Left Ventricular Mass Influences Relationship Between Filling Pressure and Early-Diastolic Ratio of Inflow Velocity to Mitral Annular Velocity (E/e′)
hypertrophic cardiomyopathy (HCM) 6 and those with severe LV dysfunction. 7 We also have reported that e′ significantly but only weakly correlated with the time constant of LV pressure decay (τ) and that E/e′ modestly correlated with filling pressure in a multicenter study in Japan (strain/strain rate vs. myocardial velocity for assessing left ventricular relaxation and filling pressure: SMAP study). 8 Accordingly, we speculated that some characteristics of the subjects in the SMAP study had
... e SMAP study had altered the relationships between τ and e′ and between LV filling pressure and E/e′, and we conducted a substudy of SMAP. Because HCM itself has idiosyncratic features associated with LV T he tissue Doppler-derived early-diastolic mitral annular velocity (e′) and the ratio of early-diastolic left ventricular (LV) inflow velocity (E) to e′ (E/e′) reflect LV relaxation and filling pressure, respectively. 1-3 It is recommended that e′ and E/e′ should not be used for the estimation of LV relaxation and filling pressure in normal subjects, or patients with significant mitral annular calcification, surgical rings, mitral stenosis, prosthetic mitral valves, moderate to severe primary mitral regurgitation, or constrictive pericarditis. 4,5 Some recent reports, however, have noted that E/e′ does not correlate with LV filling pressure even in some clinical settings such as patients with Background: Early-diastolic mitral annular velocity (e′) and the ratio of early-diastolic left ventricular (LV) inflow velocity (E) to e′ (E/e′) have been widely used as indexes of LV relaxation and filling pressure, respectively. However, many recent studies have demonstrated that they are not reliable in various clinical settings. We thus investigated the factors influencing these echocardiographic parameters in a multicenter study. Methods and Results: The study group comprised 69 patients, referred for cardiac catheterization, and enrolled in 5 university hospitals. Time constant (τ) and LV mean diastolic pressure (LVMDP) were measured using a micromanometer-tipped catheter. Although e′ only weakly correlated with τ (r=−0.35, P<0.01), E/e′ modestly correlated with LVMDP (r=0.48, P<0.001). Multivariable analysis revealed that hypertension (β=−0.33, P<0.01) and LV ejection fraction (LVEF) (β=0.44, P<0.001) were the independent determinants of e′, and LV mass index (LVMI) (β=0.37, P<0.001) and LVMDP (β=0.47, P<0.001) were those of E/e′. Additionally, E/e′ significantly correlated with LVMDP in patients with normal LVMI (r=0.74, P<0.001) but not in those with increased LVMI. Conclusions: The coincidence of hypertension and LVEF affected the relationship between LV relaxation and e′, whereas LVMI altered the relationship between LV filling pressure and E/e′. Thus, clinical conditions associated with an increase in LVMI, such as LV hypertrophy and LV dilatation, should be considered when estimating the filling pressure from E/e′.