Abstract 14178: Incidence of Bleeding Events in Patients with Atrial Fibrillation

Abstract

Objective: To describe the characteristics of patients with non-valvular atrial fibrillation (NVAF) and to estimate the incidence of major bleeding events in patients with AF.

Methods: A cohort study in newly diagnosed NVAF patients (2005-2010) was performed in the General Practice Research Database (GPRD). Linked data to the Hospital Episode Statistics and death certificates from the Office for National Statistics were included. Major bleeding was defined as primary hospital admission diagnosis based on previously published ICD-10 algorithms and stratified by intra-, extracranial and gastrointestinal (GI) bleeding. Current vitamin K antagonists (VKA) treatment was defined as a valid prescription/INR measurement during the last 28 days, recent as a treatment period from 28 to 90 days, and past treatment as more than 90 days. No treatment was the time between diagnosis of AF and first prescription/INR measurement, or the whole follow-up for patients that have never been treated.

Results: Data from 16,513 AF patients could be linked; 48% were female, the mean age (std) was 74 (12) years, the median duration of follow-up was 1.7 years (range 0-5). According to CHADS₂ score 21% were at low risk, 61% at moderate and 18% at high risk for stroke. Overall 55.8% of the patients had at least one period of VKA use, and 782 bleeding events were recorded (n=120 intra-, n=662 extracranial including n=209 GI). The crude incidence rate (IR) per 100 patient years (95% CIs)) of all bleeds was 2.5 (2.3-2.7) and for fatal bleeds 0.2 (0.2-0.3). The IR of intracranial bleeds was 0.4 (0.3-0.5), 38% of them were fatal. The IR of extracranial bleeds was 2.1 (2.0-2.3), 3% were fatal; GI bleeds occurred with an IR of 0.7 (0.6-0.8), 8% were fatal. Stratification by period of VKA use resulted in IR for current, recent, past and no use for all bleeds of 3.0 (2.6-3.3), 3.4 (2.7-4.3), 2.0 (1.5-2.5), and 2.2 (1.9-2.4).

Conclusion: Current and recent users of VKAs had higher bleeding rates than past/no users. The IR of bleedings in non-users might indicate contra-indications for VKA use and/or presence of additional risk factors for bleeding, such as e.g. concomitant medications (aspirin) or comorbidities (renal impairment). However, no causal conclusions can be drawn from these crude incidence estimates.