Oral Anticoagulant Use in Patients with Atrial Fibrillation at Low Risk of Stroke and Associated Bleeding Complications.

Abstract

The use of oral anticoagulants (OACs) in patients with atrial fibrillation (AF) and low stroke risk might cause more harm than benefit. Little attention has been given to address its prevalence and associated consequences. This study aimed to investigate the prescription rate of OACs, identify associated factors, and describe incident bleeding events in low-risk patients. We included patients with a new diagnosis of AF between 1 January 2011 and 31 December 2018 having a low risk of stroke (CHA2DS2-VASc score of 0 for males and 1 for females) from Australian general practice data (MedicineInsight). Patients were classified as OAC users if there was a recorded prescription of an OAC within 60 days of AF diagnosis, and factors associated with the prescription of an OAC were assessed using logistic regression. Recorded incident bleeding events were identified within 6 months after AF diagnosis or after OAC initiation for OAC non-users and users, respectively. The risk of bleeding was compared between the two groups by adjusting their baseline differences using propensity score matching. The study included 2810 low-risk patients (62.3% male) with a mean age of 49.3 ± 10.8 years. Of the total, 705 (25.1%) patients had a record of OAC prescription within 60 days of diagnosis of AF. Older age (odds ratio [OR] 1.03; 95% confidence interval [CI] 1.03-1.04) and diagnosis periods (2015-2016 [OR 1.46; 95% CI 1.10-1.94] and 2017-2018 [OR 1.65; 95% CI 1.17-2.23] vs. 2011-2012) were associated with higher odds of OAC initiation. Female sex (OR 0.71; 95% CI 0.59-0.85), higher bleeding risk (ORBIT score; OR 0.80; 95% CI 0.68-0.94), and higher socioeconomic index for areas (SEIFA) quintiles (SEIFA quintiles; 2 [OR 0.65; 95% CI 0.48-0.88], 3 [OR 0.74; 95% CI 0.56-0.98], 4 [OR 0.70; 95% CI 0.52-0.94], 5 [OR 0.69; 95% CI 0.52-0.91] compared with quintile 1) were associated with lower odds of OAC prescription. A total of 52 (in 1.8% of patients) incident bleeds were identified, with 18 (2.6%) among OAC users. The rate of bleeding was not significantly different between users and non-users after matching. However, within OAC users, commencement of OAC was associated with an increased risk of bleeding compared to the period before OAC initiation (p = 0.006). One in four patients at low risk of stroke received an OAC within 60 days of AF diagnosis. Older age and the period following the widespread availability of direct-acting OACs were associated with an increased likelihood of OAC prescription. Positively, using OACs was not associated with an increased rate of bleeding compared to non-users.