New ultrasonographic approaches to …kth.diva-portal.org/smash/get/diva2:280837/FULLTEXT01.pdfNew...
Transcript of New ultrasonographic approaches to …kth.diva-portal.org/smash/get/diva2:280837/FULLTEXT01.pdfNew...
New ultrasonographic approaches to
monitoring cardiac and vascular function
Anna Bjällmark
Doctoral Thesis TRITA‐STH Report 2009:7
Division of Medical Engineering, School of Technology and Health, KTH (Royal Institute of Technology)
Academic dissertation which with permission from Kungliga Tekniska Högskolan (Royal Institute of Technology) in Stockholm is presented for public review for passing the doctoral examination on Friday January 22, 2010, at 13.00. In lecture hall 3‐221, Alfred Nobels allé 10, Huddinge, Sweden. TRITA‐STH Report 2009:7 ISSN: 1653‐3836 ISRN: KTH /STH/‐‐09:7‐SE ISBN: 978‐91‐7415‐525‐9 © Anna Bjällmark, Stockholm 2009
i
Summary
Atherosclerotic cardiovascular disease is the leading cause of death worldwide. To decrease mortality and morbidity in cardiovascular disease, the development of accurate, non‐invasive methods for early diagnosis of atherosclerotic cardiac and vascular engagement is of considerable clinical interest. Cardiovascular ultrasound imaging is today the cornerstone in the routine evaluation of cardiovascular function and recent development has resulted in two new techniques, tissue velocity imaging (TVI) and speckle tracking, which allow objective quantification of cardiovascular function. TVI and speckle tracking are the basis for three new approaches to cardiac and vascular monitoring presented in this thesis: wave intensity wall analysis (WIWA), two‐dimensional strain imaging in the common carotid artery, and the state diagram of the heart. WIWA uses longitudinal and radial strain rate as input for calculations of wave intensity in the arterial wall. In this thesis, WIWA was validated against a commercially available wave intensity system, showing that speckle tracking‐derived strain variables can be useful in wave intensity analysis. WIWA was further tested in patients with end stage renal disease and documented high mortality in cardiovascular disease. The latter study evaluated the effects of a single session of hemodialysis using WIWA and TVI variables and showed improved systolic function after hemodialysis. The results also indicated that preload‐adjusted early systolic wave intensity obtained by the WIWA system may contribute in the assessment of left ventricular contractility in this patient category. Two‐dimensional strain imaging in the common carotid artery is a new approach showing great potential to detect age‐dependent differences in mechanical properties of the common carotid artery. Among the measured strain variables, global circumferential strain had the best discriminating performance and appeared to be superior to conventional measures of arterial stiffness such as elastic modulus and β stiffness index. The state diagram is a visualisation tool that provides a quantitative overview of the temporal interrelationship of mechanical events in the left and right ventricles. Case examples and a small clinical study showed that state diagrams clearly visualize cardiac function and can be useful in the detection of non ST‐elevation myocardial infarction (NSTEMI). Even though WIWA, two‐dimensional strain imaging in the common carotid artery and the state diagram show potential to be useful in the evaluation of cardiovascular function, there still remains a considerable amount of work to be done before they can be used in the daily clinical practice.
ii
iii
List of publications
The thesis is based on the following papers: I. Wave intensity wall analysis: a novel noninvasive method to measure wave intensity. Matilda Larsson, Anna Bjällmark, Britta Lind, Rita Balzano, Michael Peolsson, Reidar Winter, Lars‐Åke Brodin.
Heart and Vessels. 2009;24(5):357‐65.
II. Effects of hemodialysis on the cardiovascular system: Quantitative analysis using wave intensity wall analysis and tissue velocity imaging.
Anna Bjällmark, Matilda Larsson, Jacek Nowak, Britta Lind, Shirley Yumi Hayashi, Marcelo Mazza do Nascimento, Miguel Riella, Astrid Seeberger, Lars‐Åke Brodin.
Submitted to Heart and Vessels.
III. Ultrasonographic strain imaging is superior to conventional non‐invasive measures of vascular stiffness in the detection of age‐dependent differences in the mechanical properties of the common carotid artery.
Anna Bjällmark, Britta Lind, Michael Peolsson, Kambiz Shahgaldi, Lars‐Åke Brodin, Jacek Nowak. Submitted to European Journal of Echocardiography.
IV. State diagrams of the heart ‐ a new approach to describing cardiac mechanics. Matilda Larsson, Anna Bjällmark, Jonas Johnson, Reidar Winter, Lars‐Åke Brodin, Stig Lundbäck. Cardiovascular Ultrasound. 2009;7:22.
iv
v
Division of work between authors
I. Bjällmark, Larsson and Brodin made the outline of the study. The image acquisition was performed by Bjällmark, Larsson, Balzano and Lind. Bjällmark and Larsson performed the software development and the analysis of the patient data. Bjällmark and Larsson wrote the manuscript, and Winter critically reviewed and contributed with his clinical experience to the preparation of the manuscript. Peolsson and Brodin supervised the study.
II. Bjällmark, Larsson, Lind, Hayashi, Seeberger and Brodin designed the study. Riella and Mazza do
Nascimento organized the study setup. Bjällmark, Larsson and Brodin participated in the interpretation of the results and prepared the manuscript. Lind and Hayashi performed the ultrasound examinations and Bjällmark and Larsson performed the offline analysis. Bjällmark and Larsson wrote the manuscript, reviewed by Seeberger, Hayashi, Mazza do Nascimento, Brodin and Nowak.
III. Bjällmark and Nowak designed the study, participated in the interpretation of the results and
prepared the manuscript. Shahgaldi and Lind performed the ultrasound examinations. Bjällmark and Shahgaldi analyzed the vascular and cardiac data respectively. Bjällmark and Lind performed the intra‐ and inter‐observer variability study and Peolsson performed the principal component analysis. Bjällmark wrote the manuscript, critically reviewed by Nowak. Brodin contributed to the interpretation of the data and reviewed the manuscript.
IV. Bjällmark, Larsson, Johnson, Lundbäck and Brodin participated in contributions to conception,
analysis and interpretation of data. Software development and analysis of patient data was performed by Bjällmark and Larsson. Bjällmark and Larsson wrote the manuscript, critically reviewed by Johnson and Lundbäck. Winter was the supervisor of echo examinations and carefully reviewed the manuscript.
vi
vii
Scientific contributions not included
Patent:
Global and local detection of blood vessel elasticity (WO/2008/002257). Lars‐Åke Brodin, Håkan Elmqvist, Anna Bjällmark, Matilda Larsson.
Scientific publications:
I. Differences in myocardial velocities during supine and upright exercise stress echocardiography
in healthy adults. Anna Bjällmark, Matilda Larsson, Kambiz Shahgaldi, Britta Lind, Reidar Winter, Lars‐Åke Brodin. Clinical Physiology and Functional Imaging. 2009;29(3):216‐23.
II. Velocity tracking, a new and user independent method for detecting regional function of the left ventricle.
Carl Westholm, Anna Bjällmark, Matilda Larsson, Per Jacobsen, Lars‐Åke Brodin, Reidar Winter. Clinical Physiology and Functional Imaging. 2009;29(1):24‐31.
III. Fixation identification in centroid versus start‐point modes using eye‐tracking data. Torbjörn Falkmer, Joakim Dahlman, Tanja Dukic, Anna Bjällmark, Matilda Larsson. Perceptual and Motor Skills. 2008;106(3):710‐24.
IV. Velocity tracking ‐ a novel method for quantitative analysis of longitudinal myocardial function. Anna Bjällmark, Matilda Larsson, Reidar Winter, Carl Westholm, Per Jacobsen, Britta Lind, Lars‐Åke Brodin.
Journal of the American Society of Echocardiography. 2007;20(7):847‐56.
Conference abstracts:
I. The prevalence of right ventricular dysfunction and the acute effect of HD on right ventricular function in CKD.
Shirley Yumi Hayashi, Marcelo Mazza do Nascimento, Britta Lind, Miguel Riella, Matilda Larsson, Anna Bjällmark, Astrid Seeberger, Jacek Nowak, Bengt Lindholm, Lars‐Åke Brodin.
International society of blood purification, Stockholm 2009.
viii
II. The prevalence of intraventricular dyssynchrony, detected by tissue synchronization imaging, in hemodialysis, peritoneal dialysis and chronic kidney disease stages 3 and 4.
Shirley Yumi Hayashi, Marcelo Mazza do Nascimento, Britta Lind, Miguel Riella, Matilda Larsson, Anna Bjällmark, Astrid Seeberger, Jacek Nowak, Bengt Lindholm, Lars‐Åke Brodin.
American Society of Nephrology ‐ Renal week, San Diego 2009.
III. Improvement of left ventricular synchronicity, assessed by tissue synchronization imaging, after a single hemodialysis session in chronic hemodialysis patients.
Shirley Yumi Hayashi, Marcelo Mazza do Nascimento, Britta Lind, Miguel Riella, Matilda Larsson, Anna Bjällmark, Astrid Seeberger, Jacek Nowak, Bengt Lindholm, Lars‐Åke Brodin.
American Society of Nephrology ‐ Renal week, San Diego 2009.
IV. Velocity tracking – a new user independent method for bedside detection of myocardial ischemia.
Carl Westholm, Matilda Larsson, Anna Bjällmark, Reidar Winter, Per Jacobsen, Lars‐Åke Brodin. EuroEcho, Lisbon 2007.
V. Color coded tissue Doppler is more accurate and less sensitive to filtering and gain settings compared to spectral tissue Doppler ‐ A comparison of two commonly used tissue doppler techniques in the clinical setting.
Aristomenis Manouras, Anna Bjällmark, Reidar Winter, Lars‐Åke Brodin. EuroEcho, Lisbon 2007.
VI. Wave Intensity Wall Analysis ‐ A novel noninvasive method for early detection of cardiovascular disease.
Matilda Larsson, Anna Bjällmark, Britta Lind, Rita Balzano, Mikael Peolsson, Reidar Winter, Lars‐Åke Brodin.
EuroEcho, Lisbon 2007.
VII. A new graphical user interface module generating state diagrams of the heart. Anna Bjällmark, Matilda Larsson, Stig Lundbäck, Jonas Johnsson, Reidar Winter, Lars‐Åke Brodin. EuroEcho, Prague 2006.
VIII. Bull’s eye presentation of speckle tracking data; a simple and user independent method for detection of regional myocardial ischemia.
Per Jacobsen, Reidar Winter, Anna Bjällmark, Matilda Larsson, Magnus Nygren, Carl Westholm, Lars‐Åke Brodin.
EuroEcho, Prague 2006.
IX. To recognize faces and expressions ‐ what is the nature of the problem for people with Asperger Syndrome?
Matilda Larsson, Anna Bjällmark, Marita Falkmer, Torbjörn Falkmer. Autism Safari, 2nd World Autism Congress & Exhibition, Cape Town 2006.
X. Velocity Tracking ‐ a novel method for quantitative analysis of longitudinal myocardial function. Matilda Larsson, Anna Bjällmark, Reidar Winter, Carl Westholm, Per Jacobsen, Britta Lind, Lars‐Åke Brodin.
World Congress of Biomechanics, Munich 2006.
ix
Abbreviations
2D Two‐dimensional 3D Three‐dimensional A’ Late diastolic wave AV Atrioventricular BP Blood pressure CCA Common carotid artery DDP Dynamic displacement pump DISP Displacement E’ Early diastolic wave ECG Electrocardiogram Ep Elastic modulus ESRD End stage renal disease FR Frame rate HD Hemodialysis IMT Intima media thickness IVCT Isovolumic contraction time IVCV Isovolumic contraction velocity IVRT Isovolumic relaxation time IVRV Isovolumic relaxation velocity LV Left ventricle LVEF Left ventricular ejection‐fraction NA Negative area NSTEMI Non ST‐elevation myocardial infarction PCA Principal component analysis PSV Peak systolic velocity PWV Pulse wave velocity ROI Region of interest RV Right ventricle TVI Tissue velocity imaging W1 Wave intensity forward compression wave W2 Wave intensity forward decompression wave WI Wave intensity WIWA Wave intensity wall analysis
x
xi
Acknowledgements
The work presented in this thesis has been performed at the Division of Medical Engineering, School of Technology and Health, KTH (Royal Institute of Technology). The work has been performed in cooperation with the Department of Clinical Physiology, Karolinska University Hospital, the Division of Baxter Novum and Renal Medicine, Department of Clinical Science, Intervention and Technology, Karolinska Institute and Evangelical Medical School, Curitiba, Brazil. First of all I would like to thank Matilda Larsson, with whom I worked closely during these years. Thank you for being the most excellent co‐worker, and also for being the sweetest friend and a terrific travelling partner. Many thanks also to my supervisor Lars‐Åke Brodin for all your ideas, enthusiasm and for believing in every project. I would further like to express my gratitude to Jacek Nowak for valuable input, interesting discussions and for carefully reviewing my writing, Reidar Winter for teaching me cardiology and scientific writing, to Britta Lind for helping with the image acquisition and for answering my millions of questions about clinical ultrasound, to Shirley Yumi Hayashi and Marcelo Mazza do Nascimento for taking such good care of us when working in Brazil, to Aristomenis Manouras and Kambiz Shahgaldi for always smiling and for helping out with the image acquisition even at very short notice, to “the Professor” and “the Docent” for statistical advices and for introducing me to the lovely world of Knob Creek, to Gunilla Langvall, Monika Armuand and Karl‐Erik Jönsson for providing the best support for the employees, to Staffan Larsson for great help with Photoshop and my aching shoulder, to Michael Peolsson for pleasant research meetings with scones and tea in the city centre, and to Peta Sjölander for correcting the English. Many thanks to Stig Lundbäck and Jonas Johnson for teaching me about the dynamic displacement pump, and also for the many laughs during the long discussions about the deltaV‐function. I would like to thank my dear colleagues in the Imaging Group (Dennis, Frida, Nina and Mattias) and the Neuronics Group (Maria, Mats, Johnson, Sofia, Axel, Elin, Rickard, Daniel, Xiaogai, Madelen, Kim, Tobias, Svein and Peter) for creating a friendly and inspiring atmosphere, making the working days a pleasure. A special thanks to my friends for always being supportive and for enriching my life with late nights at Karlssons bar in Jönköping or at Harrys in Falkenberg, fika at Åbacka Café, Graz 2003/2004‐trips, long telephone conversations, fantastic time on and around the basketball court (which has since changed to more relaxed environment in the sand) and for interesting discussions about everything
xii
and nothing. Many thanks also to Britt‐Marie for being the loveliest aunt. Finally, the greatest thanks to my parents Peter and Eva, and to my brothers Per and Anders and their families for love, encouragement and for giving me the best support.
