Does oral anticoagulation reduce ischaemic major adverse cardiovascular events without increasing recurrent intracranial haemorrhage in intracranial haemorrhage survivors with atrial fibrillation?
While phase II trials suggest oral anticoagulation reduces ischaemic events in atrial fibrillation patients with prior intracranial haemorrhage, ongoing large randomized trials are needed to definitively establish its safety regarding recurrent haemorrhage.
PURPOSE: To review the current evidence on anticoagulation for intracranial haemorrhage (ICrH) survivors with atrial fibrillation (AF). METHOD: Narrative review of the literature. FINDINGS: AF and ICrH are age-related conditions whose prevalence and comorbidity is expected to increase with the ageing population. Patients with ICrH were excluded from pivotal randomized trials of anticoagulation in AF and guidelines do not provide strong recommendations on if and when to (re)initiate anticoagulation in patients with AF and ICrH. Pooled analyses of phase II randomized trials have reported reduced risk of ischaemic major adverse cardiovascular events with anticoagulation in this population, but there remains uncertainty regarding the effects of anticoagulation on recurrent ICrH and death, as well as potential heterogeneity of treatment effect in higher risk subgroups, such as patients with cerebral amyloid angiopathy. There are no reported randomized trials investigating the optimal timing of anticoagulation (re)initiation in ICrH survivors with AF and the findings from observational studies have been inconsistent. CONCLUSION: Whether or not OAC should be resumed in ICrH survivors with AF and the optimal timing of OAC (re)initiation are challenging clinical dilemmas that are becoming more frequent with our ageing population. The existing prevalence of AF in patients with ICrH and changing global demographics highlight the importance of ongoing and future randomized trials addressing unresolved questions surrounding optimal stroke prevention strategies in this vulnerable patient population.
Shoamanesh et al. (Tue,) studied this question.