This poll included a clinical vignette of an 80-year-old man with atrial fibrillation (AF), high thromboembolic risk, and multiple cardiovascular (CV) comorbidities who presented 2 months after a spontaneous lobar intracerebral hemorrhage while receiving reduced-dose apixaban, raising the question of whether anticoagulation should be restarted.

The aim of the poll was to raise awareness of the complex clinical decision of restarting anticoagulation after one of the most devastating bleeding complications. It also highlighted how evidence-based practice may sometimes challenge our initial biases, particularly when the feared complication is recurrent intracranial bleeding. The poll results were both thought-provoking and clinically relevant, reflecting the uncertainty and nuance that cardiologists frequently face in this scenario.

Currently, >6 million people in the United States are treated with oral anticoagulants (OAC), but bleeding is the most frequent and serious complication. Although prevention is fundamental, bleeding is inevitable. One of the most complex clinical decisions involves determining when, or whether, anticoagulation can be safely resumed. For patients with AF and acute ischemic stroke, there are current recommendations depending on the stroke size.¹ In cases of ICH, decision-making is more nuanced.

Balancing thromboembolic and bleeding risk in survivors of ICH with AF is a highly complex clinical decision. The potential benefits of stroke and systemic embolism prevention must be weighed against the risk of recurrent, potentially devastating intracranial bleeding. Approximately 20% of spontaneous ICHs are related to anticoagulation, with a 30-day mortality of 50%, so every cardiologist will face this dilemma repeatedly in practice. Not all ICHs are alike, and important factors to consider include mechanism (spontaneous vs. traumatic), location (lobar bleeds, often amyloid related, carry the highest recurrence risk), presence of cerebral microbleeds, and comorbidities such as uncontrolled HTN or thrombocytopenia.

According to the 2020 American College of Cardiology (ACC) Expert Consensus Decision Pathway on Management of Bleeding in Patients on OAC,² patients with high thrombotic risk, such as this one with a high CHA₂DS₂-VASc score, reinitiation of anticoagulation is favored once hemostasis is achieved and the patient is clinically stable. Nevertheless, after severe bleeds such as ICH, the bleeding risk is also high, and decisions should be individualized with a multidisciplinary approach that incorporates patient preferences. To guide this decision, it is essential to examine the available evidence, including findings from recent randomized trials and meta-analyses, which provide critical insights into risks, benefits, and timing of anticoagulation resumption.

The recently published PRESTIGE-AF (PREvention of STroke in Intracerebral haemorrhaGE Survivors With Atrial Fibrillation) trial results (n = 319; median age 79 years) showed that DOACs effectively prevent ischemic stroke in patients with AF who survive ICH, but this benefit is partially offset by an increased risk of recurrent hemorrhage. Ischemic stroke events occurred at 0.83 per 100 patient-years with DOACs and at 8.6 with no anticoagulation, whereas recurrent ICH occurred at 5 per 100 patient-years with DOACs and at 0.82 with no anticoagulation. Net clinical benefit (stroke/systemic embolism, MI, CV death, major bleeding) favored DOACs slightly (hazard ratio, 0.69; 95% confidence interval, 0.33-1.4) with the event rate in the DOAC group being slightly lower than in the no-anticoagulant group (19.2 per 100 patient-years with DOACs vs. 26.52 with no anticoagulation). Enrollment occurred between 14 days and 12 months after the index bleed; the median time to enrollment was 49 days (interquartile range, 31-93).³ The results of this study showed a particular benefit of DOACs for prevention of ischemic strokes in the first year after the index event, whereas the risk of ICH recurrence in the DOAC group appeared to continue thereafter.

