Download 2 - 张丽

Journal of Huazhong University of Science and Technology ［Med Sci］
DOI 10.1007/s11596-007-060x-x
1
27 (6): -, 2007
Assessment of Age-related Changes in Left Ventricular Twist by two
-dimensional Ultrasound Speckle Tracking Imaging*
ZHANG Li (张 丽), XIE Mingxing (谢明星)#, FU Manli(付曼丽),WANG Xin-fang(王新房), LV Qing(吕清) , HAN Wei(韩伟)，
ZHANGH Jing(张静), LIU Ying-ying(刘莹莹), WANG Jing(王静)，XIANG Fei-xiang(项飞翔)
Department of Ultrasound, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology; Hubei Provincial Key Laboratory of molecular Imaging , Wuhan 430022 , China
Summary: To assess the normal value of left ventricular twist (LVtw) and examine the changes with
normal aging by 2-dimensional ultrasound speckle-tracking imaging (STI), 121 healthy volunteers
were divided into three age groups: a youth group (19–45 y old), a middle-age group (46–64 y old )
and an old-age group (≥65 y old). Basal and apical short-axis images of left ventricular were acquired to analyse LV rotation (LVrot) and LVrot velocity. LVtw and LVtw velocity was defined as apical LVrot and LVrot velocity relative to the base. Peak twist (Ptw), twist at aortic valve closure
(AVCtw), twist at mitral valve opening (MVOtw), untwisting rate (UntwR), half time of untwisting
(HTU), peak twist velocity (PTV), time to peak twist velocity (TPTV), peak untwisting velocity
(PUV), time to peak untwisting velocity (TPUV) were separately measured. The results showed that
the normal LV performs a wringing motion with a clockwise rotation at the base and a counterclockwise rotation at the apex (as seen from the apex). The LVtw velocity showed a systolic counterclockwise twist followed by a diastolic clockwise twist. Peak twist develops near the end of systole
(96%±4.2% of end systole). With aging, Ptw, AVCtw, MVOtw, HTU and PUV increased significantly
(P<0.05) and Untw R decreased significantly (P<0.05). However, no significant differences in TPUV,
PTV and TPTV were noted among the 3 groups (P>0.05). It is concluded that LV twist can be measured non-invasively by 2-dimensional ultrasound STI imaging. The age-related changes of LVtw
should be fully taken into consideration in the assessment of LV function.
Key words：echocardiography; speckle tracking imaging; left ventricular; twist; rotation

Since the first description of left ventricular (LV)
twist motion by William Harvey in 1628 [1], the myocardium twist was being paid close attention to in the study
of cardiac mechanics. A great deal of evidence suggested
that cardiac twist plays an important role in LV contraction and relaxation. A newly developed 2-dimensional
ultrasound speckle tracking imaging (STI) technique can
trace the speckle spatial motions of myocardium. By
calculation and reconstruction of the motion and deformation of the myocardial tissue, accurate qualitative and
quantitative estimation can be achieved on cardiac mechanics features in cardiac cycle, according to motion
velocity, straining, strain rate, displacement, twist angles,
twist velocity and so on. The aim of this study was to
determine the normal value of left ventricular (LV) twist,
and to examine the effect of aging on LV twist by STI.
1 MATERIAL AND METHODS
1.1 Materials
1.1 Subjects
In this study, 121 healthy volunteers with sinus cardiac rhythm and without cardiovascular or systemic disZHANG Li, female, born in 1980, Doctoral candidate
E-mail: [email protected]
#Corresponding author
eases were subjected to physical checkup and conventional echocardiography. The study protocol was approved by the institutional review board of our hospital
and written informed consent was obtained from all participants before the enrollment.
Seven volunteers who had poor image quality on
2-dimentional echocardiography were excluded by STI
off-line. Thus, the final study group consisted of 114
subjects. To examine the effect of aging on twisting, the
subjects were divided into 3 age groups. The youth group
had 47 subjects (18–45y), middle-age group 38 (45–64y)
and old age group 29 (≥65y).
1.2 Instruments and Methods
Transthoracic echocardiograms were obtained by
using commercially available equipment (GE, USA)
(Vivid 7 Dimension, M3S probe, frequency 1.7–3.4
MHz). Echo PAC workstation with 6.0 version STI imaging analysis software was used. Volunteers in a supine
left lateral position with eupnea had ECG recording simultaneously. On short axis view, images of LV had to be
a round ring. Basal level was defined as the short axis
plane of 1/3 from basis on apex 4-chamber view in diastole. Apical level was defined as the short axis plane of
1/3 from apex on apical 4-chamber view in diastole. At
each plane, 3 consecutive cardiac cycles were acquired
during a breath hold, and were digitally stored in a hard
disk for offline analysis. The frame frequency was made
to be in line with the heart rate, whenever possible, in
storing image, so as to eliminate the inter-subject differ-
2
Journal of Huazhong University of Science and Technology ［Med Sci］ 27 (3): 2007
ences in heart rate.
