Chicago—Warfarin is used only in a minority of elderly patients after bioprosthetic aortic valve replacement (BVR) surgery, despite the fact that this procedure carries an early risk of thromboembolic events. Elderly patients undergoing BVR were prescribed postprocedure warfarin only 35% of the time in an observational study of 3-month clinical outcomes using a Medicare database. “The underuse of warfarin in these patients with a high thromboembolic risk is worrisome,” said J. Matthew Brennan, MD, of Duke Clinical Research Institute in Durham, North Carolina. “Our study showed a difference favoring warfarin over no warfarin in reducing the risk of ischemic and intracranial hemorrhagic events in a group of elderly patients undergoing prosthetic aortic valve replacement. More than half of patients in our large sample were over the age of 75 years.” The study included 25,656 Medicare-linked BVR cases from 2004-2006 at 797 hospitals in the Society of Thoracic Surgeons Adult Cardiac Surgery Database. Discharge anticoagulant strategies (aspirin plus warfarin vs aspirin only) were compared, and results were stratified according to age, sex, and thromboembolic risk factor status. At discharge, aspirin alone was the most commonly used anticoagulant strategy (48.6%), followed by aspirin plus warfarin (23.3%) and warfarin alone (11.7%); the remainder of patients used other strategies. The group that used aspirin alone numbered 12,457 patients and those taking warfarin plus aspirin numbered 5972. Warfarin was used more often in risk factor–positive patients (44.1%) than in those who were risk factor negative (25.6%). The study looked at 3 major end points: death, hemorrhage, and ischemia. At 3 months, both strategies had a similar risk of death (4.0% in risk factor–positive patients and 3.9% in risk factor–negative patients). Patients on warfarin plus aspirin were more than twice as likely to sustain some type of bleeding (mainly gastrointestinal or hemopericardium) than those on aspirin alone. There was no difference in intracranial hemorrhage among groups. A prespecified subgroup analysis showed that warfarin plus aspirin reduced the death rate among risk factor–negative patients, and a trend toward a lower death rate was observed in risk factor–positive patients. These effects were most pronounced in patients >75 years of age, Dr. Brennan stated. “Based on this study, and on current guidelines, we recommend that patients undergoing BVR should be discharged on warfarin, yet only 35% of elderly patients in this study were prescribed warfarin. Now that we have these data on the benefits of warfarin in preventing stroke, we can discuss the choice with patients. Many patients don’t want to take warfarin. But if we use this data to discuss the choice with them, telling them there is a mortality and a stroke benefit, I think they would choose a chance of bleed over stroke,” Dr. Brennan told listeners.