Objectives: To examine the use of warfarin in patients with atrial fibrillation (AF) admitted to hospital because of stroke or transient ischemic attack; and to describe the outcome of AF-associated stroke.
Design: Review of the medical records of patients, identified from a prospective registry, admitted from January 1, 1994 through December 31, 1996.
Setting: Tertiary care teaching hospital.
Results: AF was present in 92 of 722 (13%) patients at the time of admission. Only eight of 60 (13%) patients with ischemic stroke who were known to be in AF before their stroke were taking warfarin. The in-hospital case-fatality ratio for AF patients was more than double that of patients in sinus rhythm (21% versus 9%, respectively, P=0.001). AF patients were less likely to be discharged home (31% versus 59%, P=0.005). Of the 68 AF patients who survived, 74% left hospital taking warfarin. No warfarin-treated patient experienced intracranial bleeding while in hospital or during follow-up.
Conclusions: Patients with AF had more severe strokes than patients in sinus rhythm. A small proportion of patients with known AF were taking warfarin at the time of hospitalization. Bleeding complications were infrequent. Broader implementation of guidelines for the management of AF is justified to reduce the frequency of stroke in this group of patients.