Left atrial sphericity and strain augment risk prediction in patients with embolic stroke of undetermined source

European Heart Journal - Cardiovascular Imaging

16 March 2026
Organised by: Logo
ESC Journals DISEASES OF THE AORTA, PERIPHERAL VASCULAR DISEASE, STROKE Stroke IMAGING Echocardiography PREVENTIVE CARDIOLOGY Risk Factors and Prevention

Abstract

AbstractAims

Recurrent-stroke and new onset atrial fibrillation (AF) are prevalent following Embolic Stroke of Undetermined Source (ESUS). Various left atrial (LA) parameters are altered in ESUS patients vs. those with non-cardioembolic stroke (non-CES); these parameters may help risk-stratify ESUS patients for recurrent-stroke and new-AF. We comprehensively evaluated LA parameters in ESUS patients compared to non-CES patients. We further determined LA parameters associated with recurrent-stroke and new-AF in ESUS patients.

Methods and results

We prospectively recruited ischaemic stroke patients (ESUS (n = 203), non-CES (n = 256)). Comprehensive echocardiographic evaluation included LA volume, function (strain) and sphericity (circularity). ESUS patients were followed for recurrent-stroke and new-AF. LA parameters were assessed for association with outcomes. ESUS patients had increased LA volume, reduced reservoir and contractile strain, and increased circularity compared to non-CES patients (P < 0.002 for all). 36(18%) of ESUS patients developed recurrent-stroke, and 54(27%) recurrent-stroke or new-AF. Larger LA volume, reduced phasic LA strain (reservoir, contractile and conduit), and increased circularity (P < 0.05 for all) were associated with both outcomes in ESUS patients. Independent and incremental value was determined for reservoir and conduit strain, and circularity, in identifying patients at risk of recurrent-stroke and new-AF.

Conclusion

LA reservoir and conduit strain, and circularity have independent and incremental association with the occurrence of recurrent-stroke and new-AF in ESUS patients. This may represent measurable non-invasive markers of atrial cardiomyopathy in a subset of ESUS patients and enable risk-stratification and targeted intervention. Future longitudinal studies are needed to confirm these findings.

ESC 365 is supported by