Significance of Thromboembolism in Heart Failure

The incidence of thromboembolism in heart failure is increased. A number of reasons, including stasis of blood due to dilated cardiac chambers, low cardiac output and flow, increased incidence of atrial fibrillation, an activated coagulation system and endothelial dysfunction may all contribute to stroke, myocardial infarction, peripheral and pulmonary embolism in heart failure. Retrospective analyses of data bases from large-scale heart failure trials suggest an annual incidence of 1.5–3.5% 1. Because stroke, thromboembolism and myocardial infarction were generally regarded as points of secondary importance, when compared to hospital readmission, underreporting and misclassification of events may have occurred. Thromboembolism in heart failure may be a much more significant problem than currently recognized. A retrospective analysis of the SOLVD trial supports this notion by reporting an association of warfarin use with improved survival and reduced morbidity in patients with left ventricular dysfunction 2. However, the recently presented results from the Warfarin/Aspirin Study of Heart Failure (WASH) trial a prospective, multicenter pilot study comparing the outcome in 279 patients receiving no antithrombotic therapy, aspirin (300 mg/day), or warfarin (target INR of 2.5) showed no difference in combined endpoints of all-cause mortality, non-fatal MI, and non-fatal stroke between no antithrombotic therapy and use of warfarin. Aspirin even seemed to have an adverse effect on morbid events in this pilot study. The decision whether to employ anticoagulant therapy in a patient with heart failure rests on a tradeoff between the efficacy of anticoagulants in preventing thromboembolism and the risk of precipitating a hemorrhagic event. Results of controlled large-scale clinical trials assessing the risks of long term anticoagulation in this patient population have not been published yet. The Third American College of Chest Physicians Consensus Conference on Antithrombotic Therapy has estimated the general annual risk of bleeding in anticoagulation to be 2% for major bleeds and 0.8% for fatal bleeding 3. The risk of bleeding has been discussed to be increased in patients with heart failure due to hepatic congestion, altered metabolism of anticoagulant agents and possible drug interactions. Therefore, we need to identify those patient groups with heart failure that benefit most from anticoagulant therapy. Also, we need to know the thromboembolic risk factors that have the greatest impact on these patients. There is good evidence to support the use of anticoagulant therapy in patients with atrial fibrillation. Five randomized controlled trials evaluating the efficacy of warfarin in the primary prevention of stroke showed a benefit of anticoagulant therapy. An intention-to-treat analysis of the pooled data revealed an annual stroke rate of 4.5% for the control patients and 1.4% for the warfarin-treated patients (risk reduction 68%) 4. Although not significant in the multivariate analysis, the presence of heart failure further increased the risk of stroke. It is not clear whether the benefit of anticoagulation exceeds the risk of bleeding for patients in sinus rhythm. Left ventricular ejection fraction, chronic left ventricular aneurysm, left ventricular thrombi, exercise capacity and female gender have been reported to be risk factors for thromboembolism. None of these risk factors have been evaluated prospectively in patients in sinus rhythm. The incidence of arterial embolism or stroke in patients with heart failure may increase somewhat as the ejection fraction declines 5,6, but this analysis is confounded by the higher prevalence of atrial fibrillation in patients with more severe ventricular dysfunction. One of the major limitations of the retrospective analyses of large-scale heart failure trials, reporting an increased incidence of thromboembolism and stroke with a decrease in left ventricular ejection fraction, is the lack of follow up data on the presence or absence of atrial fibrillation. In some patients new onset atrial fibrillation, rather than a decreased ejection fraction, may have contributed to thromboembolism. As long as prospective data showing a clear benefit of anticoagulant therapy in patients with heart failure and sinus rhythm are missing, the evidence from available studies does not yet seem to be strong enough to recommend anticoagulant therapy for all patients with heart failure with low ejection fraction. It may prove to be more important to follow patients and to initiate anticoagulation when atrial fibrillation develops. At present the only clear-cut indications for long term anticoagulation in most patients with heart failure are atrial fibrillation and a previous thromboembolic event. A left ventricular thrombus may be an indication of temporary anticoagulation. However, long-term anticoagulation in patients in sinus rhythm remains controversial. Multicenter randomized controlled trials of long-term anticoagulant and antiplatelet therapy among patients with chronic heart failure are under way and will hopefully provide rational scientific guidelines for the future management of patients.