Abstract 15490: Unmet Medical Needs in Non-Valvular Atrial Fibrillation Patients Who Use Warfarin for Secondary Prevention of Stroke

Abstract

Background: The efficacy of warfarin in preventing stroke is well documented in controlled clinical trials of patients with non-valvular atrial fibrillation (AF). It has been widely noted that warfarin is under-utilized in routine clinical settings.

Objective: To assess whether more warfarin use translates into favorable outcomes in AF patients.

Methods: We analyzed a large administrative claims database, i3LabRx, in the US. The study period was 2004-2009. Inclusion criteria were: age≥18 years, documented non-valvular AF, and one or more risk factor(s) for stroke. Exclusion criteria were: atrial fibrillation or flutter due to reversible causes, mitral stenosis, conditions other than atrial fibrillation that require chronic anticoagulation, or active infective endocarditis. Incidence rates and 95% confidence intervals, stratified by history of events of interest and by warfarin use (% days with warfarin possession from index date to events), were calculated for the following events: stroke/systemic embolism (SE), bleeding as a hospitalization discharge code, intracranial bleeding, gastrointestinal bleeding, myocardial infarction, and combined stroke, systemic embolism and all-cause death.

Results: A total of 91,798 individuals were included in the analyses; 47% were of age 75 or older, 42% were female. Hypertension was the most prevalent co-morbid condition (80%), followed by diabetes (30%). Distribution of CHADS2 0, 1, 2, and 3+ was 0.6%, 39.3%, 36.1% and 24.0%, respectively. Prior warfain use was 38%. Average follow-up period was 2 years. Among those with prior events, the risks of recurrences of stroke/SE, stroke/SE/death, and MI were 11-, 10-, and 9-fold, and the risks of recurrences of hospitalized bleeding, ICH, and GI bleeding was 1.2-, 1.5-, and 1.7-fold for those who had 68-100% of days with warfarin possession compared to those who had >0-33%. In contrast, in those with no prior events, the incidence rates of these events were lower for those who had higher percents of days with warfarin possession.

Conclusion: More warfarin use did not translate into lower risks of recurrences of events in non-valvular AF patients, suggesting ineffective secondary prevention of stroke by warfarin in routine practice and remaining unmet medical needs.