xiii
Table of contents
Summary .......................................................................................................................................................i List of publications ......................................................................................................................................iii Division of work between authors...............................................................................................................v Scientific contributions not included.........................................................................................................vii Abbreviations .............................................................................................................................................. ix Acknowledgements.....................................................................................................................................xi Introduction ................................................................................................................................................ 1
1.1. The heart......................................................................................................................... 1 1.2. The cardiovascular system.............................................................................................. 3
1.2.1. The movement of the artery................................................................................... 4 1.3. Methods for evaluation of cardiac function ................................................................... 5
1.3.1. Tissue velocity imaging (TVI)................................................................................... 5 1.3.2. Speckle tracking ...................................................................................................... 7
1.4. Methods for evaluation of vascular function ................................................................. 8 1.4.1. Arterial stiffness ...................................................................................................... 8 1.4.2. Assessment of subclinical atherosclerosis .............................................................. 9 1.4.3. Wave intensity analysis......................................................................................... 10
Aims........................................................................................................................................................... 13 Methods and materials............................................................................................................................. 15
3.1. Wave intensity wall analysis ......................................................................................... 15 3.2. Two‐dimensional strain imaging in the common carotid artery .................................. 17 3.3. State diagram................................................................................................................ 17 3.4. Studied populations ...................................................................................................... 19 3.5. Vascular imaging ........................................................................................................... 20
xiv
3.5.1. Ultrasonographic equipment and image acquisition ........................................... 20 3.5.2. Wave intensity variables (Studies 1 and 2)........................................................... 21 3.5.3. Short‐axis speckle tracking variables and established measures of arterial stiffness (Study 3).................................................................................................................. 21
3.6. Echocardiography ......................................................................................................... 22 3.6.1. Echocardiography equipment and image acquisition .......................................... 22 3.6.2. Tissue Doppler variables (Studies 2‐4).................................................................. 23
3.7. Statistical analysis ......................................................................................................... 23 3.8. Summary of the methodology for each study.............................................................. 24
Results ....................................................................................................................................................... 27 Discussion.................................................................................................................................................. 31
5.1. Wave intensity wall analysis ........................................................................................ 31 5.2. Two‐dimensional strain imaging................................................................................... 33 5.3. Limitations of arterial measurements .......................................................................... 34 5.4. State diagram................................................................................................................ 35
Conclusions ............................................................................................................................................... 39 References ................................................................................................................................................ 41
‐ 1 ‐
Introduction
Atherosclerotic cardiovascular disease has become the leading cause of death worldwide [1]. In view of the rapid increase in incidence and prevalence of this disease, its human and economic impact can be expected to become even more significant. One way to minimize the mortality and the morbidity of cardiovascular disease is to increase the awareness of the population about important risk factors such as tobacco [2] and obesity [3]. However, a significant number of cardiovascular events occur in individuals without any presence of risk factors [4]. Therefore, the development of accurate, non‐invasive methods for early diagnosis and treatment is of considerable clinical interest. Ultrasound is a well‐established technique for imaging of the cardiac and vascular system. Its non‐invasiveness, high temporal and spatial resolution and mobility make the technique very useful in the daily clinical practice. Ultrasound in cardiac imaging was applied for the first time in 1953 by Edler and Hertz [5], who showed that anatomy and motions of the heart were detectable using a pulse‐echo system. Since then, a remarkable development of clinical ultrasound has taken place and today it is possible to obtain detailed anatomic images in 3D and parametric images showing myocardial tissue movements and deformation. Tissue velocity imaging (TVI) and speckle tracking are two relatively new ultrasound‐based techniques, primarily used for quantification of myocardial function.
1.1. The heart
The human heart is approximately the size of a fist and is located in the frontal part of the chest behind the sternum. In a normal person, the heart weighs between 250g and 300g. The heart is divided into four cardiac chambers, the left and right atrium and the left and right ventricle. The left ventricle pumps the oxygenated blood through the aortic valve into the systemic circulation to supply all the organs of the body with blood. When ejected from the heart, the blood flows into the aorta and is directed further into the arteries. The blood returns from the periphery via the low pressure venous system to the right atrium. The blood is moved forward through the tricuspid valve to the right ventricle and is then pumped through the pulmonary valve into the pulmonary arteries to be oxygenized in the lungs. Back from the lungs, the oxygenated blood re‐enters the heart through the left atrium. Through the mitral valve, the blood reaches the left ventricle and the pump cycle continues. Figure 1 illustrates the anatomy of the heart.
‐ 2 ‐
Figure 1. The anatomy of the heart [6]. The right ventricle of the heart only needs to generate the pressure required to pump the blood through the lungs. The left ventricle on the other hand is required to generate enough pressure to pump the blood to every other part of the body. This is why the left ventricle has more cardiac muscle tissue (myocardium) in its walls. The left ventricle has an approximately parabolic shape and its walls are referred to as the anteroseptal, the anterior, the lateral (antero‐lateral), the posterior (infero‐lateral), the inferior, and the septal (infero‐septal) walls. The part of the ventricle that is located closest to the atrium is known as the base of the ventricle. The lowest end of the heart is called the apex. The ventricular wall itself consists of three layers, the epicardium (external), the myocardium (middle) and the endocardium (internal). The myocardium is the cardiac muscle tissue, responsible for the pumping function of the heart, while the other two layers serve as slippery and smooth connecting tissue to the surroundings. The myocardial fibres have a helical arrangement with subepicardial fibres directed in an anti‐clockwise spiral and subendocardial fibres in a clockwise spiral from apex to base. The subepicardial fibres are predominately longitudinally oriented, transitioning to a horizontal fibre direction in the midwall, and becoming again longitudinally oriented in the subendocardial region although in the opposite direction from the subepicardial fibres. Because of the geometry of the muscle fibres the lengthening and shortening of the myocardium result in rotational movements [7, 8]. The pumping and regulating function of the heart has been differently described over the years [9‐13]. In 1986, Lundbäck refined a model of cardiac mechanics [12] described already by Leonardo da Vinci [14] implying that the heart pumps with forward and backward movements in the longitudinal direction of a piston‐like unit, referred to as the AV‐piston or the spherical AV‐plane. When the AV‐plane is drawn towards the apex of the heart, forcing the blood in the ventricles into the pulmonary and systemic circulation, this will at the same time create an inflow to the atria and its auricles. During the cardiac cycle, the outer contours of the heart are more or less unchanged, which is energy‐saving as there is no need for moving surrounding tissue. The regulating function is mostly performed by the forward and backward movements of the inter‐ventricular septum. This movement is determined by the pressure differences between the left and right ventricles [12].
‐ 3 ‐
The pumping function of the heart is electrically activated. The electrical activation starts in the sinus node located in the right atrium, and spreads in the atria. The impulse also reaches the atrioventricular (AV) node located in the septum between the atria. The cells of the AV node slow down the conduction before the impulse is transmitted to the ventricles giving the atria time to finish their contraction. The impulse passes along the bundle of His and through conduction system fascicles terminating in a set of branching Purkinje fibres which penetrate into the myocardium. These fibres allow for an almost instantaneous transmission of the cardiac impulse throughout the ventricular muscle tissue resulting in a more or less synchronous depolarization of the myocardium and subsequent contraction of the ventricle. The cardiac cycle is traditionally divided into a systolic phase and a diastolic phase. During systole, blood is ejected from the ventricles into the pulmonary and systemic circulation and during diastole the ventricles are refilled with blood. The systolic phase can be further subdivided into isovolumic contraction and ejection, whereas the diastolic phase can be subdivided into isovolumic relaxation, early diastolic filling, diastase and late diastolic filling (atrial contraction). Preload, afterload and contractility determine cardiac performance. Preload can be defined as the initial stretching of the myocardium prior to contraction. A greater preload on the cardiac muscle increases its force of contraction. According to the Frank‐Starling law, the more the heart fills with blood during diastole (within limits), the stronger is the power of the ventricular work during systole. For a healthy heart, preload may be defined as the ventricular end‐diastolic volume, since increased volume stretches the myocardium [15]. Afterload is the force that the ventricle must overcome, in order to eject the blood into the aorta. If the afterload is increased, then the heart has to generate higher intra‐ventricular pressure to be able to open the aortic valve and to maintain the blood flow during ventricular ejection [16]. Contractility is the intrinsic ability of the myocardium to contract, independent of changes in preload, afterload and heart rate. Increased contractility is associated with increased velocity of the myocardial fibre shortening and a steeper pressure rise when preload, afterload and heart rate are constant [17].
1.2. The cardiovascular system
The systemic circulation supplies the organs of the body with oxygenated blood and nutrients. The systemic circulation is composed of distribution vessels (arteries) with elastic properties and low resistance to blood flow, resistance vessels (arterioles) with less elasticity and high resistance, exchange vessels (capillary bed) and capacitance vessels (veins and venules) with high distensibility. The distribution vessels convert the pulsating flow generated by the heart into a more continuous flow. The resistance vessels regulate the blood pressure and distribute the blood flow to the organs, while the exchange of gases and metabolites takes place in the exchange vessels. The capacitance vessels contain about two thirds of the total blood volume, and have the ability to constrict, thus increasing the inflow to the heart. The blood vessels, with the exception of the exchange vessels, consist of three layers. The most inner layer is the tunica intima, followed by tunica media and the tunica adventitia. Tunica media constitutes the smooth muscle layer of the vessel wall and has the ability to contract and relax resulting in vascular constriction or dilation.
‐ 4 ‐
The blood flow is regulated according to the following equation:
RP
QΔ
= , Q = blood flow, ∆P= difference in pressure, R = peripheral resistance.
The peripheral resistance is calculated from:
4rl
R×η
= , η = viscosity of blood, l = length of lumen segment, r = lumen radius.
1.2.1. The movement of the artery
The longitudinal direction of the artery is defined as the direction of the mean blood flow and the radial direction is orthogonal to the longitudinal direction. Most methods used for characterization of artery properties are based upon measurements in the radial direction and relate to simultaneous assessment of diameter and pressure waveforms. The diameter and pressure waveforms can be obtained using invasive measurements, but such measurements have limited clinical usage. Non‐invasively, the diameter waveform can be accurately obtained using ultrasound [18]. However, the instantaneous measurement of the pressure waveform suffers from accuracy problems. Several methods for non‐invasive assessment of the pressure waveform have been developed. The pressure wave form can be obtained from the arterial diameter wave form by using an exponential relationship between the arterial cross‐section and the pressure [19]. It has also been claimed that the arterial diameter waveform is similar to the pressure waveform and a direct transformation between the two could be made with calibration of the peak and bottom values against systolic and diastolic blood pressure [20]. Radial forces induce a circumferential tension in the arterial wall. The circumferential movement is further dependent on the radius and on the thickness of the wall. Figure 2 illustrates how the different directions are defined in ultrasound images of the artery.
Figure 2. Definition of radial, longitudinal and circumferential direction in the long‐axis (left) and short‐axis views (right) of the common carotid artery.
Compared with radial function, the longitudinal movement of the artery has gained little attention hitherto. The general belief has been that the longitudinal movement of the aortic wall is negligible and mainly due to respiratory movements of the diaphragm [21]. Limitations of imaging techniques with too low temporal and spatial resolution were probably one of the reasons for this belief. With the use of different speckle tracking techniques, a distinct longitudinal movement of the inner layer
‐ 5 ‐
of the arterial wall has been observed during the heart cycle [22, 23]. The longitudinal movement has been shown to correlate with the blood flow next to the wall [24] and has approximately the same magnitude as the radial movement [25]. There is also a shear strain in the arterial wall occurring due to the larger movements of the intima‐media compared with those of the adventitia [26]. When considering movements of the artery, it should be kept in mind that arteries are not homogenous, cylindrical tubes and the mechanical behaviour can vary in different parts of the artery.
1.3. Methods for evaluation of cardiac function
Within ultrasound imaging, tissue velocity imaging (TVI) and speckle tracking are the most recent developments with the ability to quantify myocardial motion and deformation patterns in the myocardium.
1.3.1. Tissue velocity imaging (TVI)
The Doppler principle states that if an object moves away from a sound source, the reflected frequency will be lower than the transmitted frequency. Conversely, if the object is moving towards the sound source the reflected frequency will be higher than the transmitted frequency. The difference in frequency is known as the Doppler shift and is used in clinical ultrasound for estimating the velocity of moving tissue or blood cells. The Doppler equation has the following formula:
fc
)cos(v2f
α=Δ
where ∆f is the Doppler shift, f is frequency of the transmitted ultrasound (1‐15MHz), v is the velocity of the moving object, α is the angle between the moving object and the ultrasound beam and c is the speed of sound. Both tissue and blood are moving in the heart, and motion of the highly reflective slow‐moving myocardium is differentiated from low reflective blood moving with high velocity by application of amplitude and frequency filters. Thus, the Doppler technique facilitates a quantitative assessment of wall motion and velocities from an ultrasound investigation. The first attempt to assess wall motion dynamics was performed in 1989 using pulsed Doppler acquisition at a single measurement site [27], and further development resulted in two‐dimensional imaging of myocardial velocities presented as colour‐coded maps [28]. The development continued with the introduction of the TVI‐technique [29, 30], which allowed for extraction of velocity profiles in multiple selectable positions in the heart muscle. The TVI‐derived velocity information could be extracted with high temporal resolution (> 200 Hz), which was achieved by using a limited number of scan lines (8‐16 scan lines). The velocity curve is the cornerstone for all further analysis and post‐processing. Displacement is obtained by integration of the velocity curve. By relating myocardial velocities in two different locations to the distance between them, a curve reflecting the velocity of myocardial deformation (strain rate) is generated. When integrating the strain rate curve, information describing the
‐ 6 ‐
magnitude of myocardial deformation (strain) is obtained. All these variables can be colour‐coded and overlaid on 2D and 3D images (Figure 3), or presented as a function of time from specified regions in the images.
Figure 3. TVI images with different colour maps. From left: velocity (cm/s), displacement (mm), strain (%), strain rate (1/s).
