The co-occurrence of hemorrhagic-prone vasculopathies and ischemic-prone conditions requiring anticoagulation, such as atrial fibrillation (AF), constitute an increasingly common and recurrent clinical dilemma, particularly in aged populations.1,2 On one hand, this dilemma encloses the challenge of diagnosing and stratifying the hemorrhagic vasculopathy and, on the other hand, the decision to treat such patients with anticoagulants. The 2 most frequent hemorrhagic-prone vasculopathies are: (1) a deep perforator arteriolosclerosis that supplies the subcortical structures (sometimes referred to as hypertensive arteriopathy); and (2) cerebral amyloid angiopathy (CAA), which results from the progressive accumulation of β-amyloid in leptomeningeal and cortical vessels.3 Identifying such vasculopathies in asymptomatic patients largely depends on indirect imaging findings in MRI blood-sensitive sequences (T2* or susceptibility-weighted imaging) that can disclose cortical or deep intraparenchymal microbleeds (MBs) or cortical superficial siderosis (cSS). Strictly lobar MBs and cSS favor the diagnosis of CAA at least in intracerebral hemorrhage (ICH) or dementia cohort studies; and deep or mixed MBs favor a deep perforator small vessel disease.4,5 Meta-analysis data from prospective studies exploring the associations of MBs with the risk of stroke and future ICH have provided interesting information: In slightly more than 3,000 patients with ischemic stroke (IS) or TIAs, patients with MBs presented an increased risk of recurrent stroke (odds ratio [OR] = 2.25, 95% confidence interval [CI] 1.70–2.98) and an even higher OR for the risk of ICH (OR = 8.52, 95% CI 4.23–17.18), but only a minority of patients (<200) were treated with anticoagulants.6 In a more recent meta-analysis of 1,500 patients with a recent IS and documented AF, under long-term oral anticoagulation treatment, the authors found that patients with MBs presented an increased risk of future ICH (OR = 2.68, 95% CI 1.19–6.01) and that risk was particularly elevated (OR = 5.50, 95% CI 2.07–14.66) in patients with more than 5 MBs.7 Still, meta-analyses have limitations, as they compile small datasets that are not designed to answer a particular question such as the one raised here. For these reasons, there was an urgent need for solid, prospective, real-world data on whether MBs or other markers of hemorrhagic vasculopathies influence the risk of future ICH and IS, and how such information can help to stratify patients with AF for treatment with anticoagulants.

Source link