Atrial Fibrillation After Brain Bleed: DOAC vs Warfarin

A new study in Neurology looked at people with atrial fibrillation (AFib) who had previously had an intracranial hemorrhage (ICH). In this group, taking anticoagulants was associated with a lower risk of ischemic stroke and death. Direct oral anticoagulants (DOACs) were not associated with a higher risk of new brain bleeds, but warfarin was. The data are observational, so they show associations, not proof of cause and effect.

What happened

Researchers from the University of Helsinki, led by Paula Tiili, M.D., studied 1,502 people with AFib and a history of brain bleeding. The average age was 78. The brain bleed had occurred at least four months before the AFib diagnosis. Participants were followed for up to two years.

Treatment groups were uneven:

• 15% received a DOAC (apixaban, rivaroxaban, edoxaban or dabigatran)
• 29% received warfarin
• 56% received no anticoagulant

During follow-up, 57 people had an ischemic stroke, 45 had a new brain bleed and 257 died.

After adjusting for factors such as age, high cholesterol and high blood pressure, the one-year estimates after AFib diagnosis were:

| Outcome at 1 year | DOAC | Warfarin | No anticoagulant | |---|---|---|---| | Ischemic stroke | <1% | 2% | 4% | | New brain bleed | 1% | 3% | 2% | | Death from causes other than stroke | 5% | 9% | 15% |

Compared with no anticoagulation, DOAC use was associated with an 83% lower risk of ischemic stroke and an 80% lower risk of death, with no increased risk of intracranial bleeding. Warfarin was associated with a 51% lower risk of ischemic stroke and a 40% lower risk of death. It was also associated with an 85% higher risk of brain bleeds.

Why it matters

Anticoagulation for stroke prevention in AFib is a core topic in cardiology and neurology. The hard cases are patients who have already had a brain bleed. Clinicians worry that anticoagulation could cause another one. Withholding it leaves the patient exposed to cardioembolic stroke.

As Tiili put it, treatment decisions in this setting "can be complex due to concern about the possibility of another brain bleed." The study suggests that in this older, high-risk population, the untreated group had the highest rates of ischemic stroke and death. The authors say the findings may help clinicians make more individualized decisions. They also stress that more research is needed to identify who is most likely to benefit and who may still face a higher bleeding risk.

This article is educational and is not a recommendation for any individual patient.

What to know for the USMLE — AFib, DOACs, warfarin and bleeding risk

These are the high-yield concepts behind the news:

• Pathophysiology of AFib stroke: Loss of coordinated atrial contraction causes stasis, mainly in the left atrial appendage. Thrombi form and embolize, usually to the middle cerebral artery territory. This is a cardioembolic ischemic stroke.
• CHA₂DS₂-VASc estimates stroke risk in nonvalvular AFib:
• Congestive heart failure, Hypertension, Diabetes, Vascular disease, and female Sex category: 1 point each
• Age 65–74: 1 point
• Age ≥75 and prior Stroke/TIA/thromboembolism: 2 points each
• HAS-BLED estimates bleeding risk:
• Hypertension (uncontrolled), Abnormal renal or liver function, Stroke, Bleeding history, Labile INR, Elderly (>65), Drugs/alcohol
• It is mainly used to flag modifiable risk factors, such as uncontrolled blood pressure, alcohol and interacting drugs. A high score alone is not usually a reason to withhold anticoagulation.
• Warfarin: It inhibits vitamin K epoxide reductase (VKORC1), which reduces gamma-carboxylation of factors II, VII, IX, X and proteins C and S.
• It is monitored with PT/INR and acts through the extrinsic pathway first.
• Early protein C depletion can cause warfarin-induced skin necrosis.
• It is teratogenic, so pregnancy calls for heparin instead.
• Reversal uses vitamin K, 4-factor PCC or fresh frozen plasma.
• DOACs:
• Apixaban, rivaroxaban and edoxaban are direct factor Xa inhibitors.
• Dabigatran is a direct thrombin (factor IIa) inhibitor.
• They need no routine INR monitoring and have fewer dietary interactions than warfarin.
• Reversal: idarucizumab for dabigatran, and andexanet alfa (or PCC) for factor Xa inhibitors.
• DOAC vs warfarin in AFib: In nonvalvular AFib, DOACs are generally preferred over warfarin. Warfarin remains the choice for AFib with moderate-to-severe mitral stenosis or a mechanical heart valve.
• ICH background: Hypertensive hemorrhage typically involves the basal ganglia, thalamus, pons and cerebellum. Lobar hemorrhage in older adults raises the question of cerebral amyloid angiopathy. Knowing the cause of the earlier bleed matters when estimating the risk of a new one. This study summary does not report the causes of the prior bleeds.
• Question-stem clue: A patient with AFib and a remote ICH is a prompt to weigh stroke risk against bleeding risk rather than to apply a reflex "never anticoagulate." Exam questions usually reward identifying the best-supported drug class, such as a DOAC for nonvalvular AFib.

Limitations

• Observational design. The study shows associations only. The authors state that the findings do not show that anticoagulants caused the differences in risk.
• Non-random treatment. Groups were unequal (15%, 29% and 56%). Doctors choose who gets which drug, and the people who received nothing may have differed in ways the adjustment could not capture. Adjustment covered factors such as age, high cholesterol and high blood pressure. The source does not list all the variables.
• Small event counts. Only 57 ischemic strokes and 45 new brain bleeds occurred across the whole cohort. Small numbers make estimates, especially for the DOAC group, less precise.
• Immortal-time-type exclusion. Researchers included only people who did not have a stroke or die in the first three months after AFib diagnosis. The results may not apply to that earlier period.
• Short follow-up. Follow-up was up to two years, so longer-term safety is not addressed.
• Timing of the bleed. The brain bleed occurred at least four months before the AFib diagnosis. The findings may not extend to AFib diagnosed soon after a bleed.
• Death outcomes. The one-year figures reported for death refer to causes other than stroke. The overall mortality estimate was reported separately as a risk reduction.

FAQ

Are DOACs safer than warfarin after an intracranial hemorrhage in AFib?

In this observational study, DOACs were not associated with an increased risk of new brain bleeds compared with no anticoagulation. Warfarin was associated with an 85% higher risk. This is an association, not proof, and individual decisions belong to treating clinicians.

Does this study prove anticoagulants prevent strokes in people with prior brain bleeds?

No. Because it was observational, it cannot prove causation. It does show that, in this cohort, anticoagulant use was linked to fewer ischemic strokes and deaths. Randomized trials would be needed for stronger evidence.

What should I memorize about DOACs for Step 1 and Step 2 CK?

Know the targets: apixaban, rivaroxaban and edoxaban inhibit factor Xa, and dabigatran inhibits thrombin. Know the reversal agents: idarucizumab and andexanet alfa. Remember that warfarin is still preferred for mechanical valves and moderate-to-severe mitral stenosis.

Key takeaways

• In 1,502 adults with AFib and a prior brain bleed, DOAC use was associated with lower ischemic stroke and mortality and no increase in new brain bleeds. Warfarin was associated with lower stroke and mortality but 85% higher bleeding risk.
• The study is observational, with small event counts and exclusion of the first three months, so it cannot prove causation.
• For the USMLE, focus on CHA₂DS₂-VASc, HAS-BLED, the mechanisms and reversal of warfarin and DOACs, and the exceptions where warfarin is still preferred.