Serial MRI-based right ventricular mechanical wall stress measurements and their association with right ventricle function in patients with repaired Tetralogy of Fallot

Abstract

Abstract Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): Stichting Hartekind en Thorax Foundation Background Optimal timing of pulmonary valve replacement (PVR) in Tetralogy of Fallot (TOF) patients remains challenging. Wall stress is considered to be a possible early marker of right ventricular (RV) dysfunction. With patient-specific computational models, wall stress can be determined regionally and with high accuracy, especially in complex shaped ventricles such as in TOF patients. We aimed to 1) develop patient-specific computational models to assess RV diastolic wall stresses and 2) investigate the association of wall stresses and their change over time with functional parameters in TOF patients. Methods Repaired TOF patients with at least moderate pulmonary regurgitation (PR) and prior to PVR were included. MRI-based patient-specific computational ventricular models were created (figure). The ventricular geometry was created by stacking endo- and epicardial contours traced on short axis SSFP cine images. Pressure in the right ventricle was estimated from echocardiography. Mid-diastolic wall stress in the RV free wall was analysed globally and regionally (basal, mid, apical, anterior, lateral and posterior) at two time points. RV ejection fraction (RVEF), NT-proBNP and exercise tests (% maximum predicted workload) were used as outcomes for RV function. Associations between wall stresses and outcomes were investigated using linear mixed models adjusted for follow-up duration. Results Five males and five females were included with an age at baseline of 24 (IQR 16-28) years and RV end-diastolic volume of 140 (IQR 127-144) ml/m2. The period between the two time points was 7.0 (IQR 5.8-7.3) years. Global wall stress of the RV free wall combining both time points was 5.8 kPa (IQR 5.2-7.2). There was no statistical difference between baseline and follow-up global wall stress. The mean wall stresses in the mid region was 1.69 kPa (p < 0.01) higher than in the basal region and was 1.05 kPa (p = 0.03) higher than in the apical region cross-sectionally. The wall stress also increased more in the mid region compared to basal and apical region, corrected for duration of follow-up. Patients with more severe PR at baseline demonstrated a higher increase of global wall stress over time (p = 0.02), especially in lateral free wall. Higher global free wall stresses were cross-sectionally independently associated with lower RVEF, adjusted for LVEF and RVEDV (β=-1.29 % RVEF per kPa increase in wall stress, p = 0.01). This association was most prominent in the anterior, basal and mid part. No statistically significant association was found between wall stress, NT-proBNP, and exercise capacity. Conclusions This study generated a novel MRI-based method to calculate wall stress in geometrically complex ventricles. Wall stress associated negatively with RVEF in patients with TOF and PR. This promising tool for RV wall stress analysis can be used in future larger studies to validate these preliminary findings and to assess the predictive value of wall stress in TOF. Abstract Figure.