Diagnostic accuracy of speckle tracking echocardiography in transfusion-dependent beta thalassemia with atrial fibrillation
Abstract
Abstract Background Transfusion dependent β-thalassemia (TDβT) is a genetic blood disorder characterized by absent or reduced production of the β-chains of hemoglobin. Atrial fibrillation (AF) is highly prevalent in this ohort of patients. Speckle tracking echocardiography (STE) provides detailed information about left ventricular and atrial function, however its role in TDβT patients with AF is still unexplored. Purpose this study aimed to assess differences in cardiac mechanical parameters between thalassemia patients with and without atrial fibrillation. Methods β-THAL ("Atrial fibrillation in β-thalassemia") is a prospective, single-center, observational study aimed to assess the prevalence of clinical, laboratory, electrocardiographic and imaging characteristics in patients with TDβT. For the present cross-sectional analysis, a complete echocardiography evaluation comprising global longitudinal strain (GLS), peak atrial longitudinal strain (PALS) and indices of myocardial work (MW) was performed. The primary endpoint was the difference in STE indices between TDβT patients with or without atrial fibrillation. Uni and multivariate logistic regression model was used to assess the association between echocardiographic characteristics and the presence of atrial fibrillation. A p-value < 0.05 was considered significant. Results Between August 2022 and January 2025, 229 TDβT patients were enrolled. 32 patients had atrial fibrillation. Patients with AF were older (53 vs 50 years, p = 0.019) and had a higher rate of pulmonary hypertension (8 vs 3 patients, p < 0.001). Regarding echocardiographic findings, patients with AF had higher indexed values of left atrial volume and right atrial volume (43 vs 29 ml/m2, p < 0.001 and 33 vs 24 ml/m2, p < 0.001 respectively), lower values of left ventricular ejection fraction (59 vs 62%, p = 0.006), higher values of pulmonary artery systolic pressure (31 vs 25 mmHg, p < 0.001) and lower values of GLS and PALS (-19 vs -21%, p = 0.003 and 22 vs 35%, p < 0.001 respectively). No difference was found in MW parameters. In the multivariate logistic regression analysis lower values of GLS (OR 1.27, 95% CI 1.04-1.54, p = 0.016) and reduced values of PALS (OR 0.92, 95% CI 0.87-0.96, p = 0.001) were independently associated with the presence of atrial fibrillation. ROC curve analysis demonstrated that PALS had a good discriminatory ability for atrial fibrillation (AUC 0.79, 95% CI 0.71-0.87, best cut off 25.95%), outperforming GLS alone (AUC 0.65, 95% CI 0.55-0.77, best cut off -18.95%). The combined model including both GLS and PALS showed the highest diagnostic accuracy (AUC 0.80, 95% CI 0.72-0.87). Conclusions In TDβT patients with atrial fibrillation, GLS and PALS were lower and independently associated with the presence of the arrhythmia. Together, GLS and PALS showed the highest diagnostic accuracy and could represent powerful tools to early identify this cohort of patients.GLS and PALS analysis ROC curve analysis
Fields of Study & Subjects
Hemoglobinopathies and Related Disorders
Cardiovascular Function and Risk Factors
Cardiac Valve Diseases and Treatments
Publication details
- Journal / source
- European Heart Journal - Cardiovascular Imaging
- Volume
- 27
- Issue
- Supplement_1
- Publication date
- 1 January 2026
- Language
- EN