The advantage of TVI compared with conventional echocardiography is the possibility to quantify cardiac function by providing detailed regional information about myocardial velocity, displacement and strain pattern. The TVI concept has been used for evaluation of cardiac function and reported in a large number of publications, where quantitative stress‐echocardiography was one of the first clinically useful applications [31]. Amplitude values of different peaks in tissue Doppler curves have been studied in various patient populations and reference values have been established [31‐34]. The clinically most useful information in a myocardial velocity curve is the peak systolic velocity (PSV) and the two diastolic velocities, the first (E’) occuring during early diastolic filling and the second (A’) during atrial contraction. Beside the main systolic and diastolic events, there are rapid movements of short duration associated with the pre‐ and postsystolic reshaping of the ventricles that occurs during the transition between filling and ejection of the ventricles. The isovolumic contraction velocity (IVCV) is the maximal velocity during the isovolumic contraction period and the isovolumic relaxation velocity (IVRV) is the maximal velocity during the isovolumic relaxation period. In the clinical routine, ventricular displacement is most often used for evaluation of the atrioventricular plane (AV‐plane) movement during ventricular ejection. Movements that occur during the isovolumic contraction and isovolumic relaxation periods are subsequently excluded. From the myocardial velocity curve or displacement curve, different time intervals can be defined. Two commonly used time intervals are the isovolumic contraction time (IVCT) and the isovolumic relaxation time (IVRT). Prolongation of these time intervals is associated with cardiac dysfunction [35‐37]. Figure 4 illustrates commonly measured tissue velocity and displacement variables and isovolumic time intervals obtained in the tissue velocity curve.
‐ 7 ‐
Figure 4. Typical tissue velocity echocardiographically derived velocity and displacement profiles from the basal lateral wall during one cardiac cycle. IVCT = isovolumic contraction time, IVCV = isovolumic contraction velocity, PSV = peak systolic velocity, IVRT = isovolumic relaxation time, IVRV = isovolumic relaxation velocity, E’ = maximal tissue velocity during early diastolic filling, A’ = maximal tissue velocity during atrial contraction.
TVI‐derived strain and strain rate have been shown to be useful in the detection of cardiac dysfunction. However, these variables will not be considered in this thesis.
1.3.2. Speckle tracking
The most recent development in ultrasonographic imaging is called speckle tracking. Every ultrasound image has a characteristic black and white speckle pattern, which is the basis for this new diagnostic approach. The pattern is a result of interference occurring from the reflected ultrasound. Constructive interference (waves in phase) is seen as white spots, while destructive interference (waves out of phase) is seen as black spots. The speckle pattern is relatively stable over the cardiac cycle and the randomness of the speckle pattern ensures that each region of the myocardium has a unique speckle pattern, which can be used to differentiate one region from another (Figure 5).
‐ 8 ‐
Figure 5. Grey‐scale image of the left ventricle. The two enlarged areas from different parts of the ventricle each show unique speckle patterns, which can be tracked from frame to frame [38].
By comparing amplitude patterns of speckles in a small region (kernel) with the most similar speckle pattern within a search region in the succeeding frame, motion and deformation of the myocardial wall can be obtained in two directions without any angle dependency. Due to rotation, torsion and deformation, the speckle pattern is not perfectly repeated. This requires a high frame rate to capture small frame‐to‐frame changes. However, line density needs to be maintained in order to have a good spatial resolution and therefore the frame rate obtained with speckle tracking is lower compared with TVI acquisitions. Several speckle tracking algorithms have been developed and a number of validation studies have been performed [39‐44]. The technique has successfully been applied in different cardiac applications [43, 45, 46] and has been shown to be particularly useful in the assessment of regional short axis rotation patterns [47‐49]. A commercially available technique is the 2D‐strain software package from GE Vingmed (GE Healthcare, Horten, Norway) [39, 50].
1.4. Methods for evaluation of vascular function
There are several methods available for the assessment of vascular function. The aim of all these methods is to quantify the degree of arterial stiffness in the systemic circulation.
1.4.1. Arterial stiffness
Arterial stiffness is one of the most important determinants of increased blood and pulse pressure (systolic blood pressure‐diastolic blood pressure) and thus the cause of cardiovascular events, including myocardial infarction and left ventricular hyperthrophy [51‐53]. Arterial stiffness influences the ability to accommodate the blood ejected from the heart. During systole, the blood pressure rises and the diameter of the arteries increases which, during diastole, decreases again. The blood volume stored during systole is released during diastole, creating a continuous flow
‐ 9 ‐
towards the periphery of the vascular system. The ability of the arteries to distend in response to a given pressure is given by the compliance (C):
ΔPΔV
C = , ∆V = change in volume, ∆P = change in pressure
The elastic properties of the arteries have a large impact on the blood pressure and the afterload, where increased stiffness causes a reduction in distensibility and thereby elevates the systolic blood pressure and the pulse pressure, and increases the mechanical load on the heart. Additionally, arterial stiffening increases the speed of the pulse travelling along the arteries and consequently the return of the wave reflections from the peripheral parts of the vascular system occurs prematurely. The premature return causes changes in the pressure wave contours and may further disturb ventricular ejection and coronary perfusion. Ageing is accompanied by changes in arterial structure and function. Several studies, using various assessment techniques, have verified an age‐dependent increase of arterial stiffness in proximal, elastic arteries [54‐56]. The increase in arterial stiffness is considered to be largely a result of progressive degeneration of elastin fibres in the arterial media layer [57]. The same factors that underlie age‐associated structural and functional changes in the arteries have shown to be present in the pathogenesis of cardiovascular disease [58] and, interestingly, several studies published in recent years have shown that decreased elasticity of the arterial wall may be present even before the appearance of any clinical symptoms or atherosclerotic plaques [59‐61].
1.4.2. Assessment of subclinical atherosclerosis
There are several techniques available for the detection of subclinical changes of the mechanical properties of the artery. As mentioned earlier, non‐invasive methods are mainly based upon pulse transit time, analysis of arterial pressure pulse and its wave form or stiffness estimation from distending pressure and diameter measurements. Pulse wave velocity (PWV) [62] is a measure of the speed of which the pulse pressure wave travels along the arteries. The pulse pressure wave travels faster in proportion to the stiffness of the artery. Measurements of PWV can be done by measuring the pressure at two locations in the arterial system. By relating the delay between the two wave forms to the measured distance, a value of PWV is obtained. Intima media thickness (IMT) is defined as the distance between the lumen‐media interface and the media‐adventitia interface [63], and increased IMT is considered to be a valuable marker of early atherosclerosis. However, IMT is influenced by sex, age and race, which makes cut‐off values difficult to define. Local measurements of arterial stiffness can be expressed as distensibility, compliance, elastic modulus (Ep) or β stiffness index and are based upon measurements of radial movements. Even though these measurements are widely used, the validity and reproducibility procedures of the measurements show large differences, and there is, in fact, no golden standard for evaluating arterial stiffness locally [64, 65]. Much of the recent research has focused on development of methods to quantify deviations from normality, but there is still no method free of limitations available today.
‐ 10 ‐
1.4.3. Wave intensity analysis
Wave intensity analysis was introduced as an alternative method in the study of cardiovascular dynamics [66]. Ventricular ejection generates a wave that propagates throughout the circulation. The characteristics of the wave are highly dependent on the performance of the ventricle, but also of the mechanical properties of the arteries as it travels distally. Wave intensity (WI) is defined as the product of blood pressure (dP) and mean blood velocity (dU) changes (WI = dP * dU). It is a measure of the energy flux per area unit [W/m2] transported by the waves and provides information about the ventricular mechanical performance and ventricular‐arterial interaction [66]. A positive value of the wave intensity indicates that the waves travelling towards the periphery of the vascular bed are dominant, while a negative value of the wave intensity indicates that the waves originated from the peripheral site are dominant. One advantage of wave intensity analysis compared with traditional Fourier analysis, is that no assumptions of linearity or periodicity are made. Further, the results are easy to relate to particular events in the cardiac cycle since all calculations are performed in the time domain. Originally, the concept demanded invasive measurement of blood pressure and velocity but further development resulted in an ultrasound based non‐invasive wave intensity system (the Aloka system) [67]. The pressure waveform was obtained by measuring the diameter changes of the artery using an echo‐tracking technique. The maximal and minimal values of the diameter wave form were then calibrated by systolic and diastolic blood pressure measured with a cuff‐type manometer applied to the upper arm. The velocity component was obtained from Doppler measurements, and was corrected for the angle between the ultrasound beam and the flow velocity vector. The new non‐invasive wave intensity system used a time‐normalized wave intensity value (WI = dP/dt *dU/dt), which allowed data acquired at different sampling rates to be compared directly, but resulted in a complex unit [W/s2m2].
The wave intensity curve
In the wave intensity curve, two distinct positive peaks are detectable. The first peak (W1) occurs in early systole as a result of ventricular contraction. It is a forward compression wave with positive pressure and velocity changes. Forward in this context means that the wave propagates in the direction of the mean blood flow, i.e., it is generated from the proximal site. The second positive peak occurs in late systole as a result of the deceleration of the ventricular contraction. This is a forward travelling decompression wave with negative pressure and velocity changes. The magnitude of W1 has been shown to correlate with the maximal LV pressure rise (dP/dtmax) and to increase with cardiac contractility [68, 69], while the magnitude of W2 correlates with the LV pressure decay (dP/dtmin) and with the time constant τ of LV pressure decay during isovolumic relaxation [68, 70]. There are also other peaks visible in the wave intensity curve. However, they are highly dependent on measuring position and clinical diagnosis [71]. One peak often mentioned in the literature is the so called negative area (NA), defined as the integrated wave intensity curve (when it goes below zero) between W1 and W2. NA has been proposed to represent a reflected wave coming from the periphery [72]. Figure 6 shows an example of a wave intensity curve.
‐ 11 ‐
Figure 6. An example of a wave intensity curve. The first positive peak corresponds to W1 and the second positive peak to W2. NA is the integrated wave intensity curve (when it goes below zero) between W1 and W2 (the yellow area).
The input variables in wave intensity calculations, the pressure and velocity changes, are closely related and a change in pressure must be accompanied by a change in velocity. The water hammer equations explain the relationship between the pressure changes (dP) and the velocity changes (dU): For forward travelling (+) waves originating from the proximal site ++ = ρcdUdP , c = local wave speed, ρ = density of the blood and for backward travelling (‐) waves originating from the distal site
−− −= ρcdUdP The wave intensity curve shows the sum of forward and backward travelling waves. Their relationship is displayed in Table 1. Table 1. Relationship between forward and backward travelling waves. WI dP dU
Forward travelling waves > 0 positive > 0 compression > 0 acceleration
< 0 decompression < 0 deceleration
Backward travelling waves < 0 negative > 0 compression < 0 deceleration
< 0 decompression > 0 acceleration
Separation of the wave intensity curve into forward and backward travelling waves can be useful in cases when forward and backward waves have approximately the same magnitude resulting in zero net wave intensity or when the arrival time of waves are important. The local wave speed c is required in order to separate forward and backward waves in the arteries. Several ways to determine the local wave speed have been reported in the literature [73‐75]. However, the importance of separating the wave intensity into forward and backward components is debatable.
‐ 12 ‐
During the time period when W1 and W2 occur, the backward components are practically zero. On the other hand, the local wave speed c is inversely related to the local distensibility and has therefore a direct clinical value. Stiffer arteries have lower distensibility and higher wave speed.
Clinical applications of wave intensity
Today, wave intensity analysis has a limited usage in the daily clinical practice. The method is primarily used in different research areas, in order to better understand normal physiology and pathology. Wave intensity has partly contributed to a better understanding of ventricular relaxation by enabling the analysis of how the pressure and velocity waves change in end‐systole [66, 70, 76]. When the mechanisms behind these waves were studied, it was concluded that a well‐functioning heart has the ability to maintain aortic flow in end‐systole, causing a rapid decrease of left ventricular pressure before the start of isovolumic relaxation, which enhances the cardiac function [70]. Further, the aortic and microcirculatoric effects on coronary hemodynamics have been investigated using wave intensity [77]. The latest contribution to the wave intensity research area is the reservoir‐wave approach [78], which should add incremental value to the quantification of forward and backward travelling waves. By separating the pressure component into changes caused purely by waves and changes caused by the compliant structure of the artery, it has been shown that the wave intensity analysis yields different results when the volume‐related changes have been accounted for [78]. Wave intensity has also been applied in patients with various cardiovascular diseases. Dilated cardiomyopathy demonstrates a pronounced impaired systolic function and W1 has been shown to be decreased in this patient group compared with matched controls. Hypertrophic cardiomyopathy is, on the other hand, primarily detectable when evaluating diastolic function and the values of W2 have been shown to be decreased in patients suffering from hypertrophic cardiomyopathy [69]. Hypertensive patients (systolic pressure > 135 mmHg or diastolic pressure > 85mmHg) showed no significant differences in W1 or W2 compared with normal subjects, however, NA was increased in the hypertension group [69]. The magnitudes of W1 and W2 have been shown to be reduced by decreased preload [79]. This was not verified when the hemodynamic effects of nitroglycerin were studied and it was concluded that changes in W1 were not caused by preload changes [80]. Clinical manifestations of cardiac disease are frequently present in patients with end stage renal disease (ESRD), and cardiovascular disease is the leading cause of mortality in this patient group [81]. The wave intensity concept has never been applied in this patient group, but would potentially increase the understanding of ventricular‐arterial interaction and cardiovascular dynamics in a patient group with high mortality in cardiovascular disease.
‐ 13 ‐
Aims
Ultrasound imaging is today the cornerstone in the routine evaluation of cardiovascular function. TVI and speckle tracking have the ability to quantify cardiovascular function with high reproducibility [82, 83] and have positively contributed to the interpretation of ultrasound data, especially for non‐experienced sonographers [84]. Even though there are several techniques available for the evaluation of cardiovascular function, there is still much to be done within this field in order to make early detection of cardiovascular dysfunction easier, faster and more reliable. The general aim of this thesis was therefore to develop and validate new approaches to monitoring cardiac and vascular function based on TVI and speckle tracking. In more detail, the specific aims were:
To test whether speckle tracking‐derived strain data obtained in the vascular wall could be used as input for wave intensity calculations.
To evaluate the changes in cardiovascular function induced by a single session of hemodialysis
(HD) in patients with end stage renal disease, with its documented high mortality in cardiovascular disease, by a combined approach employing speckle tracking‐derived wave intensity and TVI measurements.
To apply speckle tracking‐derived vascular two‐dimensional strain measurements to the
evaluation of vascular elasticity. To compare the capacity of speckle tracking‐derived two‐dimensional strain measurements in
the detection of age‐dependent differences in vascular elasticity with that of conventionally used elasticity variables.
To find a new approach to describe the timing of global and regional mechanical myocardial
events during the cardiac cycle, and to test the capacity of the resulting state diagrams of the heart to discriminate between healthy and diseased subjects.
