Predicting the Risk of Venous Thromboembolism in Patients Hospitalized With Heart Failure

Study Questions:

What is the relationship between heart failure (HF) severity and the risk of venous thromboembolism (VTE) in hospitalized, acutely ill patients?


In the MAGELLAN study, hospitalized patients were randomized to receive rivaroxaban 10 mg daily (35 ± 4 days) or enoxaparin 40 mg daily (10 ± 4 days) plus placebo. In this subgroup analysis, 2,593 (32% of study population) patients with HF with New York Heart Association (NYHA) class III-IV symptoms were stratified based on N-terminal pro-brain natriuretic peptide (NT-proBNP) levels and plasma D-dimer concentrations. The coprimary outcomes were the composite of asymptomatic proximal deep vein thrombosis (DVT), symptomatic DVT, symptomatic nonfatal pulmonary embolism (PE), and VTE-related death by day 10 and day 35.


Patients with more severe HF (higher NT-proBNP and D-dimer) had a greater incidence of VTE compared to patients with less severe HF (4.3% vs. 2.2%, p = 0.0108 at day 10 and 7.2% vs. 4.1%, p = 0.0150 at day 35). In a multivariate analysis, NT-proBNP concentration was associated with VTE risk at day 10 (p = 0.017), whereas D-dimer concentration was associated with VTE risk at day 35 (p = 0.005). Classification of HF severity by NYHA class III or IV did not associate with VTE risk. NT-proBNP levels also correlated with symptomatic VTE risk and risk of cardiovascular death (p = 0.0248 and p < 0.001, respectively). Treatment with rivaroxaban was associated with lower risk of VTE. VTE risk did not stratify based on NT-proBNP in the rivaroxaban-treated group, but did stratify in the enoxaparin-treated group. Clinically relevant bleeding rates were not statistically increased in patients with elevated NT-proBNP (p = 0.1929 at day 10 and p = 0.2621 at day 35).


The authors concluded that hospitalized patients with more severe HF are at increased risk of VTE development and cardiovascular death, compared to patients with less severe HF. NT-proBNP may be able to predict VTE risk at day 10, whereas D-dimer may be able to predict VTE risk at day 35 in hospitalized HF patients.


This subgroup analysis of the MAGELLAN study highlights the important association between hospitalized patients with HF and the risk of VTE. For the first time, this study suggests that the severity of HF may correlate with the risk of VTE and cardiovascular death, which can reach 6% in patients receiving thromboprophylaxis. However, in many medical centers, VTE prophylaxis is not continued beyond the hospitalization period, suggesting that the rate of VTE development may be higher than reported in the MAGELLAN study. This also limits the applicability of the MAGELLAN study findings for many (most?) patients in US hospitals. In fact, prior studies have suggested that up to two-thirds of hospitalized HF patients do not receive adequate thromboprophylaxis. Thromboprophylaxis is important for all hospitalized HF patients. Balancing the thrombosis and bleeding risk is critical when determining which hospitalized HF patients might benefit from extended VTE prophylaxis.

< Back to Listings