Characteristics of altered biventricular hemodynamics after arterial switch operation for patients with d-transposition of the great arteries with preserved ejection fraction: a four-dimensional (4D) flow cardiovascular magnetic resonance (CMR) study.

Abstract

The long-term monitoring of biventricular function is essential to identify potential functional decline in patients following the arterial switch operation (ASO). The underlying pathophysiological mechanisms responsible for altered biventricular hemodynamics in ASO patients are not yet well understood. This study sought to: (I) compare the biventricular kinetic energy (KE) and vorticity of ASO patients and age- and sex-matched controls; (II) investigate the associations of four-dimensional (4D) flow biventricular hemodynamics parameters and neo-aortic root dilation, supravalvular pulmonary stenosis, and pulmonary artery transvalvular pressure difference. A total of 34 patients with dextro-transposition of the great arteries (D-TGA) who underwent ASO, and 17 age- and gender-matched healthy controls were prospectively recruited for this study. All the subjects underwent cine and 4D flow and late gadolinium enhancement scans, and all the patients underwent echocardiography within two weeks of cardiovascular magnetic resonance (CMR) imaging. The following four flow components were analyzed: direct flow, retained inflow, delayed ejection flow, and residual volume. In addition, the following six phasic blood flow KE parameters, normalized to the end-diastolic volume (EDV) and vorticity, were analyzed for both the left ventricle (LV) and right ventricle (RV): peak systolic phase, average systolic phase, peak diastolic phase, average diastolic phase, peak E-wave phase, and peak A-wave phase. The independent sample Student's t-test, Mann-Whitney U-test, univariable and multivariable stepwise regression analyses, intra and inter-observer variability analyses were used to compare patients and controls. In relation to the LV, the D-TGA patients had significantly decreased average vorticity, peak systolic vorticity, systolic vorticity, diastolic vorticity, and peak A-wave vorticity compared to the controls (all P<0.01). In relation to the RV, the pulmonary stenosis group had significantly increased peak E- and A-wave kinetic energy normalized to the end-diastolic volume (KEiEDV), and peak and average vorticity compared to the non-pulmonary stenosis group (all P<0.05). in the multivariable logistic regression model analysis, diastolic KEiEDV, peak E-wave KEiEDV peak A-wave KEiEDV, and average vorticity were associated a with transvalvular pressure difference (β=13.54, P<0.001 for diastolic KEiEDV; β=105.26, P<0.001 for peak E-wave KEiEDV; β=-49.36, P=0.027 for peak A-wave KEiEDV; and β=-56.37, P<0.001 for average vorticity). We found that 4D flow biventricular hemodynamics were more sensitive markers than the ejection fraction in the postoperative D-TGA patients. The RV diastolic KEiEDV parameters and average vorticity were risk factors for pulmonary artery obstruction in the multivariable model.