Background and Purpose: Current American Heart Association/American Stroke Association guidelines identify warfarin use as a class IA indication in patients with atrial fibrillation (AF) and ischaemic stroke (IS) or transient ischaemic attack (TIA).  However, few studies have examined factors associated with long-term antithrombotic therapy use in IS/TIA patients with AF.

Methods: We utilised the Get With The Guidelines-Stroke national quality improvement registry and the Adherence eValuation After Ischemic Stroke Longitudinal (AVAIL) Registry to examine patterns of antithrombotic use at discharge and at 12 months in IS/TIA patients with AF in the USA.  A multivariate logistic regression model was developed to identify predictors of warfarin use in this patient population at 12 months.

Results: Of the 2460 IS/TIA patients, 291 (11.8%) had AF, of whom 5.5% were discharged on aspirin alone, 49.1% on warfarin alone, 1.4% on clopidogrel alone, 34.7% on warfarin plus aspirin, 2.1% on aspirin plus clopidogrel and 1.0% on aspirin plus clopidogrel plus warfarin.  Paradoxically, there was a decrease in the rate of warfarin use in patients with a CHADS2 score higher than 3.  The only factor associated with warfarin use at 12-month follow-up was male gender (adjusted odds ratio, 2.27; CI, 1.22 to 4.35; P = 0.01).

Conclusions: Overall, the use of warfarin therapy is high at discharge in IS/TIA patients with AF; however, there was a decrease in the rate of warfarin use in patients with a CHADS2 score over 3.  Compared to women, men were more likely to be on warfarin at 1 year after the index stroke event.  Therefore, opportunities exist to improve antithrombotic use in all IS/TIA patients with AF.