Prediction of Left Ventricular Filling Pressures Using Natural Shear Waves: Incremental Value Over Conventional
Annegret Elisabeth Werner1, Stéphanie Bézy2, Laurine Wouters2
1Department of Cardiovascular Sciences, Catholic University Leuven, Belgium; Department of Cardiology, University of Rostock, Germany.
Background:
Increased myocardial stiffness can lead to diastolic dysfunction, resulting eventually in increased left ventricular filling pressures (LVFPs). The clinical assessment of left ventricular (LV) diastolic function is challenging. Shear wave elastography is a novel method based on tracking shear waves, such as those induced by mitral valve closure (MVC), using high frame rate (HFR) echocardiography. The propagation velocity of shear waves is directly related to myocardial stiffness.
Objectives:
This study investigated whether shear wave velocity at MVC is related to invasively measured LVFP.
Methods:
Eighty-five patients were prospectively enrolled. Left ventricular mid-diastolic pressure (LVMDP) and left ventricular end-diastolic pressure (LVEDP) were invasively measured. HFR (average frame rate: 1,050 ± 220 Hz) and conventional echocardiography were performed immediately after catheterization. Shear wave velocity at MVC was measured in the anteroseptal wall. The findings were compared with conventional echocardiographic parameters and the multiparametric 2021 guideline-recommended algorithm for the estimation of LVFP. The direct response of shear wave velocity at MVC on changing LV filling pressures was tested in a group of 16 patients with acutely decompensated heart failure (DHF) on admission and after diuretic therapy.
Results:
Single conventional echocardiographic parameters correlated moderately with LVFP (Pearson r between 0.39 and 0.58) and had a moderate ability to detect elevated LVFP (AUC between 0.66 and 0.79). This was similar for the multiparametric guideline algorithm (LVMDP AUC: 0.78; LVEDP AUC: 0.60). In comparison, shear wave velocities at MVC as a single measure correlated strongly with LVEDP (r = 0.71; P < 0.001) and had an similar ability to predict elevated LVMDP (AUC: 0.79; sensitivity: 87%; specificity: 65%) and an excellent ability to predict elevated LVEDP (AUC: 0.95; sensitivity: 92%; specificity: 94%) when a cutoff value of 4.8 m/s was used. In DHF patients, we observed a significant decrease in shear wave velocity after diuretic therapy (7.1 ± 1.7 m/s vs 5.3 ± 1.3 m/s; P < 0.002).
Conclusions:
In comparison with the multivariable guideline approach, end-diastolic shear wave velocity as a single measure had similar accuracy in detecting elevated LVMDP and a higher accuracy for LVEDP. Shear wave elastography might therefore be a promising new tool for the noninvasive assessment of diastolic function.
