By Jamie L. W. Kennedy, MD, FACC

Associate Professor, Division of Cardiology, Advanced Heart Failure & Transplant Cardiology, University of California, San Francisco

Dr. Kennedy reports no financial relationships relevant to this field of study.

SYNOPSIS: In the Atherosclerosis Risk in Communities (ARIC) cohort, incident heart failure hospitalization with either preserved or reduced ejection fraction was associated with long-term increased risk of venous thromboembolic events.

SOURCE: Fanola CL, Norby FL, Shah AM, et al. Incident heart failure and long-term risk for venous thromboembolism. J Am Coll Cardiol 2020;75:148-158.

Heart failure (HF) is recognized as a risk factor for a venous thromboembolic event (VTE), although the magnitude of this risk is not well-defined. Hospitalized HF patients routinely receive prophylactic measures, either pharmacological or mechanical. It is rare to continue prophylaxis following discharge, although a sizable number of HF patients are chronically anticoagulated for indications such as atrial fibrillation. Fanola et al sought to understand the risk of VTE conferred by a new diagnosis of HF in a large, prospective, population-based study.

The authors of the Atherosclerosis Risk in Communities (ARIC) study recruited 15,792 men and women between 1987 and 1989, ranging in age from 45 to 64 years. Participants primarily were African American and Caucasian. Participants have been followed via phone calls at least annually, along with periodic in-person exams. These data have provided a wealth of information on the development of HF, coronary disease, and other cardiovascular diseases.

The authors of the VTE analysis nested three separate analyses in their work, differentiated by increasingly robust information available about the ARIC participants over time. The first analysis concerned the rate of VTE in patients following an incident HF hospitalization compared to patients not diagnosed with HF, starting with enrollment in the ARIC study. They found the incidence of VTE in patients with HF was 11.8 per 1,000 patient-years compared to 2.3 per 1,000 patient-years for patients without HF, for a hazard ratio (HR) of 3.13. Understandably, this risk was highest in the month following HF hospitalization, but remained significantly elevated over long-term follow-up. Interestingly, the risk for African Americans with HF was increased 4.4-fold, compared to 2.4-fold for whites with HF.

The second analysis, starting in 2005, concerned the rate of VTE in patients with HF with preserved ejection fraction (HFpEF) vs. HF with reduced ejection fraction (HFrEF). As before, the rate of VTE was substantially higher in HF patients, 21.8 per 1,000 patient-years vs. 3.04. The rates were similar between HFrEF and HFpEF (24.4 vs. 23.1 per 1,000 patient-years; HR, 5.53 vs. 4.71). The subset of patients with unknown EF had a VTE rate of 19.0 per 1,000 patient-years. Similar to the first analysis, VTE occurred most frequently in the month following hospitalization, but the excess risk persisted over the long term. In this analysis, race was no longer associated with increased VTE risk.

The final analysis, starting in 2011, concerned the relationship between a variety of echocardiographic markers associated with the development of HF and risk of VTE in patients without documented HF. Increased left ventricular (LV) relative wall thickness and mean LV wall thickness both conferred elevated risk of subsequent VTE (HR, 1.25 and 1.32, respectively). Interestingly, left atrial volume index, perhaps the best indicator of chronically elevated filling pressures, was associated with VTE in the univariate analysis, but the association became insignificant after correction for comorbid conditions. HF with either reduced or preserved EF carries a significantly higher risk for VTE. Furthermore, patients with echocardiographic findings conferring increased risk of developing HF also carry a higher risk of VTE, suggesting a prothrombotic state is an early development in the pathway to symptomatic HF. The ARIC analysis was limited by the requirement for hospitalization for both HF and VTE diagnosis. The true VTE rate may be somewhat higher than reported.


How do clinicians use these data? Would HF patients benefit from prophylactic anticoagulation following discharge? There have been several studies of extended (four- to six-week) courses of prophylactic anticoagulation in acutely ill medical patients following discharge: EXCLAIM (enoxaparin), MAGELLAN (rivaroxaban), ADOPT (apixaban), and APEX (betrixaban). HF patients were well-represented in these studies, ranging from 18-44% of enrolled subjects. These studies have produced, at best, mixed results. Only the MAGELLAN study met its primary endpoint (demonstrating a reduction in VTE), although at the expense of significantly more bleeding complications. Based on these data, four to six weeks of pharmacological prophylaxis following discharge from a HF hospitalization does not seem to be helpful.

The ARIC data reveal the persistently elevated VTE risk; perhaps long-term prophylactic anticoagulation would be beneficial. There are no studies directly addressing this question, but the COMMANDER HF trial may shed some light on the subject. Those authors examined the long-term use of low-dose rivaroxaban (2.5 mg twice daily) vs. placebo in patients with systolic HF and coronary artery disease. The addition of rivaroxaban did not affect the primary composite endpoint of death, myocardial infarction, or stroke over a median follow-up of 21.1 months. Of these three components, the stroke rate was significantly lower in the rivaroxaban group. It is worth noting the mortality rate in this study was high at 22%.

Symptomatic deep vein thrombosis (DVT) and pulmonary embolism (PE) were two secondary endpoints in this study. There was no difference between the treatment and control arms. Interestingly, the rates were much lower than reported in the ARIC analysis (0.10 to 0.15 per 100 patient-years for DVT vs. 0.19 to 0.23 for PE). Based on this study, long-term prophylaxis does not appear beneficial, although it does seem that COMMANDER HF captured advanced stage HF patients, while the ARIC cohort represents a wider range of HFrEF severity as well as HFpEF patients.

I have to admit I have scoffed frequently at the seemingly requisite emergency room lower extremity ultrasound in decompensated HF patients with symmetric bilateral lower extremity edema. Perhaps a higher index of suspicion is one takeaway from the ARIC data. Patients with somewhat atypical symptoms, out of proportion pulmonary hypertension, or significant right heart dysfunction, and those who fail to improve with HF management, should prompt consideration of VTE. Another lesson is to review VTE prophylaxis protocols, update them if needed, assess implementation rates, and explore any barriers to implementation.