Left ventricular flow dynamics and energetics derived from 4D flow CMR in patients with low gradient severe aortic stenosis

Abstract Background Up to 40% of patients with severe aortic stenosis (SAS) determined by current guidelines present with discordant Doppler echocardiography findings and demonstrate divergent prognosis. Differentiation of true SAS who generally benefits from aortic valve intervention among this subset of patients is important albeit challenging for clinical management. Purpose Left ventricular (LV) 3-dimentional flow characterization was analyzed to illustrate altered intracardiac hemodynamics in different types of low gradient SAS and to investigate its application for the identification of true SAS in comparison with LV stroke volume index (LVSVI) and transaortic flow rate. Methods Four-dimensional flow cardiovascular magnetic resonance (4D flow CMR) was performed in 126 patients with low gradient SAS and 30 age- and gender-matched controls. All patients were categorized into 3 groups according to left ventricular ejection fraction (LVEF) and LVSVI as follows: classical low-flow low-gradient (CLFLG, n=37), paradoxical low-flow low-gradient (PLFLG, n=41), and normal-flow low-gradient (NFLG, n=48). Left ventricular flow was partitioned into 4 components: direct flow, retained inflow, delayed ejection flow, and residual volume. Average kinetic energetic parameters indexed to end-diastolic volume (KEiEDV) throughout the systolic phase were analyzed. Results Patients with CLFLG and PLFLG SAS had reduced LV direct flow proportion (CLFLG vs. control: 10±5% vs. 35±5%, P<0.001; PLFLG vs. control: 29±6% vs. 35±5%, P<0.001, Figure 1) and systolic ejection-flow KEiEDV (CLFLG vs. control: median, 5.3μJ/ml vs 8.7μJ/ml, P<0.001; PLFLG vs. control: median, 6.0μJ/ml vs. 8.7μJ/ml) compared with controls. Reduced systolic ejection-flow KEiEDV correlated with decreased LVSVI (R=0.47, P<0.001) and transaortic flow rate (R=0.72, P<0.001). Among 126 patients in the study cohort, 74 (59%) were determined as true SAS according to standardized criteria based on flow-independent parameters and demonstrated elevated systolic ejection-flow KEiEDV than patients with pseudo-SAS (median, 6.5μJ/ml vs 5.6μJ/ml, P<0.001). Systolic ejection-flow KEiEDV≥6.1μJ/ml demonstrated an optimal positive predictive value of 90% for distinguishing true SAS from pseudo-SAS among patients with low gradient SAS (Figure 2). Conclusion Alterations in LV flow components and kinetic energy were associated with the presence and type of low gradient SAS. LV systolic outflow energetic markers responsible for aortic valve opening could aid the evaluation of AS severity with Doppler-based parameters.Comparison of LV flow componentsROC analysis for discriminating true SAS