Clinical characteristics and frequency of anticoagulation in patients with atrial fibrillation and heart failure: results of a retrospective big data analysis

Abstract

Aim. To evaluate the clinical characteristics and frequency of prescribed anticoagulant therapy for patients with atrial fibrillation (AF) and heart failure (HF) in subjects of the Russian Federation based on a retrospective big data analysis using artificial intelligence technologies.Material and methods. For retrospective analysis, information was obtained from the Webiomed predictive analytics platform, which includes depersonalized data from electronic health records of outand/ or inpatients in 6 subjects of the Russian Federation, extracted using artificial intelligence technologies. From the database of patients with AF (n=144431), a group of individuals (n=20970) with an established diagnosis of HF and information on left ventricular ejection fraction (LVEF) was selected.Results. Patients with AF and HF (men, 43,7%; age 72,1±13,2 years; LVEF, 58,9±11,0%) had a history of smoking in 36,6% of cases, hypertension — in 86,7%, type 2 diabetes — in 26,6%, gout — in 2,7%, stage III and IV-V chronic kidney disease — in 50,9 and 15,6%, lower limb peripheral arterial disease — in 15,8%. The incidence of ischemic stroke, LV myocardial infarction and pulmonary embolism was 8,8, 14,7 and 2,4%, respectively. Anticoagulants, including direct oral ones, were administered to patients with AF and HF in 62,5% and 32,0% of cases, respectively. The frequency of their appointment did not significantly differ depending on LVEF.Conclusion. Patients with AF and HF are characterized by significant comorbidity, a higher incidence of cardiovascular events compared with the group of individuals with AF without HF, and an unsatisfactory percentage of anticoagulant therapy.

Keywords

• atrial fibrillation
• heart failure
• anticoagulant therapy
• retrospective analysis
• artificial intelligence