6th Dutch Bio-Medical Engineering Conference
26 & 27 January 2017, Egmond aan Zee, The Netherlands
13:00   Cardiovascular
Chair: Bart Verkerke
15 mins
Kirby Lattwein, Himanshu Shekhar, Willem van Wamel, Andrew Herr, Christy Holland, Klazina Kooiman
Abstract: Infective endocarditis is a bacterial infection of the endocardial and prosthetic valvular surfaces in the heart, and is associated with high morbidity and mortality rates.The predominant bacteria causing this infection are Staphylococci, which can bind to existing thrombi to generate vegetations (biofilms) on heart valves. Infective endocarditis is challenging to cure, because both the immune system and antibiotics are often ineffective against the biofilm-associated bacteria [1, 2]. In this in vitro study, we report a novel strategy called sonobactericide to treat infective endocarditis using ultrasound and ultrasound contrast agents in combination with antibiotics and thrombolytics. We developed an in vitro model of Staphylococcus aureus infective endocardial vegetations using retracted human whole blood clots infected with bacteria. This model was validated by histology, which revealed a biofilm outer layer consisting of fibrin embedded Staphylococci. The inner portions of the clots were comprised of fibrin, erythrocytes, and sporadic immune cells. Confocal microscopy confirmed the presence of live bacteria in the biofilm. Infected clots were treated in an established in vitro flow model and thrombolytic efficacy was assessed using analysis of time-lapse microscopy images over thirty minutes [3,4]. Infected clots were exposed to continuous human plasma flow either alone or with different combinations of the following: oxacillin (172 μg/mL), recombinant tissue plasminogen activator (rt-PA; 3.15 μg/mL), pulsed continuous- wave ultrasound (120 kHz, 0.44 MPa peak-to-peak pressure), and ultrasound contrast agent Definity (2 μl/ml). Infected clots exposed to oxacillin, rt-PA, ultrasound, and Definity achieved almost 100% loss (99.5 ± 1.7%) for all infected clots (n=6), which was more than any of the other treatment arms. The results of this study demonstrate the potential of sonobactericide as a possible novel therapeutic strategy for treating infective endocarditis.
15 mins
Sasa Kenjeres, Joost Klip
Abstract: To be able to determine regions in the vascular system where initial development of atherosclerosis could take place is a first requirement of the mathematical modeling of the atheroma plaque formation and growth. In present study, we focus on development, validation and application of a comprehensive multi-component transport model in simplified multiple-layered arterial wall configurations under various blood flow conditions. The present mathematical model includes the mass transfer of following components: low-density lipoprotein (LDL), monocyte, oxidized low-density lipoprotein (OxLDL), macrophages, cytokines, foam cells, contractive smooth muscle cells, synthetic smooth muscle cells and collagen (as presented in Ref.[1]). In addition, the plaque growth is modeled through additional equation for dynamical evolution of the mass balance of foam cells, smooth muscle cells and collagen (main constitutes of the atheroma plaque). The multi-layered structure of the arterial wall is modeled to include endothelium, intima, internal elastic layer (IEL), media and adventitia, as in Refs.[2] and [3]. The direct coupling of mass transfer with blood flow is done through the WSS dependent transfer of LDL across the lumen/endothelium interface. The analysis starts with some simplified artery geometries with pre-specified stenotic extensions of 20% (mild) 40% (intermediate) and 60% (severe) of the characteristic radius of the artery. Different values of transmural pressure are considered ranging from 70 mmHg to 160 mmHg. The fully developed laminar velocity profiles are imposed at the inlet, which correspond to typical Reynolds numbers between 100 and 500 (to mimic blood flow in the carotid artery). The comparative assessment of LDL distributions within arterial wall with experimental measurements of Ref.[4] (for identical transmural conditions) showed good agreement. Next, in addition to LDL, the multi-solutal components are introduced and their spatial and temporal evolutions are analyzed. Finally, dynamical evolution of the volume of the atheroma plaque is solved. It is concluded that the present model is able to mimick atherosclerosis progression in a carotid artery from its initial development (healthy stage) up to large time intervals (10 years).
