Young Investigator Award session – Basic ScienceIntracardiac flows and pressures70Layer specific RF ultrasound based radial strain values confirmed by a geometrical model71Evaluation of left atrial remodeling secondary to chronic post-ischemic cardiomyopathy in the mouse72Three-dimensional intracardiac blood velocity and flow from compounded 3D echo colour Doppler images

# 70 Layer specific RF ultrasound based radial strain values confirmed by a geometrical model {#article-title-2} Purpose: Cardiac deformation imaging techniques have been used to assess the local effects of cardiac regeneration therapy. Limited spatial and temporal resolution of clinically available techniques prevents detailed analysis of local myocardial deformation. Other imaging techniques might be more appropriate to assess local therapeutic effects. We used radio-frequency (RF) ultrasound to estimate local radial strain. To confirm the results analytically, we used an ellipsoidal geometrical model of the left ventricle. Methods: Data acquisition was done epicardially in 5 pigs. RF ultrasound data were acquired using a Medison Accuvix V10 with an RF interface. Data acquisition was done at 129 fps, using an L5-13 linear array transducer (center frequency = 8.5 MHz). Parameters for the analytical radial strain solution were (see figure): end diastolic outer radius (Rod = 3.5cm), end diastolic inner radius (Rid = 2.5cm), end diastolic length (Ld = 9cm), end systolic outer radius change −5%, end systolic length change −23%. Results: RF based radial strain estimation shows a transmural gradient of peak radial strain from sub-endocard to sub-epicard from 72% to 34%. The analytical solution resulted in reasonable hemodynamic parameters: EDV = 117ml, ESV = 47ml and EF = 59%. Systolic radial strain of the analytical solution was in the same range as the peak radial strain in the experiments and ranged from 80% to 36%. Conclusion: By using a geometric model based on parameters deduced from the experiments, and the assumption that cardiac tissue is incompressible, peak radial strain values in different layers of the myocardium found by RF ultrasound analysis could be confirmed analytically. We hereby show analytically and experimentally that sub-endocardial peak radial strain values can be as high as 80 to 90%.[⇓][1] ![Figure][2] # 71 Evaluation of left atrial remodeling secondary to chronic post-ischemic cardiomyopathy in the mouse {#article-title-3} Purpose: Congestive heart failure is associated in man with significant left atrial (LA) dilatation and reduced compliance (LA remodeling), secondary to left ventricular (LV) systolic dysfunction and dilatation. It is also known that LA compliance is an independent determinant of left ventricular (LV) stroke volume (SV). Because rodent models of LV myocardial infarction (MI) are widely used, but knowledge of LA anatomy and function is poor in this setting, the aim of this study was to investigate LA remodelling in a mouse model of chronic post-MI LV dysfunction. Methods: Ten C57BL/6 female mice underwent left anterior descending coronary artery ligation surgery (MI-group). Cardiac magnetic resonance was used to confirm a reduction in ejection fraction (EF) to ≤ 45% (large infarct) after 24 hours. Ten sham-operated mice were used as controls. High frame rate echocardiography (Vevo 2100 system) was used (apical 4-chamber views) to measure LA areas and compliance (reservoir function: ((maximum – minimum area)/minimum) x 100), and LV volumes, EF and SV during follow-up (48 hours, 4 and 8 weeks). Results: All mice were alive at 8 weeks. No LA remodeling was observed after 48hours, whereas after 4 weeks LV systolic dysfunction (EF: 32±11 % vs baseline 77±8, p<.001) and prevalent increase in diastolic over systolic volume (diastolic volume: 88±15 ml vs baseline 35±3, p<.001) was associated with preserved SV (27±5 ml vs baseline 28±7, p=ns), and LA dilatation and reduced compliance (maximum area: 4.4±1 mm2 vs baseline 2.7±.2, p<.001; reservoir: 39±12 % vs baseline 66±11, p<.001). After 8 weeks EF was unchanged, but a further prevalent increase in LV diastolic (148±40 ml) over systolic volume led to an increase in SV (49 ml), in parallel with an increase in LA reservoir compliance (52±14 %) secondary to a reduction of LA minimum area. A moderate positive correlation was found between SV and LA reservoir (r= .6, p=.02). Heart rate did not change at any step (475 bpm), and no echocardiographic changes were observed in controls. Conclusions: This study is the first description of LA remodeling in a chronic mouse model of post-MI LV dysfunction. In this model, a large anterior MI causes extensive eccentric LV remodeling which is associated with early LA dilatation and reduction in LA compliance followed by late partial recovery of LA compliance in parallel with an increase in LV SV. Our data confirm the positive relation between LV SV and LA compliance in the mouse, although further studies are needed to unveil mechanisms underlying SV preservation in post-MI cardiomyopathy in the mouse. # 72 Three-dimensional intracardiac blood velocity and flow from compounded 3D echo colour Doppler images {#article-title-4} Introduction: accurate 3D characterisation, quantification and visualisation of cardiac blood flow are of great clinical interest since it may improve understanding of cardiac function and diagnosis of some cardiac diseases. Currently, it can only be done with 4D MR flow, which is expensive and requires long acquisition times compared to echo. Echo Doppler is angle dependent and can only measure blood velocity along the direction of the echo beam. We propose a multiview technique for 3D velocity reconstruction from three 3D colour Doppler images. Method: we propose a method for 3D velocity reconstruction and study the impact of view angles and noise on reconstruction accuracy, from which we establish the clinical requirements for accurate 3D velocity reconstruction. Our method is tested on 3 volunteers. Data was acquired using a Philips iE33 (X3-1) 3D probe. For each volunteer 3 views were acquired (2 apical and 1 parasternal, Fig. 1). Results: we found that if angles between views are > 20° and noise in colour Doppler images is ≤10 cm/s, the reconstructed 3D velocity yields flow rates with error ≤ 10% compared to flow MR. Imaging data from 2 out of 3 volunteers had angles and noise out of these ranges, reducing the accuracy of the 3D velocity field. An example of a volunteer whose imaging data was within the ranges of noise and angles is shown in Fig. 2. Three-dimensional velocity through the aorta (mid systole) is represented as colour-coded arrows. Figure 3 represents a 2D projection of the reconstructed field in a 4-chamber view, showing the flow pattern and velocity profile. Conclusion: 3D blood velocity can be reconstructed from multiple 3D colour Doppler views if view angles and noise are within appropriate ranges. Promising results have been achieved for echogenic subjects.[⇓][3] ![Figure][2] [1]: #F1 [2]: pending:yes [3]: #F2