ENRICH-AFAtrial FibrillationOpen Access
Edoxaban Versus No Anticoagulation in Atrial Fibrillation After Intracranial Haemorrhage
Edoxaban in intracranial haemorrhage survivors with atrial fibrillation
Why the trial?
Survivors of intracranial haemorrhage with atrial fibrillation face competing risks of ischaemic stroke and rebleeding, and were excluded from the pivotal anticoagulation trials. Whether restarting oral anticoagulation with edoxaban provides net benefit in this population was a long-standing evidence-free dilemma.
Does edoxaban reduce stroke or systemic embolism in patients with high-risk atrial fibrillation and prior intracranial haemorrhage?
Population
948 patients with high-risk AF and prior intracranial haemorrhage (mean age 77)
Comparison
Edoxaban 60 mg daily (label-adjusted 30 mg) vs no anticoagulation
Design
Investigator-initiated open-label blinded-endpoint RCT (174 sites, 20 countries)
Follow-up
Mean 28 months
Key result
Edoxaban did not significantly reduce stroke or systemic embolism versus no anticoagulation in AF patients with prior intracranial hemorrhage (11.8% vs 12.8%; HR 0.88; 95% CI 0.61-1.26; p=0.48).
Authors
Discussion
Key questions
Member takes
Where experts stand
Experts read ENRICH-AF as a clear null result for routine anticoagulation after intracranial haemorrhage, with the ischaemic stroke benefit offset by a doubling of major bleeding, reinforcing the need for individualised decision-making while awaiting further trials.
The expert reaction is one-sided: no one defends routine edoxaban use in this population. Clinicians and the trial investigators alike stress that the ischaemic stroke reduction was cancelled out by excess bleeding, leaving no net benefit. The live question is whether ongoing trials comparing different anticoagulants or using individual-patient-level meta-analyses can identify subgroups who do benefit.
Agreement
Multiple experts converge on the same conclusion: edoxaban should not be used routinely in unselected AF patients with prior intracranial haemorrhage, and the path forward is individualised decision-making while awaiting additional trial data.
2 clinicians say this directly
What they’re arguing about
supportiveneutralcautiouscritical
Counts are expert takes we classified by axis. Tap a row to see the takes behind its count.
Still unclear
Whether specific subgroups, such as patients with non-lobar haemorrhage, might still derive net benefit from anticoagulation remains unanswered. The trial investigator points to the ongoing ASPIRE trial (apixaban vs aspirin) and the COCROACH individual-patient-data meta-analysis as the next sources of evidence. Until then, clinicians lack clear guidance on which of these high-risk patients should receive anticoagulation.
Key expert perspectives
No efficacy gain, bleeding more than doubled, and haemorrhagic stroke nearly tripled
Papakonstantinou highlights that edoxaban failed on efficacy, more than doubled major bleeding, and nearly tripled haemorrhagic stroke. He concludes that individualisation of anticoagulation is the key takeaway.
Distilled from their postX post
ASAshkan ShoamaneshStroke neurologist, McMaster UniversityResults readoutAug 30
Routine edoxaban is not supported, but individualised decisions and further trials are needed
Shoamanesh, the trial's lead investigator, states the findings do not support edoxaban in unselected AF patients after intracranial haemorrhage. He urges an individualised approach and says physicians should wait for more data from the ASPIRE trial and the COCROACH individual-patient-data meta-analysis before drawing broader conclusions.
KSKevin ShethNeurologist, Yale School of MedicineContextAug 30
Brain haemorrhage survivors must be enrolled in anticoagulation trials, especially in the US
Sheth emphasises that patients with prior brain haemorrhage have been excluded from all major anticoagulation trials and that it is critically important to enrol them in studies, particularly in the United States.
Distilled from their postOriginal post
Overview
Avoid edoxaban in AF patients with prior intracranial hemorrhage due to net harm; challenges assumptions favoring anticoagulation from indirect evidence.
Key Points
- To assess the efficacy and safety of edoxaban versus no anticoagulation in patients with high-risk atrial fibrillation and prior intracranial haemorrhage.
- Open-label, blinded-endpoint, event-driven randomized trial conducted across 174 sites in 20 countries with 948 participants (mean age 77 years; 39% women).
- Participants with high-risk atrial fibrillation and prior intracranial haemorrhage were randomized 1:1 to edoxaban 60 mg daily (adjusted to 30 mg) or no anticoagulation over a mean follow-up of 28 months.
- Edoxaban did not significantly reduce the primary composite endpoint of stroke or systemic embolism compared with no anticoagulation (11.8% vs 12.8%; HR 0.88; 95% CI 0.61–1.26; p=0.48).
- Reductions in ischaemic stroke and myocardial infarction were counterbalanced by an almost three-fold increase in haemorrhagic stroke, with ISTH major bleeding occurring in 11.6% with edoxaban versus 5.2% without anticoagulation (HR 2.23; 95% CI 1.39–3.59; p<0.001).
Evidence details
What drove the result?
| Outcome | Edoxaban | No OAC |
|---|---|---|
| Stroke or systemic embolism | 11.8% | 12.8% |
| HR 0.88 (95% CI 0.61-1.26); p=0.48 - fewer ischaemic strokes and MIs offset by ~3-fold more haemorrhagic strokes (counts not reported) | ||
Limitations & tradeoffs
Safety
ISTH major bleeding 11.6% vs 5.2%; HR 2.23 (95% CI 1.39-3.59); p<0.001.
Design limitations
open-label design (endpoints blinded), and enrollment of lobar intraparenchymal and convexity subarachnoid haemorrhage stopped after a 2023 DSMB review, so the final population is selected.
Statistical certainty
per-arm event counts are unavailable; results derive from the congress press release without a simultaneous publication.
Structured PICO
Does edoxaban reduce stroke or systemic embolism in patients with high-risk atrial fibrillation and prior intracranial haemorrhage?
P
Population
948 older adults (mean age 77, 39% women) with high-risk atrial fibrillation and prior intracranial haemorrhage, followed for a mean of 28 months.
I
Intervention
Edoxaban 60 mg daily (label-adjusted to 30 mg)
C
Comparator
No anticoagulation
O
Outcome
Stroke or systemic embolismcomposite
In patients with atrial fibrillation and prior intracranial hemorrhage, edoxaban did not reduce stroke or systemic embolism compared to no anticoagulation, but significantly increased major bleeding.
Main Result
Hazard Ratio: 0.88 (95% CI 0.61–1.26)
Absolute Event Rate: 11.8% vs 12.8%
p-value: p=0.48
Limitations
- No simultaneous results paper was available at presentation; source is the ESC Congress 2026 press release.
Coverage & sources
Journal, society, and media accounts. Useful signal, not independent expert judgment.
Cite This Study
Zhan Wang (2026) conducted an RCT in high-risk atrial fibrillation and prior intracranial haemorrhage (n=948). edoxaban vs. no anticoagulation was evaluated on stroke or systemic embolism (HR 0.88, 95% CI 0.61-1.26, p=0.48). Edoxaban did not significantly reduce stroke or systemic embolism versus no anticoagulation in AF patients with prior intracranial hemorrhage (11.8% vs 12.8%; HR 0.88; 95% CI 0.61-1.26; p=0.48).