After this publication, a meta-analysis of four randomized controlled trials (RCTs) in survivors of ICH with AF (n = 653; mean age 78.2 years) found that resuming OAC, predominantly with DOACs, was associated with a number needed to treat (NNT) of 8 to prevent one ischemic event and with a number needed to harm (NNH) of 22 for causing one recurrent ICH, indicating a favorable net clinical benefit in most patients.⁴ However, the authors acknowledged important trade-offs. The NNT/NNH estimates assumed uniform outcome severity, which may oversimplify the true clinical impact because recurrent ICH and ischemic strokes can differ in functional consequences. Moreover, the lack of individual patient data precluded time-to-first-event analyses that might clarify the temporal relationship between anticoagulation resumption and adverse events.

Left atrial appendage occlusion (LAAO) is a viable alternative and has shown acceptable safety in RCT and registry data.⁵ However, evidence specific to survivors of ICH with AF is lacking. Three trials are currently enrolling to address this gap: 1) A3ICH (Avoiding Anticoagulation After IntraCerebral Haemorrhage), a 1:1:1 comparison of apixaban, LAAO, and no anticoagulation; 2) STROKECLOSE (Prevention of Stroke by Left Atrial Appendage Closure in Atrial Fibrillation Patients After Intracerebral Hemorrhage), a 2:1 comparison of LAAO and no anticoagulation; and 3) CLEARANCE (Comparison of LAA-Closure vs Oral Anticoagulation in Patients With NVAF and Status Post Intracranial Bleeding), a comparison of LAAO and DOAC.

Age deserves particular focus. The median age in recent studies is near 80 years, much like this patient, and this group carries a higher prevalence of geriatric syndromes (polypharmacy, frailty, sarcopenia, frequent falls, cognitive decline). These factors are underreported in trials and not captured by most risk scores, yet they clearly influence outcomes in AF. Any recommendation to restart anticoagulation should be paired with bleeding-prevention strategies: reinforcing adherence, deprescribing nonessential drugs, reviewing drug–drug interactions, and involving family/caregivers.⁶

Taken together, the evidence supports apixaban being restarted today for this patient. He was clinically stable, and RCT and meta-analysis findings suggest a modest net clinical benefit to resuming anticoagulation. His follow-up visit (~2 months after the index ICH) fit the timing used in trials. His age aligned with the studied populations. Although neurology input is valuable within multidisciplinary care, a decision to restart could be made today if it aligned with his preferences.

References

1. Fischer U, Koga M, Strbian D, et al. Early versus later anticoagulation for stroke with atrial fibrillation. N Engl J Med. 2023;388(26):2411-2421. doi:10.1056/NEJMoa2303048
2. Tomaselli GF, Mahaffey KW, Cuker A, et al. 2020 ACC expert consensus decision pathway on management of bleeding in patients on oral anticoagulants: a report of the American College of Cardiology Solution Set Oversight Committee. J Am Coll Cardiol. 2020;76(5):594-622. doi:10.1016/j.jacc.2020.04.053
3. Veltkamp R, Korompoki E, Harvey KH, et al. Direct oral anticoagulants versus no anticoagulation for the prevention of stroke in survivors of intracerebral haemorrhage with atrial fibrillation (PRESTIGE-AF): a multicentre, open-label, randomised, phase 3 trial. Lancet. 2025;405(10482):927-936. doi:10.1016/S0140-6736(25)00333-2
4. Chi KY, Chang Y, Lee PL, et al. Anticoagulants for atrial fibrillation in survivors of intracranial hemorrhage: a meta-analysis of randomized trials. J Am Coll Cardiol. 2025;86(4):287-290. doi:10.1016/j.jacc.2025.04.069
5. Alli OO, Garg J, Boursiquot BC, et al. Racial and ethnic disparities in the use and outcomes with WATCHMAN FLX: a SURPASS analysis of the NCDR Left Atrial Appendage Occlusion Registry. J Am Heart Assoc. 2024;13(23):e036406. doi:10.1161/JAHA.124.036406
6. Schmanske N, Ngo JM, Kalra K, Nanna MG, Damluji AA. Healthy ageing in older adults with cardiovascular disease. Eur Heart J. 2025;46(26):2536-2551. doi:10.1093/eurheartj/ehaf231