Ventricular flow analysis and its association with exertional capacity in repaired tetralogy of Fallot: 4D flow cardiovascular magnetic resonance study

Abstract

Background: Four-dimensional (4D) flow cardiovascular magnetic resonance (CMR) allows quantification of biventricular blood flow by flow components and kinetic energy (KE) analyses. However, it remains unclear whether 4D flow parameters can predict cardiopulmonary exercise testing (CPET) as a clinical outcome in repaired tetralogy of Fallot (rTOF). Current study aimed to (1) compare 4D flow CMR parameters in rTOF with age- and gender-matched healthy controls, (2) investigate associations of 4D flow parameters with functional and volumetric right ventricular (RV) remodelling markers, and CPET outcome.

Methods: Sixty-three rTOF patients (14 paediatric, 49 adult; 30 ± 15 years; 29 M) and 63 age- and gender-matched healthy controls (14 paediatric, 49 adult; 31 ± 15 years) were prospectively recruited at four centers. All underwent cine and 4D flow CMR, and all adults performed standardized CPET same day or within one week of CMR. RV remodelling index was calculated as the ratio of RV to left ventricular (LV) end-diastolic volumes. Four flow components were analyzed: direct flow, retained inflow, delayed ejection flow and residual volume. Additionally, three phasic KE parameters normalized to end-diastolic volume (KEiEDV), were analyzed for both LV and RV: peak systolic, average systolic and peak E-wave.

Results: In comparisons of rTOF vs. healthy controls, median LV retained inflow (18% vs. 16%, P = 0.005) and median peak E-wave KEiEDV (34.9 µJ/ml vs. 29.2 µJ/ml, P = 0.006) were higher in rTOF; median RV direct flow was lower in rTOF (25% vs. 35%, P < 0.001); median RV delayed ejection flow (21% vs. 17%, P < 0.001) and residual volume (39% vs. 31%, P < 0.001) were both greater in rTOF. RV KEiEDV parameters were all higher in rTOF than healthy controls (all P < 0.001). On multivariate analysis, RV direct flow was an independent predictor of RV function and CPET outcome. RV direct flow and RV peak E-wave KEiEDV were independent predictors of RV remodelling index.

Conclusions: In this multi-scanner multicenter 4D flow CMR study, reduced RV direct flow was independently associated with RV dysfunction, remodelling and, to a lesser extent, exercise intolerance in rTOF patients. This supports its utility as an imaging parameter for monitoring disease progression and therapeutic response in rTOF. Clinical Trial Registration https://www.clinicaltrials.gov . Unique identifier: NCT03217240.

Keywords: 4D flow CMR; Cardiopulmonary exercise testing; Flow components; Kinetic energy; Repaired tetralogy of Fallot.

Conflict of interest statement

The authors declare that they have no competing interests.

© 2021. The Author(s).

Figures

Fig. 1

Flow chart. 4D four-dimensional, CMR cardiovascular magnetic resonance, CPET cardiopulmonary exercise testing, rTOF repaired tetralogy of Fallot

Fig. 2

Right ventricle (RV) direct flow (green) using particle tracing at peak systole, end-systole and peak early diastolic filling phases in a 30-year-old healthy subject (first row) and a 28-year-old repaired tetralogy of Fallot (rTOF) patient (second row) with respective RV direct flow of 39% and 19%. Yellow circles denote the RV contours from stacks of short axis views. RVOT RV outflow tract

Fig. 3

Differences in 4D flow right ventricular (RV) parameters according to RV function; RV remodelling; and peak oxygen uptake (VO2). RV direct flow (left), RV residual volume (middle) and peak E-wave KEiEDV (right) are presented (A) among healthy controls, rTOF with preserved RV function, and rTOF with reduced RV function; (B) among healthy controls, rTOF with preserved RV remodelling and rTOF with abnormal RV remodelling; (C) among healthy controls, rTOF with preserved peak VO2 and rTOF with abnormal peak VO2. *P < 0.05 compared with healthy controls; †P < 0.05 compared with rTOF with preserved RV function, and rTOF with preserved RV remodelling, respectively. Error bars denote median—25th percentile (lower) and 75th percentile—median (upper). KEiEDV kinetic energy normalized to end-diastolic volume (EDV), rTOF repaired tetralogy of Fallot

Fig. 4

Utility of right ventricular (RV) direct flow and RV ejection fraction (RVEF) to detect (A) moderate to severe RV remodelling (RVEDV/LVEDV ratio > 1.41) (B) exercise capacity with intermediate and high risks (% predicted peak VO2 < 65%). AUC area under ROC curve, EDV end-diastolic volume, ROC receiver operating characteristic, LV left ventricle, LVEDV left ventricular end-diastolic volume, RVEDV right ventricular end-diastolic volume, VO2 oxygen uptake

Fig. 5

Reproducibility of right ventricular (RV) 4D flow components. (A) Bland–Altman analysis of intra-observer repeated measurements of RV direct flow (first row), RV retained inflow (second row), RV delayed ejection flow (third row) and RV residual volume (last row); (B) Bland–Altman analysis of inter-observer repeated measurements of RV direct flow (first row), RV retained inflow (second row), RV delayed ejection flow (third row) and RV residual volume (last row)