Non-invasive evaluation of left ventricular filling pressures in patients with abnormal relaxation

Abstract

The aim of the present study was to assess the ability of several echocardiographic and TDI (tissue Doppler imaging) parameters to predict an elevated LVEDP (left ventricular end-diastolic pressure) in patients with abnormal relaxation. Eighty-two consecutive patients presenting with an E/A ratio (ratio of early-to-late diastolic peak transmitral velocity) <0.9 scheduled for left heart catheterization underwent echocardiography, including TDI, and measurement of LVEDP using fluid-filled catheters. The difference in duration between PVR (retrograde peak in the pulmonary veins) and A (ΔPVR-A) was calculated from pulsed Doppler recordings. VP (propagation velocity of the early mitral inflow) was determined by colour M-mode. TDI measurements included E´ (early diastolic peak myocardial velocities near the lateral mitral annulus), MVG (the early diastolic transmyocardial velocity gradient of the posterior basal wall) and the PRT (peak relaxation time), determined as the time interval between aortic valve closure and peak E´. Fifty-six patients presented with LVEDP values <15 mmHg, whereas an LVEDP >15 mmHg was found in 26 patients. The index ΔPVR-A showed a significant linear correlation with LVEDP (r=0.7, P<0.001) and provided the highest predictive accuracy for the identification of LVEDP >15 mmHg [AUC (area under receiver operating characteristic curve)=0.83], followed by PRT (AUC=0.67), whereas other TDI-derived parameters failed to reach significance. In conclusion, ΔPVR-A enabled the most accurate non-invasive estimation of LVEDP. A prolonged peak relaxation time was the only TDI-derived index that differed significantly between patient groups.