Abstract Background The clinical assessment of left ventricular diastolic function is complex, as there is no single non-invasive parameter that provides a direct measurement of myocardial relaxation, myocardial compliance, or–as a surrogate-LV filling pressure. The estimation of diastolic function involves therefore a combination of various parameters. Shear wave (SW) elastography (SWE) is a novel method based on high frame rate echocardiography. SWs are generated following mechanical excitation of the myocardium, such as after mitral valve closure (MVC), and their propagation velocity is directly linked to myocardial stiffness (MS) and is therefore a potential marker of diastolic function. Purpose The aim of this study was to investigate if the propagation velocities of natural shear waves are related to invasively measured mid-(LVMDP) and end-diastolic LV filling pressures (LVEDP) and, thus, could be used as echocardiographic estimate of left ventricular diastolic function. Methods We prospectively enrolled 95 patients with a wide range of diastolic function, scheduled for heart catheterization so that LVMDP and LVEDP could be invasively measured. Patients with myocardial pathology or dysfunction of the anteroseptal wall, as well as severe aortic stenosis, and a more than moderate mitral regurgitation were excluded. Echocardiography was performed immediately after catheterization. SW elastography in parasternal long axis views of the left ventricle (LV) was performed using an experimental scanner (HD-PULSE) at 1040±200 frames per second. An anatomical M-mode was extracted from the midline of the LV septum and color coded for tissue acceleration. The SW propagation velocity at MVC was measured as slope on the M-mode (A). Standard echocardiographic parameters were obtained with a high-end ultrasound machine (Vivid 95, GE Vingmed Ultrasound). The algorithm for evaluating diastolic dysfunction as recommended by the European Association of Cardiovascular Imaging expert consensus 2021 was applied to categorize LV filling pressure as normal or elevated. Results SW velocities correlated significantly with both, LVMDP (r=0.47; p<0,001) and LVEDP (r=0.71, p<0.001). They could excellently detect elevated LVEDP (AUC=0.95, Sensitivity=0.92, Specificity=0.94) and detected elevated LVMDP with AUC=0.79, Sensitivity=0.87, Specificity= 0.65, similar to the guideline approach (AUC=0.78, Sensitivity=0.62, Specificity=0.94). Conclusions SW velocities, measured by high frame rate echocardiography, show a strong correlation with the end-diastolic and a good correlation with mid-diastolic filling pressures. The method could excellently distinguish normal and elevated LVEDP and could–through a single measurement-differentiate normal and elevated LVMDP as good as the current guideline algorithm with its combination of four parameters in a decision tree. Our data suggest a potential clinical value of the new method for the non-invasive assessment of diastolic function.