Assessment of left ventricular systolic function using echocardiography in patients with preserved ejection fraction and elevated diastolic pressures

There is controversy regarding the nature of systolic function in patients with elevated filling pressure and preserved left ventricular (LV) ejection fraction. In this study, tissue Doppler variables and 2-dimensional echocardiographic systolic strain (SS) and systolic strain rate (SSr) were measured in patients who underwent cardiac catheterization to determine correlations with invasively measured LV end-diastolic pressure (LVEDP), dP/dt, and LV mass. Forty patients were studied. Their mean age was 55.9+/-9.9 years, and their mean LV ejection fraction was 59.8+/-5.2%. Tissue Doppler systolic annular velocity (5.4+/-1.1 vs 6.4+/-1.0 cm/s, p=0.04), SS (13.4+/-3.7% vs 18.8+/-2.3%, p0.001), and SSr (0.73+/-0.17 vs 0.98+/-0.14 s(-1), p0.001) were significantly lower in patients with LVEDP20 mm Hg compared with those with LVEDP20 mm Hg. Tissue Doppler systolic velocity, SSr, and SS were correlated with LV mass (R=0.58, R=0.57, and R=0.52, respectively, all p values0.001) and with LVEDP (R=0.49, p=0.002; R=0.79, p0.001; and R=0.70, p0.001, respectively). However, dP/dt was not significantly different between patients with LVEDP20 mm Hg and those with LVEDP20 mm Hg (1,387+/-520 vs 1,495+/-594 mm Hg/s, respectively, p=0.55) and was not correlated with LV mass (R=0.18, p=0.25). The optimum cut-off values for LVEDP20 mm Hg were SSr0.85 s(-1) (area under the curve 0.88, p0.001, positive predictive value 89%, negative predictive value 86%) and SS16% (area under the curve 0.84, p=0.002, positive predictive value 88%, negative predictive value 79%). In conclusion, as opposed to invasively measured dP/dt, tissue Doppler systolic velocity and 2-dimensional echocardiographic SS and SSr are significantly depressed in patients with preserved LV ejection fraction and LVEDP20 mm Hg, suggesting that systolic abnormalities are present in at least some of these patients. These differences are likely because invasively measured dP/dt and these echocardiographic variables measure different systolic properties in patients with preserved LV ejection fraction.