‐ 14 ‐
‐ 15 ‐
Methods and materials
This thesis is based upon three new non‐invasive approaches to cardiac and vascular monitoring, wave intensity wall analysis (WIWA), two‐dimensional strain imaging in the common carotid artery and the state diagram of the heart. Their clinical usefulness has been tested in different patient groups. This chapter starts with an introduction of the approaches used, followed by the methodology which has been common for all investigations in this thesis.
3.1. Wave intensity wall analysis
The new wave intensity concept, WIWA, is a modification of the original invasive wave intensity analysis method [66] and the further developed non‐invasive ultrasound based wave intensity system [67]. The novelty of WIWA is that the measurements of the input variables to the wave intensity calculations are obtained in the vascular wall using a speckle tracking‐based ultrasound system. Within wave intensity analysis, arterial waves are defined as sequential infinitesimal changes of blood pressure and velocity. WIWA approximates these changes by measurements of strain rate in the radial and the longitudinal direction; the pressure component (dP/dt) is approximated by strain rate in the radial direction and the flow velocity changes (dU/dt) by strain rate in the longitudinal direction. Positive strain corresponds to lengthening or stretching, and negative strain to shortening or compression relative to the original length. Thus, the radial component is sign‐reversed in the WIWA calculations. Wave intensity measured in the arterial wall is defined as:
allongitudinradialWIWA ratestrainratestrainWI ×−=
Figure 7 shows the ROI placed in the far wall of the common carotid artery (CCA) and the input variables, longitudinal and radial strain rate, to the WIWA system.
‐ 16 ‐
Figure 7. Measurements of the input variables to WIWA. a) An ROI placed in the far wall of the artery. The ROI was placed to cover the whole wall to obtain better tracking. However, the “true” ROI consists of a considerably smaller area, i.e. the four coloured dots (yellow, blue, green, pink) seen in the middle of the image of the artery. b) Longitudinal strain rate. c) Radial strain rate (not sign‐reversed).
As in other wave intensity measurements, the result of the WIWA is presented as a curve containing the two positive peaks, W1 and W2, and, in some patients, a small NA. Figure 8 illustrates an example of a wave intensity curve generated using WIWA.
Figure 8. Wave intensity curve from WIWA. W1 is the first peak occurring in early systole and W2 is the second peak occurring in late systole. NA is not visible in this example.
‐ 17 ‐
3.2. Two‐dimensional strain imaging in the common carotid artery
Two‐dimensional strain imaging in the common carotid artery has a close relationship with WIWA by also using speckle tracking for quantification of cardiovascular function. The idea is to create a tool for the characterization of arterial stiffness with high accuracy, which can be easily applied in the clinical setting. From short‐axis images of the CCA, strain variables are extracted as surrogate variables of pressure (diameter) waveforms. The strain variables (strain and strain rate) can be assessed in radial or circumferential direction, both globally and regionally. Figure 9 illustrates circumferential strain and strain rate curves, where the dotted line corresponds to global values (mean of whole circular ROI) and the yellow line to the yellow dot placed in the far wall of the CCA.
Figure 9. a) ROI corresponding to the cross‐sectional area of the CCA short‐axis image. The large, yellow dot in the far wall corresponds to the position of the regional measurements. b‐c) Typical circumferential strain and strain rate curves from CCA. Global measurements are represented by white, dotted curves, whereas regional measurements by yellow curves.
3.3. State diagram
The state diagram of the heart has been proposed as a new visualization tool for cardiac time intervals. The basic idea is to present comparative data of systolic and diastolic performance giving a more complete overview of the timing of cardiac events. The state diagram is generated from time intervals identified in tissue velocity curves at six positions in the left ventricle and one position in the right ventricle. ROIs are placed at basal segments in apical images of the heart, in order to represent the longitudinal movements of the AV‐plane. From each tissue velocity, regional information can be extracted and by calculating mean time intervals for the left and right ventricles, a fairly good indication of the global cardiac function is obtained. The cardiac cycle is divided into six main phases: Pre‐Ejection, Ventricular Ejection, Post‐Ejection, Rapid Filling, Slow Filling and Atrial Contraction. Ventricular Ejection is subdivided into an early, a
‐ 18 ‐
mid and a late phase whereas Atrial Contraction and Rapid Filling are subdivided into two phases: an early phase and a late phase. Figure 10 displays the division of phases in the typical myocardial velocity curve.
Figure 10. Definition of cardiac phases in the myocardial velocity curve. The visual interpretation of the state diagram (Figure 11) can be rather complicated for an untrained eye, and should not be confused with the bull’s eye plot, which in contrast provides a two‐dimensional map of the entire surface of the left ventricle. One heartbeat corresponds to all 360° in the state diagram, starting with Pre‐Ejection and ending with Atrial Contraction. The outermost circular segment in the state diagram corresponds to the antero‐septal wall, followed by the anterior, the antero‐lateral, the infero‐lateral, the inferior, the infero‐septum (LV) and the medial (RV) walls. The aim of the colour coding is to clearly visualize when one state of the heart’s cycle goes into the next one. If there is any time delay between the walls, the phases are shifted clockwise or anti‐clockwise in the state diagram, relative to the antero‐septal wall. The calculation of the time shift between the walls is derived from the time between the R‐wave in the ECG‐curve and the start of Pre‐Ejection.
‐ 19 ‐
Figure 11. Visual interpretation of the state diagram.
3.4. Studied populations
In total, 99 patients have been studied. The studied population in Study 1 consisted of 10 apparently healthy individuals and 10 patients with established coronary disease with significant stenosis (> 50%) in at least one major epicardial artery. In Study 2, 45 patients treated with chronic HD were studied before and after HD. None of the patients showed signs of severe vascular heart disease, congestive heart failure, pericardial disease or acute myocardial ischaemia. The causes of renal disease were diabetes (11 patients), chronic glomerulonephritis (7 patients), focal segmental glomeruloscleros (1 patient), obstructive nephropathy (6 patients), nephrolithiasis (4 patients), chronic pyelonephritis (3 patients), nephrosclerosis (9 patients), polycystic kidney disease (3 patients) and SLE (1 patient). Antihypertensive medications were being used by 37 patients (82 %) and angiotensin‐converting enzyme inhibitors and angiotensin receptor antagonists, alone or in combination, were being used by 25 patients (56 %). Thirteen patients (29 %) were being treated with beta‐adrenergic blockers, 9 patients (20 %) with calcium antagonists and 24 patients (53 %) with furosemide. On the day of the study, no medication was taken until after all investigations were finished. In Study 3, the studied subjects consisted of 10 younger and 10 older apparently healthy individuals. Inclusion criterion for the younger group was an age of < 30 years (the “Controls” in Table 2) and for the older group age > 50 years (the “Patients” in Table 2). The two age groups did not differ in BMI and arterial blood pressure, and had normal cardiac function with the exception of a slightly reduced E/A ratio reflecting an age‐dependent diastolic disturbance in the older group. The studied subjects in Study 4 were 7 patients with non ST‐elevation myocardial infarction
‐ 20 ‐
(NSTEMI) and 7 controls, individually matched according to age and sex. In Studies 1, 3, and 4, the studied population was recruited in Sweden, and in Study 2, in Brazil. The healthy individuals in Studies 1 and 3 had normal resting ECG and normal findings on the echocardiograpic examination. None of them were on medical treatment or had any history of cardiovascular disease. The control group in Study 4 had no symptoms of ischaemic heart disease at the time for the investigation but some of them were on medical treatment for different affections. The numbers of participants, sex, age and patient characteristics in each study are summarized in Table 2. All studies were approved by the local ethics committee and all participants gave their informed consent to participate. Table 2. Characteristics of the studied population. Study Volunteers (F/M) Age (mean) Patient characteristics
Patients Controls Patients Controls
1 10 (‐/10) 10 (3/7) 66 43 Established coronary disease
2 45 (17/28) ‐ 54 ‐ Treated with chronic HD
3 10 (4/6) 10 (4/6) 55 27 “Patients” > 50 years, “Controls” < 30 years
4 7 (1/6) 7 (1/6) 58 59 NSTEMI
3.5. Vascular imaging
3.5.1. Ultrasonographic equipment and image acquisition
Vascular images were acquired in order to perform wave intensity or strain measurements in the CCA. All studied individuals were examined at rest, lying in supine position. The vascular imaging equipment from Aloka Ltd is the only commercially available system for non‐invasive wave intensity measurements and was used in Study 1 for validation of the new WIWA concept. The system contained a colour Doppler system (SSD‐5500, Aloka, Tokyo, Japan) with a 7.5 MHz linear array probe, an echo‐tracking subsystem and a personal computer. The data were updated with a frequency of 1 kHz and the steering angle of the ultrasound beam never exceeded ± 20° for any recording. Cine loops of the left and right CCA were acquired, approximately 2 cm from the bifurcation. After the examination, blood pressure was measured with a cuff‐type manometer applied to the upper arm (Riester, Germany). The diameter signal was then calibrated to the blood pressure. The wave intensity measurements were averaged over a minimum of four heartbeats. Two different GE systems (GE Vingmed Ultrasound, Horten, Norway), Vingmed System Vivid 7 (Study 1) and Vivid i system (Studies 2 and 3), with 10 MHz linear transducers were used for acquisition of two‐dimensional grey‐scale harmonic cine loops of the short‐axis and long‐axis view of the CCA, approximately 2 cm from the bifurcation. In Study 1, long‐axis images of both left and right CCA were acquired with an average frame rate (FR) of 42 frames/s, in Study 2, long‐axis images of the left CCA (FR = 75 frames/s) were acquired before and after HD and Study 3 used short‐axis and long‐axis images of the right CCA (FR = 78 frames/s). Offline analysis was performed with a commercially available speckle tracking algorithm (2D‐strain, EchoPac Dimension, GE
‐ 21 ‐
Vingmed Ultrasound). Matlab 7.0.1 and R2006B were used for post‐processing of the ultrasound data extracted from EchoPac Dimension. For WIWA calculations (Studies 1 and 2), the ROI were placed in the far wall of the artery and the initial ROI‐size was approximately 5 mm2 (ROI length was 3.1 mm and ROI width was 1.6 mm), see Figure 7. The ROI‐size varied slightly during the sequence due to the tracking algorithm. The obtained strain data were exported to Matlab for WIWA calculations. In Study 3, an ROI corresponding to the cross‐sectional area of the CCA wall was created with an initial ROI width of approximately 1.6 mm (Figure 9). Long‐axis images were used for calculations of Ep and β stiffness index.
3.5.2. Wave intensity variables (Studies 1 and 2)
The two positive peaks, W1 and W2, occurring in early and late systole were measured in Studies 1 and 2. In Study 1, the following variables were also measured: NA, the time intervals between the R‐wave in the ECG curve and W1 (R‐W1) and between W1 and W2 (W1‐W2), whereas in Study 2 the wave intensity indexes were adjusted for preload differences by dividing W1 and W2 with the end‐diastolic volume. In some individuals, the placement of the ROI had to be repeated several times in order to simultaneously obtain good tracking in the radial and longitudinal directions.
3.5.3. Short‐axis speckle tracking variables and established measures of arterial stiffness (Study 3)
Circumferential peak systolic strain (%), as well as early and late systolic strain rate (strain per time unit, 1/s) variables, were measured as an average over the whole ROI giving respective “global” strain and strain rate values. The early systolic strain rate value was defined as the first peak in the strain rate curve that occurred after the QRS‐complex in the ECG curve. The late systolic strain rate value was defined as the first peak in the strain rate curve that occurred after the T‐wave in the ECG curve and was sign‐reversed compared with the early systolic strain rate value. Early systolic strain rate is a result of the blood pressure rise associated with ventricular ejection, and late systolic strain rate occurs during the pressure decrease associated with ventricular relaxation. The EchoPac software is limited to performing only “global” measurements circumferentially, therefore radial peak systolic strain, as well as early and late systolic strain rate, were measured at a point (20x20 pixels) located in the far wall of the vessel (Figure 9, the large yellow dot), where also corresponding circumferential variables were determined. During systole, circumferential strain assumes positive values due to stretching, or expansion of the vessel wall, whereas radial strain becomes negative as a result of the compression of the vessel wall. Elastic modulus (Ep) and β stiffness index are conventional measures of arterial stiffness that relate changes in arterial pressure to changes in arterial diameter. Increases in Ep and β stiffness index correspond to increased arterial stiffness. For calculation of Ep and β stiffness index, the lumen diameter was measured in the long‐axis view of the CCA as the distance between the leading edge of the lumen‐intima echo of the near wall and the leading edge of the intima‐lumen echo of the far wall [63]. Blood pressure was measured after the vascular ultrasound examination using an automatic blood pressure device (Riester, Germany).
‐ 22 ‐
Ep [85] and β stiffness index [86] were calculated as:
diastdiastsyst
diastsystp D/)DD(
)PP(E
−
−=
diastdiastsyst
diastsyst
D/)DD()P/Pln(
−=β
where Ddiast and Dsyst were the minimal diastolic and maximal systolic lumen diameters, and Psyst and Pdiast were the systolic and diastolic pressures. Ep was presented in kilopascals (kPa).
3.6. Echocardiography
3.6.1. Echocardiography equipment and image acquisition
All individuals were examined at rest, lying in the left lateral position. Two different ultrasound systems were used for cardiac imaging; Vivid i system (GE Vingmed Ultrasound, Horten, Norway) was used in Studies 2 and 3 and Vingmed System Vivid 7 (GE Vingmed Ultrasound, Horten, Norway) in Study 4 with a standard 2D transducer (M3S) and a 3D matrix transducer (3V) for triplane imaging. The commercially available software EchoPac Dimension from GE Vingmed Ultrasound (Horten, Norway) was used for off‐line analysis of the recorded ultrasound images. Matlab 7.0.1 and R2006B were used for post‐processing of ultrasound data extracted from EchoPac Dimension. LV ejection‐fraction (LVEF) and measurements of end‐diastolic volumes were obtained with the biplane Simpson’s method and LV stroke volume (LVSV) by subtracting the end‐systolic volume from the end‐diastolic volume (Studies 2‐4). The E/A ratio was obtained by dividing mitral flow peak early diastolic velocity (E) by mitral flow peak late diastolic velocity (A) (Study 3). The wall motion score index (WMSI) was calculated from 16 segments in the left ventricle, where each of the segments was assigned a score based on myocardial thickening. A normally contracting segment (or hyperkinetic segment) was assigned a score of 1; hypokinesis, 2; akinesis, 3; dyskinesis, 4; and aneurysmal, 5. WMSI was calculated by dividing the sum of scores by the number of segments visualized and, in this thesis, was performed by an experienced stress echo reader. The E/E’ ratio was obtained by dividing the flow velocity (E) over the mitral valve by the mean tissue velocity (E’) in the septal and lateral corner of the mitral annulus (Study 4). Using TVI, the LV longitudinal function was assessed from apical two chamber (Study 4), apical three chamber (Study 4) and apical four chamber (Studies 2‐4) views and RV longitudinal function from the apical four chamber view (Study 4). Optimal image quality and frame rate were considered during every imaging sequence. The frame rate of the tissue Doppler data was 145 frames/s in Study 2, > 98 frames/s in Study 3, and > 100 frames/s in Study 4. A minimum of three consecutive heartbeats was digitally stored.
