Charles Mugera 00395247 MSc Medical Ultrasound 2
Two dimensional analysis of left ventricular
myocardial regional and global contractility using
speckle tracking pre and post aortic valve
replacement for aortic stenosis.
_____________________________________________________________________
_
Dr Charles M Mugera
Echocardiography department
Hammersmith Hospital, London
MSc medical ultrasound
Clinical sciences centre
Charles Mugera 00395247 MSc Medical Ultrasound 3
Name: Dr Charles .M. Mugera
Supervisors: Mr. David Dawson M.Sc
Senior lecturer
Department of Medical ultrasound
Echocardiography
Professor Petros Nihoyannopoulos
Professor of Cardiology
National Heart and Lung Institute
Tel: +44 (0)20 8383 3948
Email: p.nihoyannopoulos @imperial.ac.uk
Dissertation title: Two dimensional analysis of left ventricular myocardial regional
and global contractility using speckle tracking pre and post aortic valve replacement
for aortic stenosis.
.
Type: Project
Research area: Echocardiography
Degree title: MSc Medical ultrasound
Department: Echocardiography
School: Clinical sciences centre
University: Imperial college London, Faculty of Medicine Office
Level 2, Faculty Building
South Kensington Campus
Imperial College
Exhibition Road
London SW7 2AZ
Charles Mugera 00395247 MSc Medical Ultrasound 4
Abstract
Introduction: Abnormalities in regional left ventricular (LV) function in severe
aortic stenosis (AS) have yet to be appropriately characterized. Often patients with
severe aortic stenosis have subclinical left ventricular systolic dysfunction despite
having preserved ejection fraction and fractional shortening on conventional
echocardiography. In these patients, the occult, systolic abnormalities are
underestimated and have been shown to contribute to symptoms, morbidity and
mortality. Two-dimensional strain ( ) and strain rate imaging (SRI), are new
ultrasound (US) indices for quantifying regional wall deformation. Mitral annular
velocities derived from tissue Doppler imaging (TDI) have already been shown to
complement established parameters in evaluating early systolic and diastolic
performance post aortic valve replacement. However, the widespread use of this
methodology remains limited. This finding can likely be attributed to the fact that
Doppler-derived velocity and deformation data are one-dimensional, that is, only the
velocity and deformation component along an image line can be assessed, resulting in
an angle –dependency of the measurements. Tissue velocity 2D strain and strain rate
imaging with speckle tracking is a novel method of assessing regional as well as
global "contractility". This method overcomes many limitations inherent in assessing
myocardial functioning with current methodology, mainly it is reproducible,
objective, and is independent of myocardial translation, tethering and furthermore as
speckle tracking is derived from B mode images is independent of Doppler angle.
Charles Mugera 00395247 MSc Medical Ultrasound 5
The applicability of this technology to patients with aortic valve stenosis and
subclinical systolic dysfunction and its clinical significance has not been evaluated.
Objectives: The general aims of this study were to compare regional displacement,
tissue velocity, strain ( ) and strain rate (SR) in severe AS pre and post operatively.
Specifically, we sought, to investigate whether Speckle derived tissue velocity and
SR could be useful to detect subtle left ventricular (LV) dysfunction in patients with
severe aortic stenosis but preserved ejection fraction, and if they can reliably detect
improvements in regional myocardial function after aortic valve replacement (AVR).
We hypothesize that those patients with severe AS will have significantly reduced
peak systolic and peak early diastolic displacement, strain and strain rates at baseline
compared with normal controls, despite having normal ejection fraction and fractional
shortening as assessed by conventional echocardiography. In addition, we
hypothesise that post AVR; deformation patterns will show an early (4-16 weeks)
improvement in the myocardial strain and strain rate preceding any changes in LV
systolic and diastolic dysfunction assessed by conventional echocardiography.
Methodology: Our study prospectively analyzed 24 consecutive patients in total.
Ten control subjects (5 women, 5 men, and mean age 29.6 ± 5.7.3 years) provided
normal values of tissue displacement, velocity strain and strain rate. We then,
prospectively analyzed 10 patients with severe aortic stenosis and preserved EF and
FS, with speckle imaging derived tissue displacement, velocity, strain and strain rate
imaging pre –operatively (mean age 71.14± 16.15) and four of these patient post
operatively (mean age 79.25 ± 1.7) for aortic valve replacement. This was done as
part of routine pre-operative and post-operative transthoracic echocardiography using
Charles Mugera 00395247 MSc Medical Ultrasound 6
standard views, with the exception that all studies would need to be done on the GE
Vivid 7 digital ultrasound system. Speckle derived imaging data is derived from
standard B mode (Grey scale images), with a frame rate of between 40- 90 frames/s.
Post-operative TTE’s were performed at 4-20 weeks after discharge. Baseline
characteristics were taken from the standard pre-operative baseline study including
basic 2 D valve area, EF, wall thickness and geometry, and Doppler flow data.
Exclusion criteria: Patients with prior cardiac surgery including CABG or other valve
replacement, more than moderate mitral valve or aortic regurgitation and chronic
kidney disease (Cr >180), emergent aortic valve replacement, or endocarditis.
Informed consent was obtained on all patients, and they were only included if they
consent. No surgery was delayed for purpose of the study if the proper hardware/GE
VIVID system was not available.
Conclusion
Our results indicate that a reduction of displacement, tissue velocity, strain, and strain
rate may be a sensitive marker of subtle, subclinical, subendocardial myocardial
dysfunction. More over, these parameters seemed to be superior to conventional
echocardiography in detecting subtle improvements in myocardial function after
AVR before LV dimensions and LV function showed improvement.
