not tabulated, but we are told that rates of intracerebral and intracranial hemorrhage were not higher with warfarin. Rates of major bleeding were in line with other trials of warfarin use (1.78 events per 100 patient-years). Thus, the benefit of warfarin remains in question. It reduced ischemic stroke—that is clear. It increased major bleeding rates, but bleeding rates overall were much lower (approximately one-fourth) than that of stroke rates.

Clinical Topics: Anticoagulation Management, Heart Failure and Cardiomyopathies, Acute Heart Failure

Keywords: Stroke, Warfarin, Heart Failure, Cerebral Hemorrhage

< Back to Listings