Externally-induced shear waves in the right ventricular free wall throughout the cardiac cycle

Funding Acknowledgements

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): 1. NWO STW/Dutch Heart foundation 2. NWO Vidi TTW

Background

The right ventricular (RV) size and function have prognostic value, not only in right heart conditions (pulmonary hypertension, congenital heart disease), but also in left ventricular (LV) disease states. There is currently no noninvasive method for measuring the elastic properties of the RV walls. Increased RV diastolic stiffness however, as determined invasively, is linked to adverse prognosis in pulmonary arterial hypertension. In the LV, new high frame rate (HFR) echocardiography applications have been used in order to determine the propagation speed of naturally-occurring or externally-induced shear waves (SW) in the septum, which is related to myocardial stiffness.

Purpose

We show for the first time that SWs in the RV free wall (RVFW) can be externally induced and imaged transthoracically with ultrasound.

Methods

SW imaging was performed using a linear array with a programmable research ultrasound system, aimed at the parasternal long-axis view of a 5-weeks-old Yorkshire-Landrace pig (Fig. 1a). SWs were induced by a high intensity focused beam (f0 = 4.5 MHz) on the RVFW (Fig. 1a). This push beam generates a downwards force at the focus, which induces SWs that propagate left and right from the focus along the RVFW, visible on tissue Doppler images (TDI) shown in Figure 1b-c. SW propagation was imaged (f0 = 5.2 MHz) using HFR ultrasound (Fig. 1b, frame rate = 3 kHz). Three acquisitions of 1 second were performed, where 14 SWs were sequentially induced during each acquisition. ECG signal was captured simultaneously and synchronized offline. SW speeds were estimated using a custom, semi-automatic pipeline that includes TDI calculation, and SW speed estimation along a manually traced spline on the RVFW. This was repeated two times to include variability due to the manual processes. Up to four SW speed estimates were made after each push beam. SW speed estimation was repeated for all push pulses throughout the cardiac cycle, and the mean and standard deviation of the estimates were plotted (Fig. 2).

Results

At least 85% of the waves were tracked successfully for all acquisitions. Diastole and systole were identified using the ECG signal (Fig. 2a). The average SW speed was 0.6 ± 0.1 m/s at end-diastole (Fig. 2b). The measured speeds ranged from 0.5 ± 0.1 m/s during diastole to 1.9 ± 0.3 m/s during systole. The changes in SW speeds correspond to the expected variation in muscle stiffness during the heart cycle as the RV relaxes and contracts.

Conclusions

We demonstrate for the first time the induction and tracking of shear waves in the RVFW of a closed-chest pig. The possibility to noninvasively quantify RV wall stiffness opens a large field of translational research, with direct applications in pulmonary hypertension, congenital heart disease and heart failure in general. Pathological increase in stiffness should be further investigated in longitudinal case/control studies. Abstract Figure. Fig1: Induction and tracking of RV SW  Abstract Figure. Fig2: SW speeds during cardiac cycle

Left ventricular high frame rate echo-particle image velocimetry: feasibility and comparison with conventional echocardiography

Funding Acknowledgements

Type of funding sources: None.

Introduction

Echo-Particle Image Velocimetry (echoPIV) tracks speckle patterns from ultrasound contrast agent (UCA) microbubbles injected intravenously, being less angle-sensitive than colour Doppler. One limitation of conventional echoPIV is the inability to accurately resolve high velocities, because of relatively low frame rates.  In contrast, high frame rate (HFR) echoPIV enables tracking of fast flow in the left ventricle (LV).

Purpose

To investigate the feasibility and precision of HFR echoPIV in patients.

Methods

19 heart failure patients were included. UCA was infused with a dedicated pump. HFR contrast images were acquired, in apical long axis view (ALAX, ensuring simultaneous visualization of LV inflow and outflow), using a fully-programmable research ultrasound system, with a phased array probe. In the same session, complete echocardiographic studies were obtained using a clinical ultrasound system, with a matrix array probe, including LV UCA. Non-contrast pulsed-wave (PW) Doppler were also obtained in ALAX (Figure 1) from the mitral valve tips (inflow) and the LV outflow tract (outflow). HFR echoPIV image quality and tracking were assessed offline by two independent observers. The peak velocity of the inflow and outflow were determined by the automated tracking algorithm of the HFR echoPIV, and measured by the peak modal velocity of the conventional PW. These velocities were compared using Pearson’s correlations and Bland-Altman plots. All patients gave their informed consent. The study was approved by the institutional review board.

