BACKGROUND: This study aimed at exploring the correlation of left atrial longitudinal function by speckle tracking echocardiography (left atrial strain) and Doppler measurements (E/E' ratio) with direct measurements of left ventricular (LV) end-diastolic pressure (LVEDP) in patients stratified for different values of ejection fraction. METHODS: The study population was 80 stable patients with sinus rhythm undergoing cardiac catheterization. This population was selected in order to have four groups of 20 patients each with different LV ejection fraction (>55%, 45-54%, 30-44%, and <30%). LVEDP was obtained during cardiac catheterization; peak atrial longitudinal strain (PALS) and mean E/E' ratio were measured in all subjects. RESULTS: Similar correlations with LVEDP of global PALS and E/E' ratio were recorded in patients with preserved (r = -0.79 vs. r = 0.72, respectively; P < 0.0001 for both) or mildly reduced ejection fraction (r = -0.75 vs. r = 0.73, respectively; P < 0.0001 for both). A closer correlation of global PALS compared to E/E' ratio was evident in patients with moderate (r = -0.78 P < 0.0001; vs. r = 0.47 P = 0.01, respectively) and severe reduction (r = -0.74 P < 0.0001; vs. r = 0.19 ns, respectively) of LV ejection fraction. In multivariate analysis of all measurements, global PALS emerged as a determinant of the LVEDP, independent on other confounding factors and, with the cutoff value of 18.0% presented the best diagnostic accuracy to predict a LVDP above 12 mmHg (AUC 0.87). CONCLUSIONS: In patients with preserved or mildly reduced LV ejection fraction, global PALS and mean E/E' ratio presented good correlations with LVEDP. In patients with moderate or severe reduction of ejection fraction, E/E' ratio correlated poorly with invasively obtained LV filling pressures. Global PALS provided an overall better estimation of LV filling pressures.
BACKGROUND: This study aimed at exploring the correlation of left atrial longitudinal function by speckle tracking echocardiography (left atrial strain) and Doppler measurements (E/E' ratio) with direct measurements of left ventricular (LV) end-diastolic pressure (LVEDP) in patients stratified for different values of ejection fraction. METHODS: The study population was 80 stable patients with sinus rhythm undergoing cardiac catheterization. This population was selected in order to have four groups of 20 patients each with different LV ejection fraction (>55%, 45-54%, 30-44%, and <30%). LVEDP was obtained during cardiac catheterization; peak atrial longitudinal strain (PALS) and mean E/E' ratio were measured in all subjects. RESULTS: Similar correlations with LVEDP of global PALS and E/E' ratio were recorded in patients with preserved (r = -0.79 vs. r = 0.72, respectively; P < 0.0001 for both) or mildly reduced ejection fraction (r = -0.75 vs. r = 0.73, respectively; P < 0.0001 for both). A closer correlation of global PALS compared to E/E' ratio was evident in patients with moderate (r = -0.78 P < 0.0001; vs. r = 0.47 P = 0.01, respectively) and severe reduction (r = -0.74 P < 0.0001; vs. r = 0.19 ns, respectively) of LV ejection fraction. In multivariate analysis of all measurements, global PALS emerged as a determinant of the LVEDP, independent on other confounding factors and, with the cutoff value of 18.0% presented the best diagnostic accuracy to predict a LVDP above 12 mmHg (AUC 0.87). CONCLUSIONS: In patients with preserved or mildly reduced LV ejection fraction, global PALS and mean E/E' ratio presented good correlations with LVEDP. In patients with moderate or severe reduction of ejection fraction, E/E' ratio correlated poorly with invasively obtained LV filling pressures. Global PALS provided an overall better estimation of LV filling pressures.