Association of non-invasive atrial cardiomyopathy markers with cerebral stroke lesions: a population-based analysis from the Hamburg City Health Study

EP Europace Journal

27 March 2026
Organised by: Logo
ESC Journals ARRHYTHMIAS AND DEVICE THERAPY DISEASES OF THE AORTA, PERIPHERAL VASCULAR DISEASE, STROKE Stroke Atrial Fibrillation (AF) IMAGING Echocardiography PREVENTIVE CARDIOLOGY Risk Factors and Prevention

Abstract

AbstractAims

Atrial fibrillation (AF) is a well-known risk factor for ischaemic stroke. Emerging evidence suggests that atrial cardiomyopathy (AtCM), independent of AF, may be a key contributor to stroke risk. We aimed to evaluate whether non-invasive AtCM markers assessed by electrocardiography (ECG), transthoracic echocardiography (TTE), and blood-based biomarkers provide incremental diagnostic value beyond established clinical risk factors for identifying cerebral stroke lesions on magnetic resonance imaging (MRI).

Methods and results

We analysed 1794 Hamburg City Health Study participants who underwent cerebral MRI with available baseline 12-lead ECG and TTE in sinus rhythm. Logistic regression analyses were performed to identify associations between non-invasive AtCM markers and stroke lesions. The incremental discriminatory performance of these markers beyond clinical risk factors was assessed. Stroke lesions were present in 152 participants (8.5%). Male sex, history of prior AF, and higher CHA2DS2-VA score were significantly associated with stroke lesions. Among AtCM markers, amplified P-wave duration (APWD), P-wave area in lead II, PR interval, left atrial volume index, left atrial ejection fraction, and NT-proBNP were also significant. Combining both clinical and AtCM markers, only CHA2DS2-VA score (OR: 1.95 per point, 95% CI: 1.49–2.55, P < 0.001) and P-wave area in lead II (OR: 0.99 per 100 µV·ms, 95% CI: 0.98–1.00, P = 0.031) remained independent predictors. However, this resulted in only marginal improvement in discrimination (ΔAUC: 0.03, 95% CI: 0.0001–0.0594) compared with the clinical risk factor model alone.

Conclusion

The incremental diagnostic value of AtCM markers beyond clinical risk factors for MRI-defined cerebral lesions is limited, likely reflecting the heterogeneous aetiology of stroke lesions in a general population cohort.

Contributors

Silvia Becker
Silvia Becker

Author

Karlsruhe Institute of Technology (KIT) Karlsruhe , Germany

Dirk Westermann
Dirk Westermann

Author

University Heart Center Freiburg-Bad Krozingen Freiburg , Germany

Axel Loewe
Axel Loewe

Author

Karlsruhe Institute of Technology (KIT) Karlsruhe , Germany

Daniel Engler
Daniel Engler

Author

The University Medical Center Hamburg-Eppendorf Hamburg , Germany

Renate B Schnabel
Renate B Schnabel

Author

University Heart and Vascular Centre Hamburg (UHZ) Hamburg , Germany

ESC 365 is supported by