‐ 23 ‐
3.6.2. Tissue Doppler variables (Studies 2‐4)
The ROI was placed at a measuring position located at the basal segment of each wall. In Study 2, the LV longitudinal function was assessed from the apical views in which segments of the septal and the lateral walls were analyzed. In the extracted myocardial velocity curve, isovolumic contraction velocity (IVCV), peak systolic velocity (PSV), isovolumic relaxation velocity (IVRV), peak early diastolic velocity (E’) and peak late diastolic velocity (A’) were identified. The IVCT and the IVRT were measured as previously described [87]. IVCT was defined as the time period between the zero crossing for the ascending limb of the positive IVCV and the zero crossing for the ascending limb of the myocardial velocity curve at the beginning of ventricular ejection when the myocardial velocity curve showed biphasic movement. In cases with negative single‐phased movement during IVC, the beginning of ICVT was defined as the time for IVCV, and the end as the zero crossing, whereas in cases with positive single‐phased movement during ICVT, the beginning was defined as the zero crossing and the end as the time of the second inflection point. IVRT was defined as the time period between the zero crossing for the descending limb of the velocity curve at the end of ventricular ejection and the zero crossing for the descending limb of the velocity curve at the beginning of ventricular filling (E’). In cases with positive or negative single‐phased movement during IVRT, its beginning and end were defined in analogy with the single‐phased movement during IVCT. In Study 4, these time intervals were chosen to be called Pre‐Ejection and Post‐Ejection. In Studies 2 and 3, the myocardial displacement was obtained by temporal integration of the myocardial velocity curve during ventricular ejection. Movements occurring during the IVC and IVR periods were subsequently excluded. In Study 4, the identification of cardiac time intervals according to the set criteria (see Section 3.3 State diagram), was performed in the antero‐septal, the anterior, the antero‐lateral, the infero‐lateral, the inferior, the infero‐septum (LV) and the medial (RV) walls.
3.7. Statistical analysis
SPSS 16.0 and 14.0 were used for the statistical analysis performed in the studies. A p‐value < 0.05 was considered significant. All variables were tested for normality using the Kolmogorov‐Smirnov test. In Study 1, correlation analyses of the input variables in the two wave intensity systems were performed; the sign reversed radial strain was correlated with the blood pressure normalized diameter changes, and the longitudinal strain with blood velocity. A correlation analysis of the wave intensity amplitude and time indexes measured with both systems was also performed. In Study 2, three variables were not normally distributed, viz. preload‐adjusted W1, W2 and preload‐adjusted W2, and hence, the Wilcoxon’s signed rank test was used to compare the differences in these variables before and after HD. The remaining WIWA and TVI variables were tested used paired t‐tests. The correlation between TVI variables and WIWA variables were tested using the Pearson (normally distributed variables) or Spearman (at least one variable not normally distributed) correlation coefficient. Systolic TVI variables (IVCV, IVCT and PSV) were correlated with
‐ 24 ‐
W1 and preload‐adjusted W1 and early diastolic TVI variables (IVRV, IVRT and E’) with W2 and preload‐adjusted W2. Both the variables’ percentage change of HD, and the absolute values of the variables in each condition (before or after HD), were considered. In Study 3, unpaired t‐tests were performed in order to test age‐dependent differences between the younger and the older groups. Principal component analysis was used to test which variables that significantly contributed to the separation of the groups, enabling a reduction of the number of measured variables. The intra‐ and inter‐observer variability of the strain measurements were assessed by calculating coefficients of variation (CV). In Study 4, paired t‐tests were performed to test the difference in percentage duration of each phase in the global state diagram between the NSTEMI and control groups. In total, twenty phases were tested. Paired t‐tests were also used to test the differences in LVEF, WMSI and E/E’ ratio between the groups.
3.8. Summary of the methodology for each study
Study 1: Validation of wave intensity wall analysis (WIWA)
In this study, WIWA was validated against a commercially available wave intensity system, the Aloka system. First, a comparison between the input variables from the two systems was performed. The longitudinal strain and sign reversed radial strain, measured with the speckle tracking algorithm, were individually compared with blood velocity and blood pressure normalized diameter changes acquired by the Aloka system. Further, the calculated amplitude values and time indexes of the two systems were compared.
Study 2: Cardiovascular function in patients with end stage renal disease
This study evaluated the effects of a single session of HD on cardiovascular function. The patients in this study were treated with HD three times per week. Duration of maintenance HD varied from 1 to 98 months (average 23, SD 24.3). HD was performed using cellulose acetate dialysers with a dialysate flow of 500 ml/min. Bicarbonate‐buffered dialysate, with calcium 1.25 mmol/l, potassium 2 mmol/l, sodium 140 mmol/l and bicarbonate 35 mmol/l, was used. The changes in TVI and WIWA variables before and after HD, as well as the correlation between TVI and WIWA, were evaluated.
Study 3: Two‐dimensional strain imaging in the CCA
This study tested whether speckle tracking‐derived strain information could be used in the detection of age‐dependent differences in mechanical properties, and compared the discriminating performance of two‐dimensional strain with that of conventional measures such as Ep and β stiffness index. Strain and strain variables, and the conventional measures of arterial stiffness, were obtained in ten young healthy individuals (< 30 years) and ten older healthy individuals (> 50 years). The variables were compared between the groups and PCA and its regression extension was used to identify the variables that best separated the two age groups. The reproducibility of the strain measurements was tested in the younger age group.
Study 4: State diagram
This study presented a new imaging tool for cardiac time intervals, the state diagram of the heart. The feasibility of the state diagram method was tested in a sample of seven patients with NSTEMI,
‐ 25 ‐
which was individually matched according to age and sex to a control group. The differences in percentage duration of each phase of the left and right ventricles, twenty phases in total, were compared for the control group and for the NSTEMI group. Conventional echocardiography variables such as LVEF, WMSI and E/E’ were also measured in the NSTEMI and control groups, in order to test their discriminating capacity. Additionally, four case examples (normal, athlete, ischaemia, dyssynchrony) were presented.
‐ 26 ‐
‐ 27 ‐
Results
This chapter presents an overview of the results. All details are found in the papers attached.
Study 1: Validation of wave intensity wall analysis (WIWA)
The validation study demonstrated good visual agreement between the two systems with W1 and W2 clearly detectable in early systole and late systole, respectively. The correlation analysis between the input variables for wave intensity analysis performed by the WIWA system and the Aloka system demonstrated good correlation between radial strain and pressure. The correlation between longitudinal strain and blood velocity was also significant, but less so. Table 3 displays the correlation analysis for the input variables of the two systems. Table 3. Correlation between the input variables of the WIWA and Aloka systems (p < 0.001). WIWA ‐ Aloka
Radial strain ‐ pressure
Correlation
coefficient
WIWA ‐ Aloka
Longitudinal strain – flow velocity
Correlation
coefficient
Healthy subgroup, left CCA r = 0.794 Healthy subgroup, left CCA r = 0.477
Healthy subgroup, right CCA r = 0.752 Healthy subgroup, right CCA r = 0.512
Diseased subgroup, left CCA r = 0.706 Diseased subgroup, left CCA r = 0.539
Diseased subgroup, right CCA r = 0.715 Diseased subgroup, right CCA r = 0.546
The time interval between the R‐wave in the ECG curve and W1, R‐W1, measured by the WIWA and Aloka systems correlated for the group as a whole (r = 0.589, p = 0.006) and the healthy subgroup (r = 0.412, p = 0.008) but not for diseased subgroup. The time intervals between the two peaks in the wave intensity curve, W1‐W2, correlated for the total group (r = 0.595, p < 0.001) and for the diseased subgroup (r = 0.797, p < 0.001) but not for the healthy subgroup. The relation of the amplitude values, W1 and W2, between the two methods showed no significant correlation, neither for the total group nor the two subgroups.
Study 2: Cardiovascular function in patients with end stage renal disease
Ventricular contraction and contractility were improved after a single session of HD with increased PSV (p < 0.01) and IVCV (p < 0.001). The displacement of the AV‐plane was decreased after HD (p < 0.001). For the measured myocardial diastolic variables, E’ was significantly decreased
‐ 28 ‐
(p < 0.001), and IVRT was significantly prolonged (p < 0.01), after HD. The results regarding the TVI variables were consistent, both when analysing the lateral and septal walls separately and as an average of both walls, but the velocities were generally higher in the lateral wall compared with the septal wall. The ventricular‐arterial interaction evaluated with the WIWA system showed increased W1 (p < 0.05) and decreased W2 (p < 0.05) after HD, and are thus results that point in the same direction as the TVI measurements. Wave intensity indexes adjusted for differences in preload may contribute in the assessment of true LV contractility and relaxation, and preload‐adjusted W1
showed to be increased after HD (p < 0.01), while preload‐adjusted W2 showed no significant change. The results of the present study suggest that the load dependency of the measured diastolic variables seems to be more pronounced, most likely as a result of the influence of passive hemodynamic forces during diastole compared with the systolic variables, which are more dependent on active forces. Figure 12 summarizes the changes in WIWA variables and the mean values of TVI variables induced by a single session of HD. E’ and A’ are presented as absolute values.
Figure 12. Summary of the measured WIWA and TVI variables before and after HD. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. before HD. W1 = Wave intensity forward compression wave, W2 = Wave intensity forward decompression wave, IVCV = isovolumic contraction velocity, PSV = peak systolic velocity, IVRV = isovolumic relaxation velocity, E’ = maximal tissue velocity during early diastolic filling, A’ = maximal tissue velocity during atrial contraction, DISP = displacement of the AV‐plane, IVCT = isovolumic contraction time, IVRT = isovolumic relaxation time.
Only a few correlations between the WIWA and TVI variables were found, viz., the percentage change of W1 and IVCV (r = 0.46, p < 0.01), the percentage change of preload‐adjusted W1 and IVCV (r = 0.36, p < 0.05), preload‐adjusted W1 and IVCT (r = ‐0.26, p < 0.05). This can mainly be explained by the fact that wave intensity and TVI variables represent different events in the cardiovascular circulation.
Study 3: Two‐dimensional strain imaging in the CCA
Global and regional circumferential systolic strain and strain rate values were significantly higher (p < 0.001, p < 0.01 for regional late systolic strain rate) in the younger individuals whereas the values of conventional stiffness variables, Ep and β stiffness index, in the same group were lower
‐ 29 ‐
(p < 0.05). Among the radial variables, only early systolic strain rate differed significantly between the two age groups (p < 0.05). Figure 13 summarizes the measurements of the strain‐based and conventional stiffness variables (the absolute values).
Figure 13. Discrimination of the two age groups by arterial stiffness variables. * p < 0.05, ** p < 0.01, *** p < 0.001 vs. the age group < 30 years.
Among all strain and conventional stiffness variables, principal component analysis and its regression extension identified global variables obtained in the circumferential direction to be superior in the detection of age‐dependent differences in mechanical properties in the CCA. Circumferential global strain had the best discriminating capacity, followed by circumferential global early systolic strain rate and global late systolic strain rate. Also circumferential, regional variables showed clinical strength but not to the same extent as the global ones. Ep, β stiffness index and radial strain variables did not significantly contribute in the separation of the two age groups. The intra‐ and inter‐observer variability of the two‐dimensional strain measurements in CCA were generally lower for the global strain and strain rate values measured in the circumferential direction (CV range 4.6%‐9.9%) compared with the regional strain and strain rate values in the same direction (CV range 5.1%‐8.4%). The variability of the radial measurements was consistently considerably higher (CV 17.1%‐361.3%).
Study 4: State diagram
There were significant differences in percentage duration between the control group and the NSTEMI group in eight of the twenty total phases in the state diagram. Most pronounced were the differences in Pre‐Ejection (p = 0.001 for the left ventricle, p = 0.041 for the right ventricle) and Post‐Ejection (p = 0.029 for the left ventricle, p = 0.002 for the right ventricle), which were prolonged in both ventricles for the NSTEMI group compared with the normal group. The conventional echocardiographic measurements (LVEF, WMSI and E/E’) showed no significant difference between the two groups. Further, the state diagram clearly visualized that the pumping function in the normal subject, the athlete, the ischemic subject and the dyssynchronic subject was very different, where the first two subjects had a short transition time between pumping and filling of the ventricles (Pre‐Ejection and Post‐Ejection) and a long time for coronary perfusion (Rapid and Slow Filling) compared with the other two subjects.
