Atrial strain as an imaging biomarker of subclinical atrial fibrillation in embolic stroke of undetermined source

Abstract

This study aims to evaluate left atrial (LA) strain as an imaging biomarker for predicting subclinical atrial fibrillation (AF) in embolic stroke of undetermined source (ESUS) patients undergoing long-term rhythm monitoring with implantable loop recorders (ILR). Retrospectively study of 67 ESUS patients who underwent ILR implantation. LA strain was assessed using speckle-tracking echocardiography (STE). Global peak atrial longitudinal strain (PALS) and peak atrial contraction strain (PACS) were quantified. Intra- and interobserver reproducibility were evaluated. Subclinical AF was considered stroke-related if detected within 12 months. Statistical analyses included regression models, ROC analysis, Kaplan–Meier survival curves, and Cox proportional hazards models. Subclinical AF was detected in 16 patients (23.9%). Patients who developed AF exhibited significantly reduced global PALS. PALS demonstrated excellent intra- and interobserver reproducibility (ICC >0.9). ROC analysis identified a global PALS cutoff <25.50% as a sensitive predictor of subclinical AF. In multivariable analysis, reduced PALS and lower left atrial volume index (LAVI) were independent predictors of AF. Time-to-event analysis confirmed shorter AF-free survival in patients with impaired PALS. Quantitative assessment of LA mechanical function using STE identifies early atrial dysfunction associated with subclinical AF in ESUS patients. Reduced PALS appears to precede atrial dilation, supporting its role as a functional imaging biomarker of atrial cardiomyopathy.