Among the 188 patients who had a stroke while taking warfarin, 15 percent of those with an INR level less than 2.0 on admission to the hospital either experienced a severe stroke or died in the hospital, compared with only 5 percent of those with an INR level greater than 2.0. For patients who had been taking aspirin, 13 percent had a severe stroke or died in the hospital, and similar results were seen in 22 percent of those taking neither warfarin nor aspirin. Death within 30 days of the stroke including those patients who died in the hospital occurred in 6 percent of warfarin-taking patient with INR greater than 2.0, 16 percent with INR less than 2.0, 15 percent of those taking aspirin, and 24 percent of those taking neither drug.

The study authors note that there has been some controversy recently about the correct target INR levels for patients with atrial fibrillation receiving anticoagulation therapy. While higher INRs reduce the likelihood of a clot forming, they also increase the risk of bleeding problems, including brain hemorrhage. Some guidelines have recently suggested that certain older patients with atrial fibrillation should have lower INR targets to reduce the risk of hemorrhage, but the researchers note that INR targets less than 2.0 increase the likelihood that a stroke will be fatal or debilitating.

"All anticoagulation therapy has risks, and balancing those risks against the possibility that insufficient therapy will lead to a severe stroke is a serious concern," Hylek says. "Our results show that the risk of intracranial hemorrhage does not increase until INR levels reach 4.0, which should assure patients and physicians that the currently advocated INR target of 2.5 a range of 2.0 to 3.0 is most likely the right balance point." Hylek is an assistant professor of Medicine at Harvard Medical School.

Other authors of this study are Daniel Singer, MD, senior author; Y
'"/>

Contact: Sue McGreevey
smcgreevey@partners.org
617-724-2764
Massachusetts General Hospital
10-Sep-2003