With standard echocardiography, on five-chamber
of apex view. velocity of mitral valve in diastole (E),
velocity of mitral valve in end-diastole (A), deceleration
time of the E wave velocity (DcT), and isovolumic relaxation time (IVRT) were measured by pulse wave
Doppler. The biplane Simpson’s method was used to
measure left ventricular eject fraction(LVEF).
1.3 Echo PAC Workstation and Data Analysis
The original data were input into workstation. STI
mode was applied. The time interval between the peak of
R wave on the electrocardiogram and the aortic valve
opening and closure, and time from the peak R wave to
the mitral valve opening and closure, were measured by
using pulse wave Doppler from the LV outflow and inflow, respectively. Form the basal and apical LV
short-axis data sets, one cardiac cycle was selected for
subsequent analysis. The endocardial border of each
short axis in the end-systolic frame was manually traced
(at this point the endocardium is most clearly revealed).
The border of the endocardium of LV was first manually
delineated and the computer software then automatically
performed speckle tracking. After the tracking, the software will automatically divided the wall of the left ventricle into 6 segments and the results of the tracking were
reported as “V” (indicating satisfactory tracking) and
“X” (indicating unsatisfactory tracking). The width of
areas of interest was adjusted to cover the whole layer of
the myocardium. Six “V” sections were chosen to be for
further analysis. The system will automatically show the
rotational angle and rotation velocity curve throughout
the cardiac cycle.
Counterclockwise rotation as viewed from the LV
apex was expressed as a positive value, whereas a
clockwise rotation as a negative value. LV twist was
defined as apical rotation relative to base. Data depicting
the basal and apical LV rotation and rotational velocities
were exported to a spreadsheet program (Excel, Microsoft Corp, Seattle, Wash, USA) and the following
parameters were calculated: peak twist (Ptw), twist at
aortic valve closure (AVCtw), twist at mitral valve
opening (MVOtw), untwisting rate (Untw R), half time
of untwisting (HTU), peak twist velocity (PTV), time to
peak twist velocity (TPTV), peak untwisting velocity
(PUV) and time to peak untwisting velocity (TPUV).
Myocardial rotation in diastole was untwisting. The
degree of untwisting, the directional reversal of systolic
counterclockwise twist during diastole, was expressed as
percentage of twist at aortic valve closure: Untw R＝
(AVCtw－Twt/AVCtw)×100%, where t is any time
point during diastole, Twt is twist at time t and AVCtw is
twist at aortic valve closure. Because isovolumic relaxation time interval varied from volunteer to volunteer, the
untwisting rate was standardized as Untw R＝{[(AVCtw
－ MVOtw)/AVCtw]×100%}/IVRT, where MVOtw is
torsion at mitral valve opening and IVRT is the time of
isovolumic relaxation. Half time untwisting was the duration of ECG R-wave peak to half of twist peak angle.
In each subject, combining the echocardiographs
and electrocardiogram, we defined time point of aortic
valve closure as end-systole and time point of the peak of
R wave in follow cardiac cycle as end diastole. Cardiac
cycle was to standardized to about 60 points by software.
1.4 Statistical Analysis
Data were expressed as ±s. To estimate parametric variables among the 3 groups, ANOVA analysis
or Chi-square test were applied. SNK method was used
for cross-over comparison. A P value less than 0.05 was
considered to be statistical significance.
2 RESULTS
Seven of 121 volunteer (5.8%) were eliminated because their STI quality control analysis software indicates “X”. With age increasing, values of IVRT, DcT,
and A wave increased gradually (P <0.05). Values of E
and E/A decreased gradually (P<0.05). Body height,
weight, heart rate and ejection fractions showed no statistically significant differences among the 3 groups
(P>0.05) (table 1).
2.1 Characteristics of Normal Left Rotation and Left
Twist Angle Curve
The changes in normal left rotation angle and left
twist angle followed some patterns throughout the cardiac cycle(fig. 1,2). As seen from the apex, the normal LV
performs an initial trivial clockwise rotation at the apex
and trivial counterclockwise rotation at the base respectively in early systolic period. And it was followed by a
counterclockwise rotation at the apex and clockwise rotation at the base. Thus, the normal global LV performs
an early systolic twist motion with an initial minimal
clockwise twist which was followed by a counterclockwise twist (fig. 2C).
