Incident heart failure tied to increased long-term risk for VTE

Incident heart failure tied to increased long-term risk for venous thromboembolism

Findings suggest the need for evidence-based strategies to prevent long-term venous thromboembolism in these patients beyond hospitalisation time.

Main Takeaway

• Incident heart failure (HF) hospitalisation was associated with more than a 3-fold higher risk for venous thromboembolism (VTE) which persisted through a long-term follow-up.
• The increased risk for VTE persisted in participants with HF with preserved ejection fraction (HFpEF) and HF with reduced ejection fraction (HFrEF).
• In participants without clinical HF, left ventricular (LV) relative wall thickness and mean wall thickness were predictors of VTE.

Why This Matters

• HF not only increases the risk for cardioembolic events and ischaemic stroke but also increases the risk for VTE.
• Hospitalised patients for acute HF are at an increased risk for short-term VTE due to prolonged immobilisation and hypercoagulability

Study Design

• This prospective population-based study included 13,728 participants from the Atherosclerosis Risk In Communities (ARIC) cohort.
• Patients with (mean age, 56.1 years) and without (mean age, 53.5 years) incident HF hospitalisation were included.
• Short- and long-term VTE risks associated with:
  • incident HF during 1987-2015 (n=13,728; design 1);
  • HF subtype during 2005-2015 (n=7588; design 2); and
  • abnormal echocardiographic measures in the absence of clinical HF during 2011-2015 (n=5438; design 3) were assessed.
• Funding: ARIC cohort was funded by the National Heart, Lung, and Blood Institute.

Key Results

• Incident HF occurred in 19.6% of patients included in the design¹.
• During a mean follow-up of 22±7 years, incident HF vs no HF was associated with an increased risk for incident total VTE (adjusted HR [aHR], 3.13; 95% CI, 2.58-3.80).
• Incident HF vs no HF was associated with a 2-fold greater risk for all VTE subtypes:
  • Pulmonary embolism: aHR, 2.57; 95% CI, 1.95-3.39;
  • Deep venous thrombosis: aHR, 3.90; 95% CI, 2.96-5.13;
  • Provoked VTE events: aHR, 2.26; 95% CI, 1.60-3.19; and
  • Unprovoked VTE events: aHR, 3.72; 95% CI, 2.94-4.72.
• Incident HF occurred in 15.3% of participants allocated to the design².
• Over a mean follow-up of 9.7±2.6 years, VTE rates in those with vs without HF were 21.8 (95% CI, 16.9-27.6) vs 3.04 (95% CI, 2.64-3.49) per 1000 person-years.
• Compared with no HF, VTE risk was reported in:
  • Any HF: aHR, 5.15; 95% CI, 3.80-6.98;
  • HFpEF: aHR, 4.71; 95% CI, 2.94-7.52;
  • HFrEF: aHR, 5.53; 95% CI, 3.42-8.94; and
  • Undetermined incident HF: aHR, 4.09; 95% CI, 2.60-6.44.
• In participants allocated to the design 3, 86 incident VTE events were reported over a mean follow-up of 3.5±0.7 years.
• In the absence of baseline HF, LV relative wall thickness (P=0.002) and mean LV wall thickness (P=0.003) were found to be predictors of incident VTE after additional adjustment for chronic kidney disease and atrial fibrillation

Limitations

• ARIC-based incident HF on hospitalisations.
• VTE validation failed to capture outpatient VTE.
• The number of VTE events for design 3 was small.
• The observed association between HF and VTE may not be causal.
• Incident VTE-related mortality was not assessed.