15 mins
Ali Akyldiz, Gustav Strijkers, Chen-Ket Chai, Cees Oomens, Frank Gijsen
Abstract: A major cause of ischemic cerebrovascular events is the atherosclerotic plaque rupture in a carotid artery [1]. Plaque rupture happens if plaque’s mechanical stability is lost [2]. The collagen network is an important determinant of tissues’ mechanical stability, however never assessed in plaques in 3D. This study aims to quantify the plaque collagen network. The collagen network in human carotid plaques (n=7, endarterectomy samples) were acquired ex-vivo with magnetic resonance diffusion tensor imaging (MR-DTI), performed with a 9.4 T horizontal scanner (Bruker Corp., Germany). A 3D spin echo sequence was used for imaging with unipolar diffusion sensitizing pulsed field gradients. Local collagen fiber orientation was qualitatively assessed with vIST/e (a tractography tool) from the MR-DTI data and histology from longitudinal plaque cross-sections. For quantitative analysis, predominant fiber alignments in 3D, and fibers’ azimuthal and elevation angles from 85 transversal cross-sections were identified from MR-DTI data. Qualitative assessment (Fig. 1) revealed two predominant fiber alignment patterns: one in the axial direction and one in the circumferential direction. This observation was confirmed by the quantitative assessment: ~35% of plaque fibers were aligned predominantly longitudinally, ~50% circumferentially and ~15% radially (Fig. 2). For both concentric (n=23) and eccentric (n=62) regions of plaques, the in-plane component of the fiber directions, identified through azimuthal angle, were mainly circumferential. Out-of-plane component, identified through elevation angle, showed two distinct orientations for concentric sections and was inclined towards in-plane orientation for eccentric ones. To the authors’ knowledge, this is the first study that imaged and quantified the collagen network in atherosclerotic plaques in 3D. The findings demonstrated both structured and unstructured fiber alignment in plaques. Although there is a general trend in predominant fiber alignment in circumferential and longitudinal directions, the fiber orientation is not uniform, which might greatly affect the mechanical stability of the plaque tissue. REFERENCES [1] C. Yuan et. al, Circulation, 105: 181-185 (2002) [2] G. C. Cheng et al, Circulation 87: 1179-1187 (1993)
15 mins
Pim van Ooij, Merih Cibis, Meike Vernooij, Aart Nederveen, Jolanda Wentzel
Abstract: Introduction: In this study we propose novel three-dimensional (3D) statistical methods to investigate and predict the relationship between wall shear stress (WSS) and wall thickness (WT) in atherosclerosis of the carotid bifurcation. Methods: Magnetic resonance imaging was used to measure 3D WT and 3D velocity averaged over the heart cycle in 11 asymptomatic individuals (mean age: 73±7 years, 6 women) with plaques in the carotid bifurcation [1]. 3D WSS was derived from the velocity data. Cohort-averaged 3D WSS and WT maps were created [2]. Linear regression was performed between the cohort-averaged WSS and WT maps to determine the slope (β) and intercept (ε). A ‘predicted’ patient-specific WT ((WT) ̂) map was created by calculation of (WT) ̂= β*WSS+ε at each point at the wall. A cohort-averaged 3D (WT) ̂ map was created. Spearman rho was determined for individual and cohort-averaged WSS and WT maps, for individual WT and (WT) ̂ maps and the cohort-averaged WT and (WT) ̂ maps. Results: The results are summarized in figure 1. The average rho of individual WSS and WT maps was -0.3±0.2, which, after the Fisher z-transformation, was significantly different from 0 (one-sample t-test, P<0.001). For the cohort-averaged WSS and WT maps rho was -0.7, indicating good co-localization of low WSS and high WT and vice-versa. For predicting (WT) ̂ the following equation was derived from the regression of the cohort-averaged WSS and WT maps: (WT) ̂=-1.0*WSS+1.8. The averaged rho for the comparison between (WT) ̂ and WT over all subjects was 0.3±0.2, which, after the Fisher z-transformation, was significantly different from 0 (one sample t-test, P<0.001). rho for the cohort-averaged 3D (WT) ̂ and WT maps was 0.7. Discussion/Conclusion: The significant co-localization of regions of low WSS and high WT in asymptomatic individuals indicates that WSS is an important factor in wall thickening in atherosclerotic disease. The application of presented techniques in more subjects may help to improve prediction of WT and aid in prediction of plaque progression over time. References: [1] Van Den Bouwhuijsen QJA et al. Determinants of magnetic resonance imaging detected carotid plaque components: The Rotterdam Study. Eur Heart J. 2012 [2] van Ooij P et al. A Methodology to Detect Abnormal Relative Wall Shear Stress on the Full Surface of the Thoracic Aorta Using 4D Flow MRI. Magn Res Med. 2015