‐ 30 ‐
‐ 31 ‐
Discussion
5.1. Wave intensity wall analysis
Wave intensity is the flux of energy per area unit [W/m2] carried by the wave as it propagates along the arteries, and is defined as the product of ∆P and ∆U. The wave intensity does not comprise the total power of ventricular work, only a small proportion of the power generated. To be able to easily compare data acquired at different sample rates, wave intensity was normalized with respect to time (WI= dP/dt * dU/dt) but with this new definition, the physical meaning of wave intensity as a measure was lost. Our WIWA system measures sign‐reversed radial strain rate instead of dP/dt, and longitudinal strain rate instead of dU/dt giving the newly‐defined wave intensity the unit 1/s2. As for the time‐normalized wave intensity, this definition has no obvious physical meaning but the same properties as for the original wave intensity were maintained by having WI > 0 for simultaneous increases or decreases in longitudinal strain and sign‐reversed radial strain, and WI < 0 for non‐simultaneous changes. Based on a study [20] showing a linear diameter‐pressure relationship, the commercially available non‐invasive wave intensity system (the Aloka system) [67] calibrates the maximum and the minimum diameter values to systolic and diastolic pressure in order to obtain a pressure wave form. The results of Study 1 showed that there is no direct transformation from strain to diameter (pressure) or flow velocity. There are physiological, methodological and technical explanations for that. Firstly, different arterial variables were measured; in the blood pool and in the arterial wall. The relation between these variables in different individuals is not fully known. Secondly, the respective variables were not measured during the same heart beat and no correction for respiratory influence was performed. The problems associated with the speckle tracking algorithm are addressed in Section 5.3. Since the wave intensity concept was introduced in 1990, a large number of studies have been carried out, primarily in order to better understand physiological events in the cardiovascular system. The wave intensity approach has been applied in the aorta [66], the coronaries [77] and in other large arteries [88]. The wave intensity pattern in the aorta is similar to the one seen in the carotid, brachial and radial arteries [88] by having two positive peaks, the first (W1) occurring in early systole and the second (W2) in late systole. Evidence has shown a smaller reflected wave to originate from the peripheral site as a result of impedance mismatch in the small arteries and arterioles. The reflected wave is visible in mid systole between the two positive wave intensity peaks, and it has been proposed that the timing of the that wave can be defined as the time interval between the onset of W1 and the onset of the NA (the backward wave) [72]. In cases of arterial stiffness or normal ageing, the reflected wave should consequently arrive earlier and would
‐ 32 ‐
display increased amplitude. However, published data on reflected waves measured with wave intensity do not seem to be congruent with the theoretic expectations. For example, the magnitude of NA measured in the CCA has been shown to be independent of age in both normal and hypertensive subjects [69]. Although W1 and W2 are clearly detectable in the WIWA curve, the effect of reflected waves returning from the periphery is scarcely visible, and when they are, the magnitudes are not reliably measurable due to very small values. Evidence has been provided showing arterial bifurcations to be well‐matched for waves generated by the heart, i.e they do not give rise to any wave reflections [89]. But if the bifurcations are well‐matched for waves from the heart they will automatically become poorly matched for waves coming from the distal site. This phenomenon is known as wave trapping. Waves generated from the heart travel along well‐matched bifurcations to the periphery where they are reflected backwards as they reach the smaller arteries and arterioles. The backward travelling waves reach the same bifurcations, but since they are poorly matched for waves travelling in the “wrong” direction, the reflections travel back towards the distal site. These re‐reflected waves will again be reflected, and so it will continue. The wave trapping phenomenon consequently makes the use of NA as a potential marker of arterial stiffness more or less impossible, and can partly explain the confusing results in studies about wave reflections in the arterial system, not only noticed with wave intensity measurements [90]. It may also explain why no clear wave reflections are detectable in the WIWA measurements. With these insights regarding wave reflections, the importance of W1 and W2 increases. But despite all efforts in trying to understand and explain how and why wave intensity patterns change during different conditions, its clinical usefulness for diagnosis and prognosis is still limited, and the main contribution of the concept is primarily a better understanding of cardiovascular hemodynamics and physiology. One reason for this may be that the waves are influenced by many different factors and, hence, it is difficult to distinguish the origin for a particular change in the wave intensity pattern. For example, the WIWA measurements are influenced by cardiac contraction and contractility, the amount of blood in the arteries, the degree of arterial stiffness etc. In Study 2, WIWA was tested in a clinical situation. The results of this study were encouraging in that the changes seen in wave intensity pattern were physiologically explainable, and similar to the results of the TVI measurements. But, besides providing the pleasure of “being able to measure” wave intensity, the obtained results also call for new efforts in order to find the role for WIWA in the clinical practice. A potential clinical application of WIWA could be to apply preload‐adjusted W1 in the assessment of LV contractility. An ideal index of contractility is sensitive to the intropic state of the heart, but insensitive to loading conditions, heart rate, and cardiac size. Contractility measures used today are, for example, the maximum value of the derivative of the LV pressure (LV dP/dtmax) and the end‐systolic elastance, which is defined as the linear slope of the end‐systolic pressure‐volume relation. The LV dP/dtmax tends to be more load dependent [91, 92] than the linear slope of the end‐systolic pressure‐volume relation that is close to an “ideal” measure of LV contractility [17, 93]. Both these measurements demand invasive registrations and are therefore seldom used in clinical studies. Invasive wave intensity measurements have shown a strong relationship between preload‐adjusted W1 and end‐systolic elastance [79]. Study 2 gave an indication of the potential use of preload‐adjusted W1 obtained from the WIWA system in the assessment of the cardiac contractility. However, further evaluation of preload‐adjusted W1 as a measure of cardiac contractility is
‐ 33 ‐
required, in which a correlation analysis of WIWA and end‐systolic elastance, and the determination of the optimal preload‐adjustment method, must be included. Another interesting idea for the development of the wave intensity concept into a technique with potential usefulness in the detection of local arterial stiffness would be to combine the data registered by the Aloka and WIWA systems. By relating the changes in pressure and flow velocity to the changes in strain measured in the arterial wall, a transfer function of wave intensity from blood to wall may be obtained. When comparing the duration of W1 registered by the Aloka system with the duration of W1 registered using WIWA, one could expect that large differences in time correspond to more elastic arteries.
5.2. Two‐dimensional strain imaging
Two‐dimensional strain imaging in the CCA is partly based on the same principle as that for WIWA by using speckle tracking‐based data obtained in the arterial wall. However, it has several advantages over the WIWA concept. Firstly, the method is easy to apply in the clinical setting with short acquisition time and short post‐processing. The analysis can be performed at any workstation with EchoPac, without any extraction of data to other programs. Additionally, the speckle tracking algorithm appears to perform very well in the circumferential direction in the short‐axis view of the artery. Secondly, the results provided are easy to interpret. Even though the two age groups in Study 3 did not differ in BMI or arterial blood pressure, and had normal cardiac function with the exception of slightly reduced E/A ratio in the older group, they displayed significant differences in circumferential strain and strain rate, indicating that this method is sensitive in the detection of pure mechanical wall property changes. Third, the method adds incremental value in the detection of age‐dependent differences compared with established methods for detection of arterial stiffness such as Ep and β stiffness index. A future application would be to investigate if this approach can be used for identification of rupture‐prone vulnerable plaques. The detection of these plaques is one of the major future challenges in cardiovascular research. Intravascular ultrasound elastography [94] has shown potential to detect vulnerable plaques but have some limitations in being invasive, expensive and can only be applied to patients that already undergo surgical intervention. At the time being, there is no standardized non‐invasive method for identifying vulnerable plaques, but the commonly used plaque classification is based on evaluation of the echolucency of the plaque in the grey‐scale image. The CCA is well suitable for non‐invasive evaluations since it is superficially located and therefore easy to image with high quality. Additionally, it has consistently been shown that atherosclerotic abnormalities detectable in the CCA are related to atherosclerosis in the coronary arteries, the aorta and other large arteries [95‐98]. The two‐dimensional strain imaging approach offers the possibility of a direct measurement of the vascular elastic properties in different parts of the plaque area and in the adjacent vascular sectors. This, in turn, may lead to improved assessment of plaque vulnerability and optimally to a more efficient anti‐atherosclerotic treatment and thereby also to the prevention of cerebral vascular events. An additional step would be to combine the strain information from long‐axis and short‐axis images to enable the assessment of a full 3D strain tensor of the artery, which potentially would further refine the diagnostic characterisation of vulnerable plaques.
‐ 34 ‐
5.3. Limitations of arterial measurements
The speckle tracking‐based input data to WIWA and to short‐axis measurements in the CCA are extracted from 2D‐strain software from GE Vingmed [39]. Applications based on a commercially available speckle tracking system are usually associated with some problems. For example, the user has normally only limited access to the full technical details of how the strain data are obtained and how the tracking of speckles is performed, or to what extent the in‐build spline functions and the temporal and the spatial filtering affect the strain and strain rate curves. Additionally, the 2D‐strain software uses different algorithms to obtain estimates of strain variables in the radial and longitudinal directions, which further complicates the interpretation of the obtained strain data. The estimation of longitudinal strain (corresponding to circumferential strain in short‐axis images of the artery) is based on the Lagrangian formula [99] and strain rate is its first derivative. Radial strain is obtained by integrating the radial strain rate curve, which is based on velocities of all tracking points. These different ways to perform strain calculations would, according to GE Vingmed, result in better tracking in the longitudinal than in the radial direction. It is a well‐known fact that the major limitation of speckle tracking‐based estimation of velocity values is that lateral estimations, i.e. estimations in the longitudinal direction of the artery, are intrinsically noisier than axial estimations, i.e. estimations in the radial direction of the artery, due to lower spatial resolution in the lateral direction of the image [100]. As a result, the lateral strain rate values will be noisier as well. The results in Study 1 showed a stronger correlation between pressure‐radial strain compared with that between blood velocity‐longitudinal strain. Accordingly, it appears that the better longitudinal tracking provided by the 2D‐strain software can not fully compensate for the lower spatial resolution in the lateral direction. The speckle tracking technique has been applied and validated foremost in heart applications [43, 45, 46] and the accuracy of speckle tracking‐based measurements in the arteries is not fully understood. So far, radial and circumferential function has been the main focus in arterial speckle tracking‐based research, and estimates of motion and strain in these directions have shown good accuracy in different experimental setups [101‐104]. The first studies on longitudinal movement were performed relatively recently and it was shown that longitudinal motion could be assessed using speckle tracking [22, 26]. The feasibility of simultaneous assessment of radial and longitudinal strain was tested in an in‐silico model of the carotid artery [102], showing better accuracy for radial than longitudinal strain estimates. The latter findings partly confirm the results showing higher correlation coefficients of pressure‐radial strain compared with that of blood velocity‐longitudinal strain as seen in Study 1. The above‐mentioned studies clearly show encouraging results regarding the possibility to perform strain measurements of the artery. However, it should be kept in mind that each study uses unique tracking algorithms and this does not necessarily mean that the speckle tracking algorithm currently provided by GE Vingmed has the same accuracy and quality in the assessment of arterial strain patterns. Some changes of the WIWA system have been addressed since the validation study (Study 1) was performed. The placement of the ROI in the artery is now performed differently. In Study 1, the size of the ROI placed in the arterial wall was approximately 5 mm2 (ROI length = 3.1 mm and ROI width = 1.6 mm). In Study 2, the ROI was placed to cover the whole far wall of the artery (see Figure 7) but data were extracted from four adjacent points giving a “true” ROI of the same size as in Study 1. This change in processing appears to give a better tracking of the speckles frame‐to‐frame. In Study 1, a low frame rate (average 42 frames/s) was pointed out as one of the major limitations of
‐ 35 ‐
the WIWA method. With recent updates in the software, and changes of acquisition settings, the frame rate for long‐axis (Study 2) and short‐axis (Study 3) images has increased to around 80 frames/s. Due to the differences in axial and lateral resolution the speckle tracking technique is angle dependant to some degree, and an increase in frame rate results in reduced lateral resolution and consequently, an increased angle dependency of the speckle tracking technique. However, with increasing computational power of the ultrasound equipment, it can be expected that cine loops with high frame rate and preserved lateral resolution will be generated with the speckle tracking technique in the near future. Performing WIWA can be very time‐consuming. In some individuals, the tracking works well instantly, but in others, the placement of the ROI has to be repeated several times in order to simultaneously obtain good tracking in the radial and longitudinal direction. We have tried to verify in which images optimal tracking can be obtained by comparing gain settings, acquisition angle or speckle pattern, but no conclusion can been drawn as yet. This problem essentially vanished with the introduction of two‐dimensional strain imaging in the short‐axis view of the CCA, where the tracking works more or less perfectly at once, resulting in stable strain curves with little noise. The low intra‐ and inter‐observer variability for circumferential global and regional measurements in Study 3 confirms the good quality of the extracted strain data. However, the good quality seems to be limited to circumferential short‐axis measurements since the radial short‐axis measurements showed a relatively high intra‐ and inter‐observer variability. When extrapolating these findings to WIWA, the high radial variability would question the accuracy of the WIWA measurements. In the intra‐ and inter‐observer variability study, the ROI was placed once in the images, with few further adjustments (if no bad tracking was apparent). As mentioned before, the tracking in the WIWA measurements is a problem, and presumably, if more effort had been put into the placement of the ROI in the short‐axis as well, the radial variability would have been lower. On the other hand, the way the intra‐ and inter‐observer variability study was performed closely matches the reality of the clinical routine where there is no time for long post‐processing.
5.4. State diagram
The quantitative analysis of cardiac function using ultrasound techniques available today does not allow for simultaneous analysis in different projections. The state diagram is a useful tool for visual and quantitative analysis of the timing of mechanical events in the heart. It provides an understanding of the temporal interrelationship within the left ventricle and between the ventricles allowing for a rapid approach for the assessment of myocardial abnormalities. For many years, cardiac time intervals have been described as a measurement of cardiac performance and as a tool for understanding and illustrating mechanical events in the heart. Prolongation, shortening and delay of the different time intervals have been evaluated as markers of cardiac dysfunction [105, 106]. Normally, the heart cycle is divided into time intervals separated by the opening and closing of the valves. These global events can be detected by conventional ultrasound registrations of blood flow through the aortic and mitral valves using pulsed or continuous Doppler, or by M‐mode images of the aortic and mitral valves. TVI provides information of regional myocardial function. The timing of regional events differs from the timing of global events due to regional differences in activation and relaxation of heart motion and shape [107]. Further, the regional myocardial velocity can either be an active movement contributing to global function or passive, i.e., as a result of movements occurring at other sites in the heart. There is also
‐ 36 ‐
a time delay between mechanical events, seen as myocardial movements, and their hydraulic consequences, i.e., the opening or closure of the mitral and aortic valves and the flow over the valves. Several studies have attempted to identify global events in myocardial movement curves but have come to different conclusions [87, 108‐110]. A recently performed study [108] clearly emphasizes this issue by investigating the moment of aortic valve closure in the tissue velocity curve based on four different suggestions found in the literature. High frame rate B‐mode images were compared with simultaneously recorded apical TVI images resulting in aortic valve closure positioned at the first negative peak after ventricular ejection. The discrepancy in the definition of global events in the myocardial movement curves can partly be explained by the different views of when the valves are considered to be open or closed. But despite the different thoughts regarding this issue, the pre‐ and post‐ejection phases are important considerations when evaluating cardiac function, and prolongation of these time intervals are associated with ventricular dysfunction [35‐37]. In the state diagram, the Pre‐Ejection and Post‐Ejection phases are defined exactly like the isovolumic phases previously suggested [87]. However, these phases can not be considered to be truly isovolumic since they include volume redistributions between the atria, the ventricles and the systemic circulation, a view which was partly confirmed in the study mentioned above [108]. The names IVCT and IVRT become consequently misleading and are the reason why these time intervals were called Pre‐Ejection and Post‐Ejection in Study 4. Pre‐Ejection is initiated with the beginning of the closure of the tricuspid and mitral valves, i.e. the generation of muscle tension and blood volume redistributions. The phase ends with a powerful tension increase that initiates the opening of the pulmonic and aortic valves. The flow through the valves does not occur at this moment, instead occurring in early Ventricular Ejection. Post‐Ejection starts when the pulmonic and aortic valves are about to close, just before a backflow into the ventricles is generated. The forces of the backflow overcome the tension in the muscles and the ventricular volumes increase. This phase ends when the rapid relaxation starts and the tricuspid and mitral valves are about to open, i.e., the muscle tension decreases, enabling the return of the AV‐plane. Consequently, the flow over the tricuspid and mitral valves occurs first in early Rapid Filling. The phases Ventricular Ejection, Rapid Filling and Atrial Contraction were divided into sub‐phases, in order to improve the diagnostic power of the state diagram. The duration of early Ventricular Ejection gives information about the pressure condition in the circulation. A short acceleration phase may be associated with low systemic pressures in analogy with the shape of the flow profiles seen in the pulmonalis and aorta. Mid Ventricular Ejection may be dependent on the preload in the ventricle. High preload and large stroke volumes would prolong mid Ventricular Ejection but the duration is also influenced by the muscle power of the myocardium. Late Ventricular Ejection is closely related to the appearance of mid Ventricular Ejection. Previous studies have shown that degeneration of the elastic properties of the ventricles provokes a restrictive ventricular filling pattern characterized by a shortened early diastolic mitral flow deceleration time [111]. This time corresponds to early Rapid Filling in the state diagram. The division of arterial contraction into sub‐phases would provide information about ventricular filling pressures. Exactly how the phase durations change during different conditions needs to be extensively evaluated in future studies. The reproducibility of the state diagram has not been evaluated. In velocity curves with relatively good quality, the division of the cardiac phases is easy to perform in most subjects. The only exception would be the division of the sub‐phases in Ventricular Ejection, which sometimes can be a matter of confusion.