2.2 Characteristics of Normal Left Rotation and Left
Twist Velocity Curve
The changes in normal left rotation velocity and left
twist velocity followed some patterns throughout the
cardiac cycle (fig. 1, 2). As seen from the apex, the curve
of rotation velocity direction of the normal LV had an
initial trivial clockwise rotation at the apex and a trivial
counterclockwise rotation at the base respectively in
early systolic period. The change was followed by a
counterclockwise rotation at the apex and clockwise rotation at the base. Thus, the curve of the twist velocity
direction of normal global LV had an early systolic twist
motion with an initial trivial clockwise twist followed by
a counterclockwise twist. During diastole clockwise twist
dominated (fig. 2D) .
2.3 Comparison of Parameters of LV Twist and Rotation
With age increasing, Ptw, AVCtw, MVOtw, HTU
and PUV increased (P<0.05) and Untw R decreased
gradually (P<0.05). Whereas there were no statistical
differences in TPUV, PTV, TPTV and the time to peak
twist angle among the 3 groups (P>0.05). Thus, the time
to peak twist angle (the point at 96%±4.2％ of systole)
appeared in the end of the systole among the subjects.
TPTV appeared in mid-systole (the point at 56%±14%
of systole), whereas TPUV appeared at early diastole
(the point at 107%±12％ of the end of isovolumic relaxation).
3 DISCUSSION
3.1 The Principle of STI
Ultrasound STI technology is a novel technique for
the ultrasound quantitative analysis. It can quantitate the
Journal of Huazhong University of Science and Technology ［Med Sci］ 27 (6): 2007
velocity and straining of myocardial segments in
two-dimensional gray scale images. In the images, tiny
structures that shorter than the ultrasonic wavelength can
form scattering spots. These scattering bodies are echo
speckles which are about 20–30 pixels each. They are
distributed diffusely in the myocardial tissue uniformly
and closely follow the myocardial movement. The STI
software system can recognize the echo spots, and track
the location in each frame image in a real-time fashion,
map myocardial trajectory of the same location at different frame frequencies, calculate the rotational angle
and speed parameters of cardiac cycle.
3.2 Features of Anatomic Mechanics of Left Ventricular Twist
Torsion deformation of the left ventricle is a spiral
twist motion, which is related to the contraction of its
obliquely spiraling fibers[7]. Torrent-Guasp et al [8]found
that the orientation of the myofibers varies across the LV
wall: subepicardial fibers follow the path of a left-handed
helix around the cavity, fibers in the mid-wall are circumferentially oriented, whereas subendocardial fibers
follow a right-handed helical path. Stevens et al
[9]
examined the direction of cardiac muscle fibers of pigs
and found that the fibers from the endocardial to the epicardial ranged from +80° to –60°. In the normal LV,
because of the specific helical myofiber pattern, LV
twisting generated by the subepicardial fibers is partly
counteracted by the subendocardial myofibers. This
study showed that LV twist mainly manifestated as a
clockwise rotation at the base and counterclockwise rotation in the LV apex. Thus, the normal LV performed
mainly a counterclockwise twist throughout the cardiac
cycle. This result was consistent with previous findings[10-13]. The direction and size of ventricular twist was
dictated by transmural gradients of fiber strain and the
movement dominance of epicardial fiber relative to subendocardial fiber. Transmural straining in systolic period
was decreasing gradually from endocardial side to epicardial, from cardiac apex to base, which caused endocardial movement towards the cavity. Towards the apex,
the ratio between the radius of the epicardial and endocardial cavity of left ventricle became increasingly
greater. Therefore, the reversed strength of epicardial is
predominate. Therefore, the reversing force of the epicardial fibers dominated, resulting a greater counterclockwise torsion during systole. As a result, in normal
person, the rotation of base of left ventricle (clockwise)
and apex (counterclockwise) goes in opposite direction
[8,9]
.
3.3 The Effects of Aging on Left Ventricular Twist
The study found that the with age increasing, all indices of left ventricular twist increased and the results
were consistent with the findings by STI[11], TDI[12] and
MRI[13]. The reason could be the aging-related subepicardial layer myocardial fibrosis. This fibrotic degeneration leads to functional impairment because of myocardial hypoperfusion, which caused a dominance of epicardial counterclockwise rotation, relative to the clockwise rotation of subendocardial myocardial fibers and
thus ultimately strengthening the counterclockwise rotation in global left ventricular motion. Some studies
showed that patients who had left ventricular hypertrophy secondary to aortic stenosis[2], hypertrophic cardio-
3
myopathy[3] and asymptomatic type 1 diabetes with
blood sugar control[4] also had strengthened twist during
systole. The mechanism of this phenomenon is not fully
understood. This abnormal torsion may be one of the
sensitive prediction for the early identification of abnormal cardiac morphology.
