Clinical phenotype of anticoagulant discontinuation and the long-term prognosis in atrial fibrillation patients with high bleeding risk: A report from the GLORIA-AF Registry

Background: Discontinuation of oral anticoagulants (OACs) in patients with atrial fibrillation (AF) at high risk of bleeding is commonly seen. Objective: To explore risk factors leading to OAC discontinuation and the impact on clinical events in patients with AF at high bleeding risk. Methods: From the prospective, multi-center Global Registry on Long-Term Antithrombotic Treatment in Patients with Atrial Fibrillation (GLORIA-AF) study, we analyzed, by OAC discontinuation, all-cause death, cardiovascular death, major adverse cardiovascular events (MACE), thromboembolism, major bleeding, stroke and myocardial infarction (MI) during follow-up. The hazard ratio (HR) and confidence interval (CI) was determined by Cox regression. Results: Of 3,131 patients (age 74.9 ± 7.3 years, 39.5% male) with a HAS-BLED score ≥3, 633 (20.2%) stopped OAC during follow-up. After multivariate adjustment, age ≥80 years (HR: 1.26, 95% CI: 1.05-1.50), older with frailty (HR: 1.33, 95% CI: 1.12-1.94), chronic kidney disease (CKD, HR: 1.36, 95% CI: 1.14-1.63), chronic obstructive pulmonary disease (COPD, HR: 1.56, 95% CI: 1.20-2.03), peripheral arterial disease (PAD, HR: 1.38, 95% CI: 1.03-1.87), multimorbidity (HR:1.68, 95% CI:1.20-2.36) and polypharmacy (HR:1.59, 95% CI: 1.24-2.05) were associated with OAC discontinuation. The patients who stopped OACs during follow-up had a higher risk of all cause death (HR:1.74 , 95% CI:1.35-2.25), cardiovascular death (HR: 2.01, 95% CI:1.34-3.03), MACE (HR: 2.37, 95% CI:1.79-3.14), thromboembolism (HR: 1.96, 95% CI:1.28-2.99), major bleeding(HR: 4.35, 95% CI:2.79-6.78), stroke (HR: 2.42, 95% CI:1.57-3.73)and MI (HR: 2.02, 95% CI:1,12-3.66) after OAC discontinuation. The incidence of OAC discontinuation was higher in the patients with VKA compared with NOACs (23.1% vs. 19.3, P = 0.026). OAC discontinuation rate was consistent across regions and the risk of clinical events was similar in VKA and NOACs groups. Conclusion: Being an older patient with frailty, or with CKD, COPD, PAD, multimorbidity and polypharmacy were risk factors of OAC discontinuation. OAC discontinuation was independently associated with higher risk of mortality, major bleeding, thromboembolism and MACE in AF patients at high risk of bleeding.