Echocardiographic methods
All transthoracic echocardiographic data were performed by a GE Vivid E95 ultrasound machine (GE Healthcare, Oslo, Norway) with cardiac sector probe M5S and 4V (1.0–1.5 Hz), and using Echopac workstation (Version:201) for offline analyses. Connect the ECG to record the moving image of at least five cardiac cycles (frame rate>40fps), to ensure the image is clear (endocardium) and complete (including epicardium), and adjust the frame rate to 40% of the patient’s heart rate. All echocardiograms of each patient were obtained by the same ultrasound machine.
Standard ultrasoundexamination.
Along the parasternal LV long-axis view measured the LV end-systolic diameter (LVIDS), LV end-diastolic diameter (LVIDD), LV posterior wall thickness (LVPW), end-systolic volume (ESV), end-diastolic volume (EDV). LVEF as indicators to assess left ventricular systolic function was calculated by the modified Simpson biplane method, with images recoded from apical two-chamber and apical four-chamber views. Fractional shortening (FS) was derived from M-mode imaging. In the apical four-chamber section, the mean value of early diastolic peak velocity (e’) of septal and lateral mitral annulus was computed by tissue Doppler and pulsed Doppler imaging. The E/e′ ratio was calculated by the ratio of the peak early transmission velocity (E) to the average e’. Then FS and E/e′ ratio could be used to evaluate left ventricular diastolic function [16]. Moreover, we used M-mode and tissue Doppler imaging echocardiography methods to measure isovolumic relaxation time (IRT), isovolumic contraction time (ICT) and ejection time (ET), and then Tei index((IRT+ICT)/ET) was computed automatically by the ultrasonic system as a reliable estimate of left ventricular global function[17]. These data were expressed as averages of three consecutive cycles.