Clinical and biomarker determinants for recurrent stroke in patients with atrial fibrillation: a systematic review and meta-analysis

<p dir="ltr"><b>Background and objectives:</b><b> </b>Despite effective secondary prevention, including oral anticoagulant (OAC) therapy, the risk of recurrent stroke (RS) in patients with atrial fibrillation (AF) remains substantial with an annualized risk of 3.2%-6.5% per year. The reasons for this high residual risk are unclear. There is growing need for improved risk prediction tools to identify patients at greatest risk of RS in AF and to find new therapeutic targets for secondary prevention. Our objective was to perform a systematic review to investigate the association of clinical factors and echocardiographic, blood, and neuroimaging biomarkers, with stroke recurrence after AF-related stroke. <p dir="ltr"><b>Methods:</b><b> </b>We searched Embase/Ovid Medline until August 2023. Studies were included irrespective of OAC use. Risk ratios (RRs) were pooled using random effects. Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed. <p dir="ltr"><b>Results:</b><b> </b>Of 5,427 records searched, 42 reports from 28 studies including 52,798 patients (5,046 RS events during follow-up) were identified. In addition to the CHA₂DS₂-Vasc score (RR 1.22, 95% CI 1.15-1.30, per point) and its individual components, the clinical factors associated with recurrence were a qualifying stroke despite OAC (RR 1.55, 1.23-1.94 [vs OAC-naïve]), sustained AF (RR 1.44, 1.19-1.75 [vs paroxysmal]), hyperlipidemia (RR 1.27, 1.05-1.53), chronic kidney disease (RR 1.86, 1.19-2.92), and malignancy (RR 4.36, 1.85-10.78). NIH Stroke Scale scores (RR 0.97, 0.95-0.99) and Asian ethnicity (RR 0.59, 0.36-0.97) were associated with a reduced risk of recurrence. Known AF was not associated with a higher risk of stroke compared with AF detected after stroke (RR 1.20, 0.93-1.55). Neuroimaging markers associated with RS included chronic lacunar (RR 1.91, 1.29-2.84) or embolic-appearing (RR 2.76, 1.32-5.77) infarcts and cerebral microbleeds (RR 1.29, 1.04-1.59). Echocardiographic markers included atrial size (RR 1.40, 1.12-1.75 per cm/m²), intracardiac thrombus (RR 1.99, 1.38-2.87), spontaneous left atrial appendage (LAA) echocardiographic contrast (RR 2.91, 1.21-6.17), or low LAA intensity variation (RR 2.81, 1.48-5.32). Data were limited for blood biomarkers. <p dir="ltr"><b>Discussion:</b><b> </b>We identified several clinical factors associated with recurrence after AF-related stroke, with AF burden and stroke despite OAC of particular clinical relevance. Biomarkers of atrial cardiopathy, small vessel disease, or previous infarction were also associated with RS. Collaborative efforts are needed to identify and validate new risk factors and biomarkers of RS in AF.