15 mins
Emanuele Rondanina, Frans van de Vosse, Peter Bovendeerd
Abstract: Cardiac growth is a natural process through which heart enlarges, maintaining its proportions, due to an increase of blood flow demand of the growing body. This phenomenon is partially caused by the response of the cardiac myocytes to changes in local mechanical load, as induced by the changing demands of the body. Several diseases, like aortic stenosis or aortic regurgitation, can however result in pathological hypertrophy or dilation. Nevertheless it is not always clinically clear whether replacing the malfunctioning valve will lead to a reversal of this pathological growth. The aim of our research is to create a mathematical model to predict the cardiac growth output and the reversing process the heart might undergo after a surgery. This can be helpful to clinicians to understand how the disease will evolve and what will be the impact of surgery. The mechanical model [1] is composed of a truncated ellipsoidal left ventricle embedded in a lump parameter model of circulation. The myocardial tissue is defined by myofibers, capable of generating active force, which are embedded in a non-linear and transversely isotropic extracellular matrix. Moreover the myofiber orientation is taken in account by considering both longitudinal and transmural directions. A growth model [2], dependent on both local and global perturbations, works in symbiosis with the mechanical model. From the mechanical model a local load, proportional to the active work exerted by myofibers, is extracted and the difference between its homeostatic value is used as growth stimulus. Following a simple model, this stimulus is then translated in tissue volume change. Initial simulations to recover the geometry of a healthy heart revealed the need for adding a global haemodynamic feedback to obtain better physiological end states. We are currently working on improving the growth model [2] in this respect. In subsequent simulations we will investigate cardiac growth in case of valve pathologies.
15 mins
Ruoyu Xing, Astrid Moerman, Yanto Ridwan, Anton van der Steen, Frank Gijsen, Kim van der Heiden
Abstract: Wall shear stress (WSS) WSS is the frictional force that blood flow exerts on the endothelium of the vessel wall. It is known to initiate atherosclerotic plaque growth, which is the accumulation of lipids, fibrous tissue and inflammatory cells in the arterial wall. As the disease advances, plaques can progress and may become vulnerable. Vulnerable plaques are characterized by thin fibrous cap, large lipid pool and increased plaque inflammation. Rupture of a vulnerable plaque is the main cause of stroke and myocardial infarction. However, factors determining plaque progression and plaque vulnerability remain unclear. We hypothesized that WSS plays a role in atherosclerotic plaque progression and plaque stability. We study this correlation by 1) monitoring WSS over time during plaque growth in an atherosclerotic mouse model and 2) linking changes in WSS patterns to plaque composition. A tapering cast was placed around the right common carotid artery (RCCA) of ApoE-/- mice (n=7). As a result, a vulnerable plaque develops proximal to the cast. We analysed WSS over this growing plaque 5, 7, and 9 weeks after cast placement. Frist, detailed RCCA geometry was obtained using contrast-enhanced micro-CT. Vessel surface was reconstructed using an in-house developed segmentation protocol. Next, we measured blood velocity and vessel diameter in RCCA by Doppler Ultrasound to determine flow rate. Finally, we computed WSS in RCCA. At week 9, animals were sacrificed for histological analysis of plaque composition. At week 5, a lumen-intruding plaque was visible proximal to the cast with low WSS of 4.6 ± 1.9 Pa. Over time, plaque progressed and further intruded into the lumen, accompanied by decreasing WSS of 4.0 ± 1.3 Pa at week 7 and 3.1 ± 3.3 Pa at week 9. Histological analysis revealed that plaque composition at week 9 varied at different WSS locations. Circumferentially-averaged WSS along RCCA was co-registered with local plaque composition. Data showed that low WSS at week 7 can significantly predict accumulation of macrophages in plaque (R = -0.32, p<0.05). Large lipid pool were positively associated with high WSS (R = 0.34, p<0.05). Besides, high WSS was associated with thinner fibrous caps (R = -0.22, p<0.05). In this study, we demonstrated for the first time how WSS patterns changed over time during plaque progression. We established causal relationship between WSS and plaque composition. Our results suggested that accumulation of macrophages in low WSS regions were related to plaque progression. A larger necrotic core area and thinner fibrous cap were found in high WSS regions, indicating the link between high WSS and plaque vulnerability.