Discontinuation of anticoagulation and its associated factors in atrial fibrillation

Abstract

Abstract Background/Introduction Oral anticoagulation has been shown to reduce the incidence of embolic events associated with atrial fibrillation, however the discontinuation of these drugs observed in clinical trials may not be representative of the real-world setting due to the inclusion of selected populations. Purpose The objective of the study was to compare whether there is a difference in the incidence rate of discontinuation between vitamin K antagonists and direct anticoagulants and to evaluate the factors associated with non-adherence in patients newly diagnosed with atrial fibrillation. Methods This was a prospective cohort study. Adult patients (age >18 years) who had newly started anticoagulation therapy for atrial fibrillation or atrial flutter were enrolled. Patients with previous oral anticoagulation, mechanical prosthetic heart valves or creatinine clearance less than 30 mL/min were excluded. The follow-up period was 12 months. As the treatment allocation was not randomized, a propensity score weighting was performed considering baseline characteristics potentially associated with exposure and outcome. Factors associated with anticoagulant discontinuation were evaluated using a weighted Cox model. Results A total of 379 patients were included (mean age 78±9 years, 58% females). The median follow-up was 362 days (IQR 347–370). Loss to follow-up was 1%. The anticoagulation discontinuation rate was 24.6% in the direct anticoagulant group and 15.6% in the vitamin K antagonist group. A weighted model of time to discontinuation of anticoagulation treatment showed a crude HR of 1.40 (95% CI 0.79–2.48) for the direct anticoagulant therapy group compared to the vitamin K antagonist group, and a model adjusted for age, type of atrial fibrillation, radiofrequency ablation, bleeding, number of chronic drugs, and cardiology consultations during follow-up showed an adjusted HR of 1.26 (95% CI 0.75–2.12). The main reason for discontinuation of anticoagulation was high risk of bleeding in the vitamin K antagonist group and performance of a radiofrequency ablation procedure in the direct anticoagulant group. There was no statistically significant difference in the discontinuation rate between both groups when the baseline characteristics of the patients were considered. Conclusion(s) There was no statistically significant difference in the discontinuation rate between both groups when the baseline characteristics of the patients were considered. The study showed that the discontinuation of anticoagulants in atrial fibrillation in our setting was not associated with the type of drug used, the age of patients, or the type of arrhythmia. Funding Acknowledgement Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Boehringer Ingelheim Figure 1