Abstract 2122: Head-to-head Comparison Between Tissue Doppler Imaging And Velocity Encoded Magnetic Resonance Imaging For The Measurement Of Myocardial Velocities And The Assessment Of Left Ventricular Dyssynchrony And Diastolic Function.

Abstract

Aim of the study. To compare echocardiographic tissue Doppler imaging (TDI) and velocity encoded (VE) magnetic resonance imaging (MRI) for the measurement of myocardial velocities and the assessment of left ventricular (LV) dyssynchrony and diastolic function. Methods. Ten healthy volounteers (5 men, 31±5 yrs) and 25 heart failure (HF) pts (17 men, 58±11 yrs) underwent both cardiac MRI and TDI. Longitudinal myocardial peak systolic (PSV) and diastolic (PDV) velocities and time to PSV (Ts) and to PDV (Td) were measured at basal and mid level of LV septum and lateral wall. To quantify LV dyssynchrony, delay in Ts between basal septum and lateral wall was calculated (SLD) and pts were categorized in 3 groups according to the extent of LV dyssynchrony by TDI: minimal (SLD<30 ms), intermediate (30 – 60 ms) and extensive (>60 ms) LV dyssynchrony. As a measure of LV diastolic function, the ratio of transmitral E wave velocity and mitral annulus septal early velocity (E/E′) was also calculated and pts were classified in 3 groups: normal LV diastolic function (E/E′ <8), abnormal LV diastolic function (E/E′= 8 –15) and severe LV diastolic dysfunction (E/E′>15). Results. PSV measured with VE-MRI showed excellent correlation with PSV measured with TDI (r=0.94, P<0.001) in both normals and HF pts and Bland-Altman analysis revealed a small bias (MRI-TDI) of 1±1 cm/s (P< 0.001) between both techniques. A strong correlation was also observed between Ts measured with TDI and VE-MRI (r=0.97, P<0.001) without significant bias. Excellent agreement between TDI and VE-MRI was found for LV dyssynchrony classification with a weighted κ=0.96. e′ and A′ PDV measured with VE-MRI correlated very well with PDV measured with TDI (r=0.95, P<0.001) with a small bias of 0.4±1 cm/s (P<0.001) between the 2 techniques. Excellent agreement between TDI and VE-MRI was also observed in classifying LV diastolic function with a weighted k=0.91. A very good correlation has been observed also for Td of E′(r=0.96, P<0.001) and Td of A′ (r=0.88, P<0.001) measured with both techniques. Conclusions. TDI and VE-MRI can be used interchangeably to measure LV myocardial systolic and diastolic velocities and excellent agreement exists between the 2 techniques for LV dyssynchrony and diastolic function classification.