Prognostic Value of Left Atrial Strain for Atrial Fibrillation After Cryptogenic Stroke and Embolic Stroke of Undetermined Source

Le résumé fourni par la source

Subclinical atrial fibrillation (AF) is increasingly recognized as an important underlying mechanism of cryptogenic stroke (CS) and embolic stroke of undetermined source (ESUS). Early identification of patients at increased risk of AF remains a major clinical challenge. Assessment of left atrial strain (LAS) by transthoracic echocardiography has emerged as a promising method for evaluating atrial function and may provide incremental information for AF risk stratification. We performed a systematic review in accordance with Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines to evaluate the association between LAS and AF detection in patients with CS and ESUS. PubMed, Embase, the Cochrane Library, and Google Scholar were searched up to December 7, 2025. Thirteen studies, including 2132 patients, were eligible for qualitative synthesis. Random-effects meta-analyses using the inverse-variance method were conducted where appropriate. Eleven studies reporting LAS as mean values with standard deviation were included in a comparative meta-analysis, comprising 525 patients with AF and 1185 without AF. LAS was significantly lower in patients with AF (standardized mean difference -0.80; 95% confidence interval, -1.09 to -0.50; P < 0.001), although heterogeneity was substantial (I2 = 83%). In an exploratory prognostic meta-analysis, 3 studies reporting multivariable-adjusted odds ratios from logistic regression models were pooled. Lower LAS was associated with AF detection (pooled odds ratio 0.87 per 1% increase; 95% confidence interval, 0.79-0.97; P = 0.009), with considerable heterogeneity across studies (I2 = 86.6%). Overall, LAS is generally lower in patients with AF and may support risk stratification for AF detection in patients with CS and ESUS; however, substantial heterogeneity and methodological differences across studies should be acknowledged.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.