Results

Conventional echo image quality was good in 12 (63%), medium in 5 (26%) and bad in 2 (11%). EchoPIV tracking was good in 12 (63%), medium in 2 (10%) and bad in 5 (26%). In the 12 patients where echoPIV tracking was good, the direction and velocity of intracavitary vortices could be visualized (Figure 1). The inflow velocity could be determined by echoPIV  in 17/19 (89%), and outflow in 14/19 (74%). EchoPIV tended to underestimate the maximal velocity as determined by PW (Figure 2), with a bias of 0.19 m/s (inflow) and 0.28 m/s (outflow). This negative bias is expected as the PW is assessing maximum velocity in the interrogation kernel whereas echoPIV returns the mean velocity. The correlation of the two methods was good for the inflow (R2 = 0.77, p < 0.001) and moderate for the outflow (R2 = 0.54, p < 0.001). This may be explained by the position of the LV outflow tract deeper in the image, leading to increased attenuation, clutter and reduced lateral resolution.

Conclusion

HFR echoPIV has comparable feasibility to routine echocardiography, and the ability to correctly estimate intraventricular flow velocity. It can provide in one acquisition all the functional information that can be detected by routine echocardiography, PW and color Doppler, as well as contrast. It succeeds in surpassing the shortcomings of Doppler (angle dependency) and classical contrast imaging (low frame rate). Abstract Figure 1: HFR echoPIV in study patients  Abstract Figure 2: study results

On the interplay of loading, myocardial stiffness and contractility in transthoracic acoustic radiation force-induced shear wave measurements in pigs

Funding Acknowledgements

Type of funding sources: Public Institution(s). Main funding source(s): Research Foundation Flanders (FWO grant number 1211620N) & TTW-Dutch Heart Foundation partnership program "Early recognition of cardiovascular diseases" (project number 14740)

Background

Acoustic radiation force-based shear wave elastography (SWE) is a promising technique to non-invasively assess mechanical properties of the heart based on the propagation speed of acoustically induced shear waves. However, the interpretation of cardiac SWE measurements remains complex, and it is unclear how other factors such as loading affect shear wave propagation speed (SWS) measurements in diastole and systole.

Purpose

We applied transthoracic SWE in a pig model to investigate the dependencies of diastolic and systolic SWS on pressure-volume (PV) loop derived indices of loading, myocardial stiffness and contractility.

Methods

In 7 pigs, loading conditions were altered (increasing or decreasing preload; increasing afterload) and myocardial stiffness was changed (LAD occlusion for 60-100 minutes followed by 40 minutes of reperfusion). For each intervention, transthoracic SWE measurements were performed in a parasternal long-axis view with a high frame rate ultrasound system (> 6.2 kHz). Recordings of 28 ms were repeated at 34 Hz during 1.5 s to track shear waves throughout the cardiac cycle. To determine systolic and diastolic SWS in a robust manner, a piecewise linear model was fitted to the SWS data of each intervention representing multiple acquisitions, heartbeats and M-lines (fig. 1a). PV loops were recorded simultaneously with SWE measurements to estimate end-diastolic pressure (EDP), end-systolic pressure (ESP), end-diastolic pressure volume relation (EDPVR with exponential coefficient β in fig. 1b) and preload-recruitable stroke work (PRSW). Passive chamber stiffness was evaluated as the local slope of the EDPVR, i.e. β·EDP. Linear regressions and Pearson’s correlation coefficients were calculated.

