Association of CMR-Derived Strain with Risk Stratification in Patients with Tetralogy of Fallot (ToF)

Background: Tetralogy of Fallot (ToF) is the most common cyanotic congenital heart disease, typically repaired in childhood with the ventricular septal defect (VSD) closure and relief of right ventricular outflow tract obstruction. Although outcomes have improved, patients remain at risk for complications such as arrhythmias, pulmonary valve dysfunction, and heart failure. Cardiac magnetic resonance (CMR) imaging allows reproducible assessment of ventricular volumes and strain, providing insight into subclinical dysfunction. This study evaluated CMR-derived strain in relation to clinical risk status to support predictive modeling for high-risk cardiovascular patients with ToF.

Methods: We included 45 patients with Tetralogy of Fallot (ToF) who underwent cardiac MRI after 2020. Images were analyzed using Circle Cardiovascular Imaging (cvi42). Left ventricular (LV) endocardial and epicardial contours were traced from the short-axis, 2-chamber, 3-chamber, and 4-chamber views. Right atrial (RA) contours were delineated from the 4-chamber view, left atrial contours from the 2- and 4-chamber views, and right ventricular (RV) contours from the short-axis and 4-chamber views. Global circumferential, radial, and longitudinal strain values were derived from these tracings. An importance factor (IF) was calculated from clinical history (age at repair, postoperative cardiac/respiratory symptoms, history of Blalock–Taussig shunts, and repair type). Patients with IF >2 were classified as high-risk. Strain values and chamber volumes were compared to risk status using ROC curves. A multivariable logistic regression model including LV longitudinal strain and minimum RA volume index was constructed to predict risk status.

Results: LV longitudinal strain showed a trend toward association with high-risk status (p = 0.068) (A). Minimum RA volume index was significantly associated with high-risk status (p < 0.001) (B). The predicted probability from the logistic regression model demonstrated superior discrimination compared to either variable alone (AUC = 0.919) (C).

Conclusion: CMR-derived functional and volumetric measures can improve risk stratification in patients with ToF. Combining LV longitudinal strain with RA volume index enhanced predictive performance and may provide a clinically useful tool for identifying high-risk patients.

Strain Parameters and Predictive Performance

Parameter	AUC	P-Value
LA Global Longitudinal Strain	0.559	0.623
LA Global Radial Strain	0.576	0.543
LV Global Longitudinal Strain	0.715	0.068
LV Global Radial Strain	0.720	0.064
LV Global Circumferential Strain	0.645	0.230
RA Global Longitudinal Strain	0.527	0.833
RA Global Radial Strain	0.576	0.577
RV Global Longitudinal Strain	0.559	0.524
RV Global Radial Strain	0.652	0.171

Figure 1. Receiver operating characteristic (ROC) curves of left ventricle (LV) longitudinal strain, minimum right atrial (RA) volume (indexed), and logistic regression predicted probability of LV longitudinal strain and minimum RA volumes (indexed) in predicting high-risk patients (importance factor score >2).