Tissue Doppler imaging: a new technique for assessment of pseudonormalization of the mitral inflow pattern

Left ventricular diastolic dysfunction (LVDD) is a frequent cause of heart failure. Doppler echocardiography has become the method of choice for the noninvasive evaluation of LVDD. However, pseudonormalization (PN) of the mitral inflow often presents a diagnostic challenge in clinical practice. In this setting, we sought to define the role of tissue Doppler imaging (TDI) of the septal mitral annulus. Echocardiography was performed in 36 consecutive subjects (age 59 +/- 10 years). Eighteen of these had diagnosed coronary artery disease (CAD) with recent onset of symptoms (within 3 months), 18 had clinical suspicion of CAD, and 15 had symptoms of heart failure (New York Heart Association [NYHA] Class 2.4 +/- 0.5). The mitral inflow profile (E, A, E/A) was measured by pulsed Doppler, and the deceleration time (DT) and the isovolumic relaxation time (IVRT) were calculated. Peak diastolic velocities of the septal mitral annulus (E(T), A(T), E(T)/A(T)) and the time interval from Q in the ECG to the onset of E(T) were derived from pulsed TDI. Left heart catheterization was performed for direct measurement of left ventricular end-diastolic pressure (LVEDP). PN defined by an E/A ratio > 1 and an LVEDP > or = 16 mmHg was found in nine patients. All patients with PN had symptoms of heart failure (NYHA Class 2.8 +/- 0.5). Patients with and without PN did not differ with respect to the E/A ratio (1.29 +/- 0.44 vs 1.16 +/- 0.23, P = ns), DT (182 +/- 38 msec vs 205 +/- 42 msec, P = ns), and IVRT (88 +/- 24 msec vs 92 +/- 18 msec, P = ns). In the group with PN, a significant reduction of E(T) (5.6 +/- 1.8 cm/sec vs 8.8 +/- 2. 9 cm/sec, P < 0.05) and E(T)/A(T) (0.5 +/- 0.16 vs 0.82 +/- 0.37, P < 0.05) was detected. In the PN group, the Q-E(T) interval was prolonged (404 +/- 48 msec vs 346 +/- 50 msec, P < 0.05). Receiver operating characteristic curve analysis for E(T) yielded an area under the curve of 0.78 +/- 0.06 (P = 0.034) for separating patients with versus without PN. When the combination of E(T) < 7 cm/sec and E(T)/A(T) < 1 was used as cutpoint, PN could be identified with a sensitivity of 83% and a specificity of 79%. There was no significant relation between LVEDP and either E(T) (r = 0.32, P > 0. 2) or the Q-E(T) interval (r = 0.14, P > 0.5). In conclusion, E(T) and the Q-E(T) interval appear to be useful parameters for assessing LV diastolic dysfunction in symptomatic patients with a pseudonormal mitral inflow pattern and elevated filling pressures.