Results

  Diastolic SWS was significantly correlated to EDP when altering loading (blue in fig. 2a: R = 0.55; p < 0.01) and stiffness (orange in fig. 2a: R = 0.66; p < 0.01). A similar correlation is found between SWS and passive chamber stiffness β·EDP (fig. 2b). Diastolic SWS is more sensitive to changes in stiffness than in loading, as reflected by the larger slope of the regression line (0.79 vs. 0.28 in fig. 2b). Furthermore, systolic SWS significantly correlated with measures of contractility during loading alterations: ESP in fig. 2c (R = 0.69; p < 0.001) and PRSW in fig. 2d (R = 0.63; p = 0.02). However, no significant correlation was found between systolic SWS and contractility during infarct/reperfusion.

Conclusion

This study shows that diastolic SWS reflects the instantaneous stiffness of the myocardium, but is not a load-independent measure of the intrinsic passive mechanical properties of the heart. Instantaneous stiffness, and thus diastolic SWS, might be altered by loading (due to material non-linearity) or intrinsic mechanical changes. Furthermore, loading experiments suggest that systolic SWS is related to contractility. The relation between SWS and contractility in the presence of myocardial infarct deserves further study. Abstract Figure. Fig. 1: SWS and PV analysis.  Abstract Figure. Fig. 2: SWS vs. PV-derived indices.

A spatial and temporal characterisation of single proton acoustic waves in proton beam cancer therapy

An in vivo range verification technology for proton beam cancer therapy, preferably in real-time and with submillimeter resolution, is desired to reduce the present uncertainty in dose localization. Acoustical imaging technologies exploiting possible local interactions between protons and microbubbles or nanodroplets might be an interesting option. Unfortunately, a theoretical model capable of characterising the acoustical field generated by an individual proton on nanometer and micrometer scales is still missing. In this work, such a model is presented. The proton acoustic field is generated by the adiabatic expansion of a region that is locally heated by a passing proton. To model the proton heat deposition, secondary electron production due to protons has been quantified using a semi-empirical model based on Rutherford's scattering theory, which reproduces experimentally obtained electronic stopping power values for protons in water within 10% over the full energy range. The electrons transfer energy into heat via electron-phonon coupling to atoms along the proton track. The resulting temperature increase is calculated using an inelastic thermal spike model. Heat deposition can be regarded as instantaneous, thus, stress confinement is ensured and acoustical initial conditions are set. The resulting thermoacoustic field in the nanometer and micrometer range from the single proton track is computed by solving the thermoacoustic wave equation using k-space Green's functions, yielding the characteristic amplitudes and frequencies present in the acoustic signal generated by a single proton in an aqueous medium. Wavefield expansion and asymptotic approximations are used to extend the spatial and temporal ranges of the proton acoustic field.

Tapering of the interventricular septum can affect ultrasound shear wave elastography: An in vitro and in silico study

Shear wave elastography (SWE) has the potential to determine cardiac tissue stiffness from non-invasive shear wave speed measurements, important, e.g., for predicting heart failure. Previous studies showed that waves traveling in the interventricular septum (IVS) may display Lamb-like dispersive behaviour, introducing a thickness-frequency dependency in the wave speed. However, the IVS tapers across its length, which complicates wave speed estimation by introducing an additional variable to account for. The goal of this work is to assess the impact of tapering thickness on SWE. The investigation is performed by combining in vitro experiments with acoustic radiation force (ARF) and 2D finite element simulations, to isolate the effect of the tapering curve on ARF-induced and natural waves in the heart. The experiments show a 11% deceleration during propagation from the thick to the thin end of an IVS-mimicking tapered phantom plate. The numerical analysis shows that neglecting the thickness variation in the wavenumber-frequency domain can introduce errors of more than 30% in the estimation of the shear modulus, and that the exact tapering curve, rather than the overall thickness reduction, determines the dispersive behaviour of the wave. These results suggest that septal geometry should be accounted for when deriving cardiac stiffness with SWE.

Natural shear wave propagation speed is influenced by both changes in myocardial structural properties as well as loading conditions

Funding Acknowledgements

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Research Foundation - Flanders (FWO)

Background

Shear wave elastography (SWE) is a promising tool for the non-invasive assessment of myocardial stiffness. It is based on the evaluation of the propagation speed of shear waves by high frame rate echocardiography. These waves can be induced by for instance mitral valve closure (MVC) and the speed at which they travel is related to the instantaneous stiffness of the myocardium. Myocardial stiffness is defined by the local slope of the stress-strain relation and can therefore be altered by both changes in structural properties of the myocardium as well as loading conditions.

