Newer Anticoagulants Linked to Fewer Strokes and Deaths After Brain Bleeds in AFib Patients

Of 1,502 people in a Finnish-led study, all of them living with atrial fibrillation and a brain bleed in their past, 257 died during follow-up. Fifty-seven had an ischemic stroke. Forty-five bled into the brain again. The numbers come from a study published October 21, 2026, in Neurology, the journal of the American Academy of Neurology, and they bear on a hard question: whether to give blood thinners to people who have already had a brain bleed.

The bind is real. Atrial fibrillation, an irregular heartbeat, raises the risk of ischemic stroke, in which blood flow to part of the brain is blocked, often by a clot. Anticoagulants, which keep blood from clotting easily, can help prevent that. But for someone who has had an intracranial hemorrhage, a brain bleed, doctors worry that the drugs could cause another. Paula Tiili, MD, of the University of Helsinki, said treatment decisions in this group can be complex for exactly that reason, and her team wanted to understand the risks of both stroke and bleeding with newer and older drugs.

Participants had an average age of 78. Their brain bleeds had occurred at least four months before their AFib diagnosis, and they were followed for up to two years. Fifteen percent received newer anticoagulants, known as direct oral anticoagulants, a group that includes apixaban, rivaroxaban, edoxaban and dabigatran. Twenty-nine percent received warfarin, an older anticoagulant. The remaining 56 percent received no anticoagulant.

After adjusting for factors such as age, high cholesterol and high blood pressure, the researchers compared outcomes one year after the AFib diagnosis. Fewer than 1 percent of people on newer anticoagulants had an ischemic stroke, against 2 percent on warfarin and 4 percent on no anticoagulant. New brain bleeds occurred in 1 percent of the newer-drug group, 3 percent of the warfarin group and 2 percent of those taking nothing. Deaths from causes other than stroke in that first year reached 5 percent among newer-drug users, 9 percent with warfarin and 15 percent with no anticoagulant.

Against no medication, newer anticoagulants were associated with an 83 percent lower risk of ischemic stroke and an 80 percent lower risk of death, with no increase in intracranial bleeding. Warfarin was associated with a 51 percent lower stroke risk and a 40 percent lower risk of death, but also an 85 percent higher risk of brain bleeds.

The study is observational, so it cannot show that the drugs caused these differences, only that they were associated with them. There is a further limit: researchers included only people who neither had a stroke nor died in the first three months after their AFib diagnosis. The results may therefore not apply to that earlier window.

Tiili said the findings may help clinicians make more individualized decisions about preventing stroke in people with AFib and prior brain bleeds. She also cautioned that more research is needed to identify who is most likely to benefit and who may still face a higher bleeding risk. That question, rather than whether to treat at all, is where the work now points.