Charles Mugera 00395247 MSc Medical Ultrasound 13
Chapter 1 Introduction and literature review
1.1 Introduction: Abnormalities in regional left ventricular (LV) function in severe
aortic stenosis (AS) have yet to be appropriately characterized1. Often patients with
severe aortic stenosis have subclinical left ventricular systolic dysfunction despite
having preserved ejection fraction and fractional shortening on conventional
echocardiography. However, when they have preserved ejection fraction (EF) and
fractional shortening, subtle, subclinical, LV systolic dysfunction may be
underestimated by the standard methods on routine echocardiography. In these
patients, the occult, insidious systolic abnormalities have been shown to contribute to
symptoms, morbidity and mortality. 2-5
Nihoyannopoulos P et al. showed that most of these patients show an early
improvement of ventricular diastolic function and LVMI regression after aortic valve
replacement assessed by conventional echocardiography6-8. However, assessment of
early improvement in the LV systolic dysfunction, by conventional
echocardiography, post aortic valve replacement, remains to be appropriately
elucidated. Therefore, a sensitive instrument is required to detect such subtle, occult,
dysfunction, in order to provide a guide as to the optimal timing of AVR as well as to
provide a tool to asses the beneficial effect of AVR on LV systolic function.
Mitral annular velocities derived from TDI have already been shown to complement
established parameters in evaluating systolic and diastolic performance in various
cardiac disorders9. Strain and Strain rate derived from tissue Doppler
Charles Mugera 00395247 MSc Medical Ultrasound 14
echocardiography have also been used as new indexes of assessing local myocardial
systolic and diastolic function post aortic valve replacement10. However, despite the
publication of numerous studies showing the additional information that can be
provided by Doppler-derived myocardial velocity and deformation data in both the
experimental and clinical setting, the widespread use of this methodology remains
limited11, 12. This finding can likely be attributed to the fact that Doppler-derived
velocity and deformation data are one-dimensional, that is, only the velocity and
deformation component along an image line can be assessed, resulting in an angle –
dependency of the measurements13. Although careful data acquisition can avoid any
major difficulty in data analysis and interpretation, it requires a certain level of
expertise and the associated training. Moreover, this angle-dependency decreases the
reproducibility of the measurement between observers and between studies. On top of
that, extracting meaningful deformation data requires manual tracking of the region of
interest throughout the cardiac cycle, which is a tedious and time-consuming task14-16.
Prior experimental studies using either magnetic resonance imaging (MRI) tagging
techniques or implanted microcrystals have analysed changes in regional deformation
in the settings of both acute and chronic pressure overload17, 18. However, these
studies, based on the analysis of rotational myocardial strains, documented only an
increased systolic torsion of the LV and delayed and prolonged diastolic untwisting.
The clinical data provided by MRI tagging on two dimensional abnormalities of
radial and longitudinal deformation in AS patients is limited and has been restricted
only to the MRI tagging evaluation of regional strain values, and not strain rates.
Charles Mugera 00395247 MSc Medical Ultrasound 15
This is because MRI currently has insufficient temporal sampling rates to resolve
regional strain rate (SR) profiles19.
All of the above-mentioned problems can be overcome if a method were available
that not only allows one to measure the velocity along the image line (cf. Doppler-
based methods) but also perpendicular to the image line, that is, in two dimensions.
Indeed, this strategy would allow reconstruction of any in-plane velocity/deformation
component, thereby solving the angle-dependency problem. Moreover, it would
enable automated in-plane tracking of the region of interest, speeding up the analysis
process tremendously 20.
Two-dimensional (2-D) motion, that is, velocity, estimation using ultrasound has
been an active field of research for many years, and multiple approaches have been
proposed, such as methods based on speckle tracking, multiple-beam Doppler, and
spatial modulation of the sound field21, 22. However, it has not been until relatively
recently that ultrasound image quality and computer capacity were adequate to have
some of these methods mature into practical research tools or commercial products.
1.1 Principle of speckle tracking
The fundamental principle of 2-D velocity estimation based on speckle tracking is as
follows: a particular segment of myocardial tissue shows in the ultrasound image as a
pattern of gray values. Such a pattern, resulting from the spatial distribution of gray
values, is commonly referred to as a speckle pattern. This pattern characterizes the
underlying myocardial tissue acoustically and is unique for each myocardial segment.
Charles Mugera 00395247 MSc Medical Ultrasound 16
It can, therefore, serve as a fingerprint of the myocardial segment within the
ultrasound image23. This unique pattern (fingerprint) is called Speckle. In the
Speckle tracking technique, a defined region (Kernel) is tracked, following a search
algorithm based on optical flow method. This simply means that if the position of the
myocardial segment within the ultrasound image changes, one can assume that the
position of its acoustic fingerprint will change accordingly. Tracking of the acoustic
pattern during the cardiac cycle within the ultrasound image thus allows one to follow
the motion of this myocardial segment within the (2-D) image24. The algorithm
searches for an area with the smallest difference in the total sum of pixel values,
which is the smallest sum of absolute differences25.
1.2 Physical origin of speckle
Ultrasound imaging is based on the pulse-echo technique: an ultrasound pulse is
transmitted, and subsequently the reflected echo signal is detected. Reflections occur
at transitions between different types of tissue (e.g., blood-muscle) or at specific sites,
much smaller than the wavelength, where the local sound velocity or mass density is
different from its surroundings (i.e., collagen fibres within the myocardium). The
latter reflections are relatively small in amplitude and are commonly referred to as
scatter reflections. The sites at which scattering occurs are defined as scattering sites
or simply scatterers26.
Each scatter will reflect the incident wave as it receives it but at lower amplitude
(determined by the exact acoustical and geometrical characteristics of the individual
scatterer). As myocardial tissue contains many scattering sites, the signal detected by
Ricevi informazioni sui nostri servizi, sulle offerte e non perdere news e consigli su università e lavoro.