Purpose

The aim of this study was to investigate how changes in myocardial structural properties as well as loading conditions affect shear wave speed after MVC.

Methods

Until now, 8 pigs (weight: 33.6 ± 5.4 kg) were included. The following interventions were performed: 1) preload was reduced by balloon occlusion of the vena cava inferior, 2) afterload was increased by balloon occlusion of descending aorta, 3) preload was increased by intravenous administration of 500 ml of saline and 4) ischemia/reperfusion injury (I/R injury) was induced in the septal wall by balloon occlusion of the LAD for 90 min. with subsequent reperfusion for 40 min. Echocardiographic and left ventricular pressure recordings were simultaneously obtained during each intervention. Left ventricular parasternal long-axis views were acquired with an experimental high frame rate ultrasound scanner (average frame rate: 1279 ± 148 Hz). Shear waves were visualized on tissue acceleration maps by drawing an M-mode line along the interventricular septum. Shear wave propagation speed after MVC was calculated by assessing the slope of the wave pattern on the tissue acceleration map (Figure A).

Results

The change in left ventricular end-diastolic pressure (LVEDP) and shear wave speed after MVC between baseline and each intervention are shown in Figure B and C, respectively. Preload reduction resulted in significant lower LVEDP compared to baseline (p < 0.01), while the other loading changes did not have a significant effect. Shear wave speed after MVC significantly increased by afterload and preload increase (p < 0.01). I/R injury resulted in increased shear wave speed (p < 0.01) without significantly altering LVEDP. There was a good positive correlation between the change in LVEDP and the change in shear wave speed induced by loading changes (r = 0.76; p < 0.001) (Figure D). However, the correlation became less strong if data of I/R injury was taken into account as well (r = 0.63; p < 0.001).

Conclusion

Our results suggest that SWE is capable to characterize myocardial tissue properties and besides has the potential as a novel method for the estimation of left ventricular filling pressures. However, in the presence of structural changes of the myocardium, care should be taken when estimating filling pressures based on shear wave propagation speed.

Abstract Figure.

Closed-chest measurement of diastolic and systolic shear wave speed to assess myocardial stiffness

Funding Acknowledgements

Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Research Foundation Flanders (FWO): grant 1211620N TTW – Dutch Heart Foundation partnership program "Earlier recognition of cardiovascular diseases": project number 14740

Background

Echocardiographic shear wave elastography (SWE) encompasses all ultrasound techniques tracking shear wave (SW) motion in the cardiac wall, of which the propagation speed is linked to the intrinsic mechanical properties. SWs can be induced naturally, for example by valve closure, or externally by using an acoustic radiation force (ARF). Although the latter is technically more demanding, it enables instantaneous stiffness assessment throughout the entire cardiac cycle (fig. a). However, it is unknown how factors such as cardiac loading and contractility, next to intrinsic mechanical properties, affect ARF-based SW speeds.

Purpose

We performed transthoracic SWE measurements in pigs to study the effects of hemodynamic alterations, inotropic state and myocardial infarction (MI) on diastolic and systolic SW speeds.

Methods

Different cardiac conditions were considered in three pigs: (i) baseline (BL), (ii) preload decrease (PD), (iii) afterload increase (AI), (iv) preload increase (PI), (v) administration of dobutamine (DOB), (vi) BL2, (vii) MI through 60-100 min. occlusion of the LAD and (viii) 40 min. reperfusion (REP). For each condition, transthoracic high frame rate ARF-based SWE acquisitions were taken in a parasternal long-axis view with a research ultrasound system. SWs were induced in the septum at 34 Hz during 1.5 s to track SW speeds throughout the cardiac cycle (fig. a&b). Systolic and diastolic SW speeds were determined from the 10% highest and lowest median values per condition, respectively. Left ventricular pressure-volume (PV) loops were recorded to estimate end-diastolic pressure (EDP), end-systolic pressure (ESP) and passive chamber stiffness (dPdV). dPdV was determined as the slope of the tangent to the fitted end-diastolic PV relationship at mean ED volume. Linear regressions and Pearson’s correlation coefficients were computed.

