2.3 Echocardiographic evaluation
The echocardiographic evaluation was performed by a single operator (LL). A commercially available system (Vivid 7, GE Vingmed ultrasound AS) was used for all patients. Standard parasternal and apical views were usedand acquired images were stored digitally in high analysis still images and in cine loops (in a format of three consecutive beats for analysis). A single observer blinded of the patients’ identity (not the same person who performed the echocardiographic examination) performed offline analysis using EchoPac (version 113 - GE Vingmed ultrasound AS). On the day of the examination, echocardiographic assessment was performed in a two-staged approach. Initially, a basic echocardiogram was performed. All classic LV function related systolic and diastolic indices were obtained, according to the European Society of Cardiology and European Association of Cardiovascular Imaging guidelines [22]. Left atrial volume and LV mass were both indexed to patients’ body surface area.
In order to obtain 2DSTE data, both parasternal and apical views (at frame rates 60-90Hz) were acquired. Thus, adequate spatial and temporal resolution and accurate frame to frame tracking (for three consecutive cardiac cycles) was ensured. The endocardial LV borders were manually traced (region of interest). When tracking was poor in more than two consecutive myocardial regions, the acquired data were declined. The timings of aortic and mitral valve opening and closure were manually defined by the use of pulsed-wave Doppler. No patient was excluded based on poor 2DSTE-related echocardiographic data. 2DSTE analysis included assessment of GLS and strain rate as also global radial and circumferential strain and strain rate. LV twist was calculated as the difference between apical and basal LV rotation as it was assessed from equivalent short-axis views. Studies with >2 consecutive segments (out of a total amount of 17) not adequately tracked were deemed as inappropriate for the measurement of LV twist and rotation. Untwist rate was measured as the peak negative time derivative of twist during diastole. The time interval from R wave peak to the maximal untwist was then calculated.
Following the baseline echocardiographic evaluation, infusion of dipyridamole for 6 minutes (0.84mg/kg) was performed. CFR was calculated as the ratio between hyperemic and basal coronary flow at the LAD area. Just before the end of dipyridamole infusion a new echocardiographic assessment (focused mainly on LV systolic and diastolic function indices) was performed. At the end of the dipyridamole infusion, 125-250mg of aminophylline was administered to the patient, to counteract any dipyridamole negative effect. The dose was dependent on the patient’s status after dipyridamole infusion. Beverages containing methylxanthines such as coffee, tea, chocolate and coke were prohibited for at least 24 hours before the study. Patients with CFR values<2 were referred for another ischemia stress test and/or coronary angiographyto exclude potential significant epicardial stenosis in the LAD area. None of the enrolled patients with CFR values<2 had confirmed significant CAD. Intra-observer variability for all indices of interest included in this study has been previously reported [23].