History of bleeding is associated with higher risk of cardiovascular outcomes in patients with atrial fibrillation: a report from the GLORIA-AF registry phase III

Abstract Background Oral anticoagulants (OACs) are the cornerstone of thromboembolic risk prevention in patients with atrial fibrillation (AF), but associated with an increased risk of bleeding. Although non-vitamin K antagonist oral anticoagulants (NOACs) have improved safety, a significant proportion of patients with AF still develop bleeding. We aimed to assess the impact of a previous history of bleeding on the natural history of a contemporary cohort of patients with AF. Methods From the prospective GLORIA-AF Registry Phase III, which enrolled patients with a recent diagnosed of AF and a CHA2DS2-VASc score ≥1, we analysed patients according to their history of previous bleeding, as reported at enrolment. We analysed the association of prior bleeding with OAC and NOAC use using multivariable logistic regression model. Second, we evaluated the association of previous bleeding with a primary composite outcome of all-cause death and major adverse cardiovascular events (MACE) over a 3-year follow-up, using multivariable Cox-regression models. Other secondary outcomes were explored, including thromboembolic events, and subsequent major bleeding. Results 20,899 patients (mean age 70.2±10.3 years, 45.0% females) were included in this analysis, of which 1130 (5.4%) were reported a previous bleeding at baseline. Patients with prior bleeding events were older, and with a higher prevalences of cardiovascular comorbidities and risk factor; they also presented with higher mean CHA2DS2-VASc scores (4.6±1.5 vs. 3.1±1.5, p0.001) and HAS-BLED scores (2.5±0.9 vs. 1.3±0.9, p0.001). On multivariable logistic regression, older age, previous thromboembolic events, coronary artery disease and peripheral artery disease, as well as history of cancer and abnormal kidney function, were all associated with higher odds of having a previous history of bleeding (Figure 1). OAC were less used in patients with previous bleeding (OR: 0.66, 95%CI: 0.57-0.78). Specifically, NOAC were not more used than VKA in patients with previous bleeding (OR: 0.98, 95%CI: 0.83-1.15), but when used, the odds of receiving a reduced dose was higher (OR: 1.37, 95%CI: 1.14-1.64). During a median follow-up of 3.0 IQR: 2.9-3.1 years, patients with previous bleeding had higher incidence of the primary composite outcome of all-cause death and MACE (Figure 2), with higher risk observed at multivariable Cox regression analysis (HR: 1.18, 95%CI: 1.01-1.37). Among secondary outcomes, patients with previous bleeding showed higher risk of death (HR: 1.21, 95%CI: 1.02-1.44), MACE (HR: 1.29, 95%CI: 1.05-1.59) and subsequent major bleeding (HR: 1.92, 95%CI: 1.49-2.46), but not thromboembolic events (HR: 1.22, 95%CI: 0.92-1.60). Conclusions In a contemporary cohort of patients with AF, history of bleeding was associated with reduced use of OAC for thromboembolic risk prevention. Prior bleeding events were associated with a higher risk of subsequent cardiovascular events and major bleeding.Figure 1 Figure 2