‐ 37 ‐
The state diagram in its present form does not take into account the differences in heart rate between subjects. A true direct comparison of subjects, or evaluation of individual changes in cardiac function after interventions, is therefore difficult. Table 4 summarizes the heart rate dependency of the phases in the state diagram based on data from normal, young adult subjects during supine bicycle tests [112, 113]. The duration (ms) of the diastase, or the Slow Filling, demonstrates the most significant change, having a non‐linear dependency on heart rate with the largest changes at a heart rate of < 80 beats/minute [113]. The heart rate dependency of the other phases shows a slight linear decrease in duration; Ventricular Ejection having the highest changes and Pre‐ and Post‐Ejection the lowest. Worth noticing is that the percentage duration of Pre‐Ejection and Post‐Ejection during changes in heart rate remains more or less constant (|β| < 0.01). Table 4. Heart rate dependency of the phases in the state diagram. β is the gradient of the linear regression line. Phase Duration β Percentage duration β
Pre‐Ejection Linear decrease ‐0.4 Constant ‐0.001
Ventricular Ejection Linear decrease ‐1.3 Linear increase 0.16
Post‐Ejection Linear decrease ‐0.4 Constant ‐0.009
Rapid diastolic filling Linear decrease ‐0.6 Linear increase 0.08
Slow filling Non‐linear decrease ‐ Linear decrease ‐0.16
Atrial contraction Linear decrease 1.0 Linear decrease ‐0.07
As can be seen in Table 4, the heart rate dependency of the phases in the state diagram shows a quite complex pattern. The data presented are only valid for normal, young adult subjects and how the interrelationship of the phases changes during cardiac dysfunction and normal ageing is not fully known at present. Therefore, a correct normalization of the phases in the state diagram according to heart rate is more or less impossible to perform. Besides, to normalize the state diagram is perhaps not the optimal solution. The state diagram is normally based on data acquired at rest and consequently gives the “baseline” performance of the heart. Once a state diagram is generated, other information related to the cardiac function may be added to get a more complete overview of the cardiac performance. The longitudinal displacement of the AV‐plane and the forward and backward movements of the septum, representing the pumping and regulating function respectively, are important considerations.
‐ 38 ‐
‐ 39 ‐
Conclusions
Speckle tracking‐derived strain data can be useful as input for wave intensity calculations, where the correlation between radial strain and pressure (diameter) was stronger compared with that of longitudinal strain and velocity.
When applied in patients with end stage renal disease, the speckle tracking‐derived wave
intensity, the WIWA, indicates, similar to the results of the TVI measurements, that a single session of hemodialysis improves systolic function.
Preload‐adjusted W1 obtained by the WIWA system has the potential to be useful in the
assessment of true cardiac contractility. Two‐dimensional short‐axis strain imaging is sensitive in the assessment of vascular elasticity
in the carotid artery and appears to be superior to conventional measures in the detection of age‐dependent differences in arterial wall properties with global circumferential peak systolic strain being the most sensitive variable, followed by global circumferential early systolic strain rate and global circumferential late systolic strain rate.
The state diagram has the potential to be an efficient tool in the visualization of normal and
disturbed cardiac performance and was shown to be useful in the clinical setting of non ST‐elevation myocardial infarction (NSTEMI).
WIWA, two‐dimensional strain imaging in the common carotid artery and the state diagram
show potential to be useful in the evaluation of cardiovascular function but their clinical value needs to be further evaluated.
‐ 40 ‐
‐ 41 ‐
References
1. Lloyd‐Jones D, Adams R, Carnethon M et al. Heart Disease and Stroke Statistics ‐ 2009 Update: A Report From the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2009; 119: e21‐e181 (e142). 2. Jabbour S, Reddy KS, Muna WFT, Achutti A. Cardiovascular disease and the global tobacco epidemic: a wake‐up call for cardiologists. International Journal of Cardiology 2002; 86: 185‐192. 3. Rashid MN, Fuentes F, Touchon RC, Wehner PS. Obesity and the Risk for Cardiovascular Disease. Preventive Cardiology 2003; 6: 42‐47. 4. Herrington DM, Brown WV, Mosca L et al. Relationship Between Arterial Stiffness and Subclinical Aortic Atherosclerosis. Circulation 2004; 110: 432‐437. 5. Edler I, Hertz CH. The use of ultrasonic reflectoscope for continuous recordings of movements of the heart wall. Kungl Fysiogr Sällsk Lund Förhandl 1954; BD 24: 1‐19. 6. Permanente K. http://www.kpep.org/normal_heart_function/index.htm 2009‐09‐23. 7. Greenbaum R, Ho S, Gibson D, Becker A, Anderson R. Left ventricular fibre architecture in man. Br Heart J 1981; 45: 248‐263. 8. Sengupta P, Krishnamoorthy V, Korinek J et al. Left Ventricular Form and Function Revisited: Applied Translational Science to Cardiovascular Ultrasound Imaging. Journal of the American Society of Echocardiography 2007; 20: 539‐551. 9. Brecher G, Galletti PM. Functional anatomy of cardiac pumping. In: Hamilton WF, Dow P, editors. Handbook of Physiology. Washington D.C; 1963. 10. Böhme W. Über den aktiven Anteil des Herzens an der Förderung des Venenblutes. Ergeb Physiologie 1936; 38: 1324‐1329. 11. Harvey W. An Anatomical Disputation Concerning the Movement of the Heart and Blood in Living Creatures (Reviewed by Lester S. King). JAMA 1977; 238: 630. 12. Lundbäck S: Cardiac Pumping and Function of the Ventricular Septum. Stockholm: Karolinska Institute; 1986. 13. Torrent‐Guasp F, Buckberg G, Clemente C, Cox J, Coghlan H, Gharib M. The structure and function of the helical heart and its buttress wrapping. I. The normal macroscopic structure of the heart. Semin Thorac Cardiovasc Surg 2001; 13: 301‐319. 14. Wandt B. Long‐Axis Contraction of the Ventricles: A Modern Approach, but Described Already by Leonardo da Vinci. Journal of the American Society of Echocardiography 2000; 13: 699‐706. 15. Katz A. Ernest Henry Starling, His Predecessors, and the "Law of the Heart". Circulation 2002; 106: 2986‐2992.
‐ 42 ‐
16. Sonnenblick E, Evans Downing S. Afterload as a primary determinant of ventricular performance. Am J Physiol 1963; 204: 604‐610. 17. Bombardini T. Myocardial contractility in the echo lab: molecular, cellular and pathophysiological basis. Cardiovascular Ultrasound 2005; 3. 18. Hoeks APG, Brands PJ, Smeets FAM, Reneman RS. Assessment of the distensibility of superficial arteries. Ultrasound in Medicine & Biology 1990; 16: 121‐128. 19. Meinders J, Hoeks A. Simultaneous assessment of diameter and pressure waveforms in the carotid artery. Ultrasound in Medicine and Biology 2004; 30: 147‐154. 20. Sugawara M, Niki K, Furuhata H, Ohnishi S, Suzuki S. Relationship between the pressure and diameter of the carotid artery in humans. Heart Vessels 2000; 15: 49‐51. 21. Patel DJ, Mallos AJ, Fry DL. Aortic mechanics in the living dog. Journal of Applied Physiology 1961; 16: 293‐299. 22. Cinthio M, Rydén‐Ahlgren Å, Jansson T, Eriksson A, Persson HW, Lindström K. Evaluation of an ultrasonic echo‐tracking method for measurements of arterial wall movements in two dimensions. IEEE Ultrasonics, ferroelectrics and frequency control society 2005; 58: 1300‐1311. 23. Golemati S, Sassano A, Lever MJ, Bharath AA, Dhanjil S, Nicolaides AN. Carotid artery wall motion estimated from B‐mode ultrasound using region tracking and block matching. Ultrasound in Medicine and Biology 2003; 29: 387‐399. 24. Sunagawa K, Kanai H, Tanaka M. Simulaneous measurement of blood flow and arterial wall vibrations in radial and axial directions. IEEE Ultrasonics Symposium 2000; 2: 1541‐1544. 25. Persson M, Rydén‐Ahlgren Å, Jansson T, Eriksson A, Persson H, Lindström K. A new non‐invasive ultrasonic method for simultaneous measurements of longitudinal and radial arterial wall movements: first in vivo trial. Clinical Physiology and Functional Imaging 2003; 23: 247‐251. 26. Cinthio M, Ryden‐Ahlgren Å, Bergkvist J, Jansson T, Persson HW, Lindström K. Longitudinal movements and resulting shear strain of the arterial wall. Am J Physiol Heart Circ Physiol 2006; 291: 394–402. 27. Isaaz K, Thompson A, Ethevenot G, Cloez JL, Brembilla B, Pernot C. Doppler echocardiographic measurement of low velocity motion of the left ventricular posterior wall. Am J Cardiol 1989; 64: 66‐75. 28. McDicken W, Sutherland G, Moran C, Gordon L. Colour Doppler velocity imaging of the myocardium. Ultrasound Med Biol 1992; 18: 561‐564. 29. Brodin L‐Å, van der Linden J, Jensen‐Urstad M, Olstad B. New functional imaging options with echocardiography based on myocardial velocity curves. IEEE Computers in Cardiology. 1998: 253‐256. 30. Brodin L‐Å, van der Linden J, Olstad B. Echocardiographic functional images based on tissue velocity information. Herz 1998; 23: 491‐498. 31. Mädler C, Payne N, Wilkenshoff U et al. Non‐invasive diagnosis of coronary artery disease by quantitative stress echocardiography: optimal diagnostic models using off‐line tissue Doppler in the MYDISE study. Eur Heart J 2003; 24: 1584‐1594. 32. Edvardsen T, Gerber BL, Garot J, Bluemke DA, Lima JAC, Smiseth OA. Quantitative Assessment of Intrinsic Regional Myocardial Deformation by Doppler Strain Rate Echocardiography in Humans. Circulation 2002; 106: 50‐56. 33. Gorcsan J, Strum DP, Mandarino WA, Gulati VK, Pinsky MR. Quantitative Assessment of Alterations in Regional Left Ventricular Contractility With Color‐Coded Tissue Doppler Echocardiography Circulation 1997; 95: 2423‐2433. 34. Søgaard P, Egeblad H, Kim WY et al. Tissue doppler imaging predicts improved systolic performance and reversed left ventricular remodeling during long‐term cardiac resynchronization therapy. Journal of the American College of Cardiology 2002; 40: 723‐730.
‐ 43 ‐
35. García‐Fernández MA, Azevedo J, Moreno M et al. Regional diastolic function in ischaemic heart disease using pulsed wave Doppler tissue imaging. Eur Heart J 1999; 20: 496‐505. 36. Hayashi S, Rohani M, Lindholm B et al. Left ventricular function in patients with chronic kidney disease evaluated by colour tissue Doppler velocity imaging. Nephrol Dial Transplant 2006; 21: 125‐132. 37. Hirschfeld S, Meyer R, Korfhagen J, Kaplan S, Liebman J. The isovolumic contraction time of the left ventricle. An echographic study. Circulation 1976; 54: 751‐756. 38. Stoylen A. http://folk.ntnu.no/stoylen/ 2009‐09‐23. 39. Leitman M, Lysyansky P, Sidenko S et al. Two‐Dimensional Strain ‐ A Novel Software for Real‐time Quantitative Echocardiographic Assessment of Myocardial Function. Journal of the American Society of Echocardiography 2004; 17: 1021‐1029. 40. Reisner SA, Lysyansky P, Agmon Y, Mutlak D, Lessick J, Friedman Z. Global longitudinal strain: a novel index of left ventricular systolic function. J Am Soc Echocardiogr 2004; 17: 630‐633. 41. Amundsen BH, Helle‐Valle T, Edvardsen T et al. Nonivasive Myocardial Strain Measurement by Speckle Tracking Echocardiography ‐ Validation Against Sonomicrometry and Tagged Magnetic Resonance Imaging. Journal of the American College of Cardiology 2006; 47: 789‐793. 42. Bohs LN, Trahey GE. A Novel Method for Angle Independent Ultrasonic Imaging of Blood Flow and Tissue Motion. IEEE Transactions on biomedical engineering 1991; 38: 280‐286. 43. Helle‐Valle T. New noninvasive method for assessment of left ventricular rotation: speckle tracking echocardiography. Circulation 2005; 112: 3149‐3156. 44. Popovic ZB, Benejam C, Bian J et al. Speckle‐tracking echocardiography correctly identifies segmental left ventricular dysfunction induced by scarring in a rat model of myocardial infarction. Am J Physiol Heart Circ Physiol 2007; 292: 2809‐2816. 45. Becker M. Analysis of myocardial deformation based on pixel tracking in 2D echocardio‐graphic images allows quantitative assessment of regional left ventricular function. Heart and Vessels 2005; 92: 1102‐1108. 46. Suffoletto MS, Dohi K, Cannesson M, Saba S, Gorcsan J. Novel Speckle‐Tracking Radial Strain From Routine Black‐and‐White Echocardiographic Images to Quantify Dyssynchrony and Predict Response to Cardiac Resynchronization Therapy. Circulation 2006; 113: 960‐968. 47. Gustafsson U, Lindqvist P, Mörner S, Waldenström A. Assessment of regional rotation patterns improves the understanding of the systolic and diastolic left ventricular function: an echocardiographic speckle‐tracking study in healthy individuals. Eur J Echocardiogr 2009; 10: 56‐61. 48. Kim H, Sohn D, Lee S et al. Assessment of left ventricular rotation and torsion with two‐dimensional speckle tracking echocardiography. J Am Soc Echocardiogr 2007; 20: 45‐53. 49. van Dalen BM, Soliman O, Vletter W, ten Cate F, Geleijnse M. Insights into left ventricular function from the time course of regional and global rotation by speckle tracking echocardiography. Echocardiography 2009; 26: 371‐377. 50. Artis NJ, Oxborough DL, Williams G, Pepper CB, Tan LB. Two‐dimensional strain imaging: A new echocardiographic advance with research and clinical applications. International Journal of Cardiology 2007; 123: 240‐248. 51. Chae C, Pfeffer M, Glynn R, Mitchell G, Taylor J, Hennekens C. Increased pulse pressure and risk of heart failure in the elderly. JAMA 1999; 281: 634‐639. 52. Girerd X, Laurent S, Pannier B, Asmar R, Safar M. Arterial distensibility and left ventricular hypertrophy in patients with substained essential hypertension. Am Heart J 1991; 122: 1210‐1214. 53. Störk S, van den Beld AW, von Schacky C et al. Carotid Artery Plaque Burden, Stiffness, and Mortality Risk in Elderly Men. A Prospective, Population‐Based Cohort Study Circulation 2004; 110: 344‐348.