Dong et al[5] reported that untwisting predominantly
occurred during isovolumic relaxation. It was consistent
with our results. Ventricular compliance and suction dictates the characteristics of untwisting of the ventricle.
Untwisting could directly regulate ventricular filling. The
rapid untwisting in early ventricular diastole is conducive
to ventricular pressure gradient, which encourages ventricular filling. The change of cardiac function from the
youth to old age, from normal function to lowered compliance and impaired function during systole, is a natural
physiological process. Normally, with aging, the heart of
the middle-aged and old people will suffer from amyloid
deposition and collagen degeneration, thereby leading to
decreased ventricular compliance and impaired diastolic
heart function. The study found that with increase of age,
untwisting rate was decreased, untwisting time prolonged,
untwisting velocity delayed. The exact mechanism by
which aging results in the mechanical and anatomical
changes of left ventricle awaits further study. However,
our study suggests that in the evaluation of the diastolic
function left ventricle, the age should be taken into consideration.
3.4 Limitations
Application of STI is still confined to the
two-dimensional observation, and it can not fully follow
the spatial movement and it is not accurate enough. The
poor quality of two-dimensional image and the unclear
border of endocardium are two other limitations. But the
information it provides is highly correlated with the
findings of TDI and MRI[12,13]. Our study indicated that
STI can serve as a reliable technique in the study of cardiac mechanics.
REFERENCES
1
Harvey W. An anatomical disposition on the motion of the
heart and blood in animals, 1628. In: Willis F A, Keys T E,
eds. Cardiac Classics. London, England: Henry Kimpton.
1941.19-79
2
Nagel E, Stuber M, Burkhard B et al. Cardiac rotation and
relaxation in patients with aortic valve stenosis. Eur Heart
J, 2000,21:582-589
3
Young A A, Kramer C M, Ferrari V A et al. Three- dimensional left ventricular deformation in hypertrophic cardiomyopathy. Circulation,1994,90:854-867
4
Chung J, Abraszewski P, Yu X et al. Paradoxical increase
in ventricular torsion and systolic torsion rate in type 1
diabetic patients under tight glycemic control. J Am Coll
Cardiol, 2006,47:384-390.
5
Dong S, Hees P, Hees P et al. MRI assessment of LV relaxation by untwisting rate: a new isovolumic phase
measure of tau. Am J Physiol Heart Circ Physiol,
2001,281:H2002-H2009
6
Meunier J, Bertrand M. Ultrasonic texture motion analysis:theory and simulation. IEEE Trans Med Imaging,
1995,14:293-300
7
Taber L A, Yang M, Podszus W W. Mechanics of ventricular torsion. J Biomechanics, 1996,29:745-752
8
Torrent-Guasp F, Buckberg G D, Clemente C et al. The
structure and function of the helical heart and its buttress
4
Journal of Huazhong University of Science and Technology ［Med Sci］ 27 (3): 2007
wrapping. I.The normal macroscopic structure of the heart.
Semin Thorac Cardiovasc Surg, 2001,13：301-319.
9
Stevens C, Remme E, LeGrice I et al. Ventricular mechanics in disastole : material parameter sensitivity. J
Biomech, 2003,36:737-748
10 Luo A G, Yin L X, LI C M et al. Measure of left ventricular torsion with cardiac pacing and right bundle branch
block patients by two-dimension ultrasound speckle
tracking imaging. Chin J Ultrasonogr (Chinese),
2006,15:641-645
11 Takeuchi M, Nakai H, Kokumai M et al. Age-related
changes in left ventricular twist assessed by
two-dimensional speckle-tracking imaging. J Am Soc
Echocardiogr, 2006,19:1077-1084
12 Notomi Y, Srinath G, Shiota T et al. Maturational and
adaptive modulation of left ventricular torsional biomechanics: Dopple tissue imaging observation from infancy
to adulthood. Circulation, 2006,113,2524-2533
13 Oxenham H C, Young A A, Cowan B R et al. Age-related
changes in myocardial relaxation using three-dimensional
tagged magnetic resonance imaging. J Cardiovasc Magn
Reson, 2003,5:421-430.
(Received March 29, 2007)