Abstract
Purpose
Thromboembolic risk of atrial fibrillation (AF) in heart failure with preserved (HFpEF), mid-range (HFmrEF), and reduced ejection fraction (HFrEF) is not well identified. This study assessed the thromboembolic risk of AF in patients with HFpEF, HFmrEF and HFrEF.
Materials and methods
Within the CODE-AF prospective, outpatient registry (COmparison study of Drugs for symptom control and complication prEvention of Atrial Fibrillation), a total of 10476 patients with non-valvular AF including 929 (8.8%) patients with HF was analyzed. Multivariable cox regression was used to evaluate the risk of thromboembolic event, including stroke, systemic embolism and transient ischemic attack. Hazard ratio (HR) was adjusted by each component of CHA2DS2-VASc risk score and the use of oral anticoagulant (OAC).
Results
The median age of the overall population was 68.0 (interquartile range, 60.0–75.0); 63.9% were male. The proportion of HFpEF, HFmrEF and HFrEF was 43.6%, 26.7% and 29.7%, respectively. CHA2DS2-VASc risk score was higher in HF group than no-HF group. OAC was more commonly used in HF group than no-HF group (85.2% vs. 68.9%, p<0.001). The rate of OAC usage was 85.1%, 86.6%, and 84.0% in HFpEF, HFmrEF, and HFrEF group, respectively. During follow-up period of median 14.3 months, 15 patients experienced thromboembolic event in HF group with incidence rate of 1.39 events per 100 person-years, while 94 patients did in no-HF group with 0.87 events per 100 person-years. In patients without OAC, incidence rate of thromboembolic event was 1.31, 2.77, and 6.24 events per 100 person-years in HFpEF, HFmrEF, and HFrEF, respectively. Compared with no-HF group, HF was associated with increased risk of thromboembolic event with clinical variable adjusted HR of 3.04 (95% CI, 1.12–8.26, p=0.03). Among 3 types of HF, HFrEF increased the risk of thromboembolic event (adjusted HR 7.39, 95% CI 2.15–25.44, P=0.002), while HFmrEF or HFpEF did not. Finally, in patients with optimal OAC, risk of thromboembolic event was not increased by HF or HFrEF.
Conclusion
In OAC-naïve non-valvular AF, HF was associated with increased risk of thromboembolic event. Among 3 types of HF, HFrEF increased the risk of thromboembolic event, while HFmrEF or HFpEF did not. However, in patients with optimal OAC, even HFrEF was not associated with increased risk of thromboembolic event. These results support current OAC strategy in HF patients, especially emphasizing optimal OAC in HFrEF population.
Acknowledgement/Funding
The National Research Foundation of Korea