‐ 44 ‐
54. Benetos A, Waeber B, Izzo J et al. Influence of age, risk factors, and cardiovascular and renal disease on arterial stiffness: clinical applications. Am J Hypertens 2002; 15: 1101‐1108. 55. McEniery CM, Yasmin, Hall IR, Qasem A, Wilkinson IB, Cockcroft JR. Normal Vascular Aging: Differential Effects on Wave Reflection and Aortic Pulse Wave Velocity. J Am Coll Cardiol 2005; 46: 1753‐1760. 56. Mitchell GF, Parise H, Benjamin EJ et al. Changes in Arterial Stiffness and Wave Reflection With Advancing Age in Healthy Men and Women: The Framingham Heart Study. J Hypertens 2004; 43: 1239‐1245. 57. Avolio A, Jones D, Tafazzoli‐Shadpour M. Quantification of alterations in structure and function of elastin in the arterial media. Hypertension 1998; 32: 170‐175. 58. Lakatta E. Arterial and cardiac aging: major shareholders in cardiovascular disease enterprises: Part III: cellular and molecular clues to heart and arterial aging. Circulation 2003; 107: 490‐497. 59. Glasser S, Arnett D, McVeigh G et al. Vascular compliance and cardiovascular disease: a risk factor or a marker? Am J Hypertens 1997; 10: 1175‐1189. 60. Laurent S, Boutouyrie P. Arterial stiffness: a new surrogate end point for cardiovascular disease? J Nephrol 2007; 20: S45‐50. 61. Mattace‐Raso FUS, van der Cammen TJM, Hofman A et al. Arterial Stiffness and Risk of Coronary Heart Disease and Stroke. The Rotterdam Study. Circulation 2006; 113: 657‐663. 62. Bramwell JC, Hill AV. Velocity of transmission of the pulse wave and elasticity of the arteries. Lancet 1922; 1: 891‐892. 63. Wendelhag I, Gustavsson T, Suurküla M, Berglund G, Wikstrand J. Ultrasound measurement of wall thickness in the carotid artery: fundamental principles and description of a computerized analysing system. Clin Physiol 1991; 11: 565‐577. 64. Hamilton PK, Lockhart CJ, Quinn CE, McVeigh GE. Arterial stiffness: clinical relevance, measurement and treatment. Clinical Science 2007; 113: 157‐170. 65. Pannier B, Avolio A, Hoeks A, Mancia G, Takazawa K. Methods and devices for measuring arterial compliance in humans. Am J Hypertens 2002; 15: 743‐753. 66. Parker K, Jones CJH. Forward and Backward Running Waves in the Arteries: Analysis Using the Method of Characteristics. Journal of Biomechanical Engineering 1990; 112: 322‐326. 67. Harada A, Okada T, Sugawara M, Niki K. Development of a Non‐invasive Real‐time Measurement System of Wave Intensity. IEEE Ultrasonics Symposium 2000; 1517‐1520. 68. Ohte N, Narita H, Sugawara M et al. Clinical usefulness of carotid arterial wave intensity in assessing left ventricular systolic and early diastolic performance. Heart Vessels 2003; 18: 107‐111. 69. Sugawara M, Niki K, Ohte N, Okada T, Harada A. Clinical usefulness of wave intensity analysis. Med Biol Eng Comput 2009; 47: 197‐206. 70. Sugawara M, Uchida K, Kondoh Y et al. Aortic blood momentum ‐ the more the better for the ejecting heart in vivo? Cardiovascular Research 1997; 14: 433‐446. 71. Parker K. http://www.bg.ic.ac.uk/research/intro_to_wia/ 2009‐08‐12. 72. Koh T, Pepper J, DeSouza A, Parker K. Analysis of wave reflections in the arterial system using wave intensity: a novel method for predicting the timing and amplitude of reflected waves. Heart Vessels 1998; 13: 103‐113. 73. Davies JE, Whinnett ZI, Francis DP et al. Use of simultaneous pressure and velocity measurements to estimate arterial wave speed at a single site in humans. Am J Physiol Heart Circ Physiol 2006; 290: H878‐H885. 74. Harada A, Okada T, Niki K, Chang D, Sugawara M. On‐line noninvasive one‐point measurements of pulse wave velocity. Heart and Vessels 2002; 17 61‐68.
‐ 45 ‐
75. Khir A, Swalen M, Feng J, Parker K. Simultaneous determination of wave speed and arrival time of reflected waves using the pressure–velocity loop. Medical and Biological Engineering and Computing 2007; 45: 1201‐1210. 76. Parker KH, Jones CJ, Dawson JR, Gibson DG. What stops the flow of blood from the heart? Heart and Vessels 1988; 4: 241‐245. 77. Sun Y‐H, Anderson TJ, Parker KH, Tyberg JV. Wave‐intensity analysis: a new approach to coronary hemodynamics. J Appl Physiol 2000; 89: 1636‐1644. 78. Tyberg J, Davies J, Wang Z et al. Wave intensity analysis and the development of the reservoir–wave approach. Medical and Biological Engineering and Computing 2009; 47: 221‐232. 79. Nakayama M, Itoh H, Oikawa K et al. Preload‐adjusted 2 wave‐intensity peaks reflect simultaneous assessment of left ventricular contractility and relaxation. Circulation journal 2005; 69: 683‐687. 80. Niki K, Sugawara M, Chang D et al. Effects of sublingual nitroglycerin on working conditions of the heart and arterial system: analysis using wave intensity. J Med Ultrason 2005; 32: 145‐152. 81. Levey A, Eknoyan G. Cardiovascular disease in chronic renal disease. Nephrology Dialysis Transplantation 1999; 14: 828‐833. 82. Gaballa M, Lind B, Storaa C, Brodin L‐Å. Intra‐and Interobserver reproducibility in off‐line extracted cardiac tissue doppler velocity measurements and derived variables. Engineering in Medicin and Biology Society Proceedings of the 23rd Annual International Conference of the IEEE 2001; 1: 160‐162. 83. van Dalen B, Bosch J, Kauer F et al. Assessment of Mitral Annular Velocities by Speckle Tracking Echocardiography versus Tissue Doppler Imaging: Validation, Feasibility, and Reproducibility. J Am Soc Echocardiogr 2009; 22 1302‐1308. 84. Caiani EG, Lang RM, Korcarz CE et al. Improvement in echocardiographic evaluation of left ventricular wall motion using still‐frame parametric imaging. Journal of the American Society of Echocardiography 2002; 15: 926‐934. 85. Peterson L, Jensen R, Parnell J. Mechanical properties of Arteries in vivo. Circulation Research 1960; 8: 622‐633. 86. Kawasaki T, Sasayama S, Yagi S, Asakawa T, Hirai T. Non‐invasive assessment of the age related changes in stiffness of major branches of human arteries. Cardiovasular research 1987; 21: 678‐687. 87. Lind B, Nowak J, Cain P, Quintana M, Brodin L‐Å. Left ventricular isovolumic velocity and duration variables calculated from colour‐coded myocardial velocity images in normal individuals. Eur J Echocardiogr 2004; 5: 284‐293. 88. Zambanini A, Cunningham S, Parker K, Khir A, Thom SM, Hughes A. Wave‐energy patterns in carotid, brachial, and radial arteries: a noninvasive approach using wave‐intensity analysis. Am J Physiol Heart Circ Physiol 2005; 289: H270‐276. 89. Parker KH. An Introduction to Wave Intensity Analysis. Medical and Biological Engineering and Computing 2009; 47: 175‐188. 90. van der Bos G, Westerhof N, Elzinga G, Sipkema P. Reflection in the systemic arterial system: effects of aortic and carotid occlusion. Cardiovascular Research 1976; 10: 565‐573. 91. Mahler F, Ross Jr J, O'Rourke R, Covell J. Effects of changes in preload, afterload and inotropic state on ejection and isovolumic phase measures of contractility in the conscious dog. Am J Cardiol 1975; 35: 626‐634. 92. Mason D. Usefulness and limitations of the rate of rise of intraventricular pressure (dp‐dt) in the evaluation of myocardial contractility in man. Am J Cardiol 1969; 23: 516‐527. 93. Schertel E. Assessment of left‐ventricular function. Thorac Cardiovasc Surg 1998; 46: 248‐254.
‐ 46 ‐
94. de Korte CL, van der Steen AFW. Intravascular ultrasound elastography: an overview. Ultrasonics 2002; 40: 859‐865. 95. del Sol AI, Bots ML, Grobbee DE, Hofman A, Witteman JCM. Carotid intima‐media thickness at different sites: relation to incident myocardial infarction. The Rotterdam Study. European Heart Journal 2002; 23: 934‐940. 96. Hollander M, Bots M, Del Sol A et al. Carotid plaques increase the risk of stroke and subtypes of cerebral infarction in asymptomatic elderly: the Rotterdam study. Circulation 2002; 105: 2872‐2877. 97. Nowak J, Nilsson T, Sylvén C, Jogestrand T. Potential of Carotid Ultrasonography in the Diagnosis of Coronary Artery Disease. Stroke 1998; 29: 439‐446. 98. O'Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson SK. Carotid‐Artery Intima and Media Thickness as a Risk Factor for Myocardial Infarction and Stroke in Older Adults. The new England journal of medicine 1999; 340: 14‐22. 99. D’hooge J, Heimdal A, Jamal F et al. Regional Strain and Strain Rate Measurements by Cardiac Ultrasound: Principles, Implementation and Limitations. Eur J Echocardiography 2000; 1: 154–170. 100. D'hooge J, Bijnens B. The principles of ultrasound based motion and deformation estimation. In: Sutherland GR, Hatle L, Claus P, D'hooge J, Bijnens BH, editors. Doppler Myocardial Imaging ‐ A textbook. 1st edn. Hasselt: BSWK; 2006. p. 23‐44. 101. Hansen HH, Lopata RG, de Korte CL. Noninvasive carotid strain imaging using angular compounding at large beam steered angles: validation in vessel phantoms. IEEE transactions on medical imaging 2009 28: 872‐880. 102. Larsson M, Kremer F, Claus P, D'Hooge J. Ultrasound‐based 2D Strain Estimation of the Carotid Artery: an in‐silico feasibility study. IEEE International Ultrasonics Symposium 2009; In press. 103. Mercure E, Cloutier G, Schmitt C, Maurice R. Performance evaluation of different implementations of the Lagrangian speckle model estimator for non‐invasive vascular ultrasound elastography. Med Phys 2008; 35: 3116‐3126. 104. Ribbers H, Lopata RGP, Holewijn S, Pasterkamp G, Blankensteijn JD, Korte CLd. Noninvasive Two‐Dimensional Strain Imaging of Arteries: Validation in Phantoms and Preliminary Experience in Carotid Arteries In Vivo. Ultrasound in Medicine & Biology 2007; 33: 530‐540. 105. Sogaard P, Egeblad H, Pedersen AK et al. Sequential Versus Simultaneous Biventricular Resynchronization for Severe Heart Failure. Evaluation by Tissue Doppler Imaging. Circulation 2002; 106: 2078‐2084. 106. Weissler AM, Harris WS, Schoenfeld CD. Systolic Time Intervals in Heart Failure in Man. Circulation 1968; 37: 149‐159. 107. Sutherland GR, Hatle L, Claus P, Herbots K, Šeparović J. Normal Data. In: Sutherland GR, Hatle L, Claus P, D'hooge J, Bijnens BH, editors. Doppler Myocardial Imaging ‐ A textbook. 1st edn. Hasselt: BSWK; 2006. p. 49‐102. 108. Aase SA, Torp H, Støylen A. Aortic valve closure: relation to tissue velocities by Doppler and speckle tracking in normal subjects. European Journal of Echocardiography 2008; 9: 555‐559. 109. Opdahl A, Remme EW, Helle‐Valle T, Vartdal T, Smiseth OA. Accurate timing of mitral valve opening by tissue Doppler imaging for an all in one beat analysis. European Journal of Echocardiography 2006; 7: S122. 110. Remme EW, Lyseggen E, Helle‐Valle T et al. Mechanisms of Preejection and Postejection Velocity Spikes in Left Ventricular Myocardium. Interaction Between Wall Deformation and Valve Events. Circulation 2008; 115: 373‐380.
‐ 47 ‐
111. Møller JE, Søndergaard E, Poulsen SH, Egstrup K. Pseudonormal and Restrictive Filling Patterns Predict Left Ventricular Dilation and Cardiac Death After A First Myocardial Infarction: A Serial Color M‐Mode Doppler Echocardiographic Study. J Am Coll Cardiol 2000; 36: 1841‐1846. 112. Bjällmark A, Larsson M, Johnson J et al. The Importance of Time variables for describing the Mechanics of the Heart in healthy individuals at different heart rates. Medicinteknikdagarna. Örebro. 2007. 113. Chung CS, Karamanoglu M, Kovács SJ. Duration of diastole and its phases as a function of heart rate during supine bicycle exercise. Am J Physiol Heart Circ Physiol 2004; 287: 2003‐2008.