Results

Diastolic SW speed was correlated to EDP for conditions with changes in loading, and to dPdV for conditions with changes in chamber stiffness (fig. c).  Both relationships were significant, with a moderate positive correlation for EDP (R = 0.48, p = 0.02) and a strong positive correlation for dPdV (R = 0.76, p < 0.01). Furthermore, the observed change in diastolic SW speed was smaller when altering EDP compared to dPdV (0.4 m/s vs. 1.0 m/s). For systolic SW speed, very strong positive correlations were found with ESP (R = 0.91, p < 0.01), and with dPdV (R = 0.81, p < 0.01) in fig. d.

Conclusion

This study shows that both diastolic and systolic SW speed are related to passive chamber stiffness. Moreover, loading also influenced systolic SW speed and, to a lesser extent, diastolic SW speed, presumably because of material nonlinearity. Systolic SW speed is linked to contractility as well. Thus, while SWs after valve closure occur at a certain moment in the cardiac cycle, the timing of ARF-based SWs can be chosen such to assess specific aspects of the cardiac (structural and functional) status.

Abstract Figure.

Fundamental modeling of wave propagation in temporally relaxing media with applications to cardiac shear wave elastography

Shear wave elastography (SWE) might allow non-invasive assessment of cardiac stiffness by relating shear wave propagation speed to material properties. However, after aortic valve closure, when natural shear waves occur in the septal wall, the stiffness of the muscle decreases significantly, and the effects of such temporal variation of medium properties on shear wave propagation have not been investigated yet. The goal of this work is to fundamentally investigate these effects. To this aim, qualitative results were first obtained experimentally using a mechanical setup, and were then combined with quantitative results from finite difference simulations. The results show that the amplitude and period of the waves increase during propagation, proportional to the relaxation of the medium, and that reflected waves can originate from the temporal stiffness variation. These general results, applied to literature data on cardiac stiffness throughout the heart cycle, predict as a major effect a period increase of 20% in waves propagating during a healthy diastolic phase, whereas only a 10% increase would result from the impaired relaxation of an infarcted heart. Therefore, cardiac relaxation can affect the propagation of waves used for SWE measurements and might even provide direct information on the correct relaxation of a heart.

P1138Cardiac shear wave velocity in healthy individuals

Background

The closure of the valves generates shear waves in the heart walls. The propagation velocity of shear waves relates to stiffness. This could potentially be used to estimate the stiffness of the myocardium, with huge potential implications in pathologies characterized by a deterioration of the diastolic properties of the left ventricle. In an earlier phantom study we already validated shear wave tracking with a clinical ultrasound system in cardiac mode.

Purpose

In this study we aimed to measure the shear waves velocity in normal individuals.

Methods

12 healthy volunteers, mean age=37±10, 33% females, were investigated using a clinical scanner (Philips iE33), equipped with a S5-1 probe, using a clinical tissue Doppler (TDI) application. ECG and phonocardiogram (PCG) were synchronously recorded. We achieved a TDI frame rate of >500Hz by carefully tuning normal system settings. Data were processed offline in Philips Qlab 8 to extract tissue velocity along a virtual M-mode line in the basal third of the interventricular septum, in parasternal long axis view. This tissue velocity showed a propagating wave pattern after closure of the valves. The slope of the wave front velocity in a space-time panel was measured to obtain the shear wave propagation velocity. The velocity of the shear waves induced by the closure of the mitral valve (1st heart sound) and aortic valve (2nd heart sound) was averaged over 4 heartbeats for every subject.

Results

Shear waves were visible after each closure of the heart valves, synchronous to the heart sounds. The figure shows one heart cycle of a subject, with the mean velocity along a virtual M-mode line in the upper panel, synchronous to the ECG signal (green line) and phonocardiogram (yellow line) in the lower panel. The slope of the shear waves is marked with dotted lines and the onset of the heart sounds with white lines. In our healthy volunteer group the mean velocity of the shear wave induced by mitral valve closure was 4.8±0.7m/s, standard error of 0.14 m/s. The mean velocity after aortic valve closure was 3.4±0.5m/s, standard error of 0.09 m/s. We consistently found that for any subject the velocity after mitral valve closure was higher than after aortic valve closure.

Conclusion

The velocity of the shear waves generated by the closure of the heart valves can be measured in normal individuals using a clinical TDI application. The shear wave induced after mitral valve closure was consistently faster than after aortic valve closure. Abstract P1138 Figure.

Abstract P1138 Figure.

Nonspherical vibrations of microbubbles in contact with a wall: a pilot study at low mechanical index

Radially oscillating microbubbles can deform when in contact with a wall. These nonspherical shapes have a preferential orientation perpendicular to the wall. Conventional microscope setups for microbubble studies have their optical axis perpendicular to the wall (top view); consequently they have a limited view of the deformation of the bubble. We developed a method to image the bubble in a side view by integrating a mirror in the microscope setup. The image was recorded at 14.5 million frames per second by a high-speed camera. When insonified by a 1-MHz, 140-kPa ultrasound pulse, a 9-microm diameter coated bubble appeared spherical in the top view, but strongly nonspherical in the side view. Its shape was alternatively oblate and prolate, with maximum second order spherical harmonic amplitude equal to the radius.

A minimax procedure in the context of sequential testing problems in psychodiagnostics

The purpose of this paper is to derive optimal rules for sequential testing problems in psychodiagnostics. In sequential psychodiagnostic testing, each time a patient is exposed to a new treatment, the decision then is to declare this new treatment effective, ineffective, or to continue testing and exposing the new treatment to another random patient suffering from the same mental health problem. The framework of minimax sequential decision theory is proposed for solving such testing problems; that is, optimal rules are obtained by minimizing the maximum expected losses associated with all possible decision rules at each stage of testing. The main advantage of this approach is that costs of testing can be explicitly taken into account. The sequential testing procedure is applied to an empirical example for determining the effectiveness of a cognitive-analytic therapy for patients suffering from anorexia nervosa. For a given maximum number of patients to be tested, the appropriate action is indicated at each stage of testing for different numbers of positive reactions to the cognitive-analytic therapy. The paper concludes with a simulation study, in which the minimax sequential strategy is compared for the anorexia nervosa example with other procedures that exist for similar classification decision problems in the literature in terms of average number of patients to be tested, classification accuracy and average loss.

A Simultaneous Approach to Optimizing Treatment Assignments with Mastery Scores

A model for simultaneous optimization of combinations of test-based decisions in psychology and education is proposed using Bayesian decision theory. The decision problem addressed consists of a combination of a placement and a mastery decision. Weak and strong decision rules are distinguished. As opposed to strong rules, weak rules are allowed to take prior test scores in the series of decisions into account. The introduction of weak rules makes the placement-mastery problem a multivariate decision problem. Conditions for optimal rules to take monotone forms are derived. Results from an empirical example of instructional decision making are presented to illustrate the differences between a simultaneous and a separate approach.

Countercurrent multistage fluidized bed reactor for immobilized biocatalysts: II. Operation of a laboratory-scale reactor

In Part I of this series,(1) we derived a model and made simulations for a multistage fluidized bed reactor (MFBR). It was concluded that the MFBR can be an attractive alternative for a fixed bed reactor when operated with a deactivating biocatalyst. In Part II of this series, the design of a laboratory-scale MFBR and its evaluation to investigate the practical feasibility of this reactor type, will be described. Experiments with a duration as long as 10 days were carried out successfully using immobilized glucose isomerase as a model reaction system. The results predicted by the model are in good agreement with the measured glucose concentration and biocatalyst activity gradients, indicating perfect mixing of the particles in the reactor compartments.The diameters of the biocatalyst particles used in the experiments showed a large spread, with the largest being 1.7 times the smallest. Therefore, an additional check was carried out, to make sure that the particles were not segregating according to size. Particles withdrawn from the reactor compartments were investigated using an image analyzer. Histograms of particle size distribution do not indicate segregation and it is concluded that the particles used have been mixed completely within the compartments. As a result, transport of biocatalyst is nearly plug flow.