(Circulation. 1995;92:842-853.)
© 1995 American Heart Association, Inc.
Articles |
Assessment of Cardiac Function by Three-dimensional Echocardiography Compared With Conventional Noninvasive Methods
Presented in part at the 66th Scientific Sessions of the American Heart Association, Atlanta, Ga, November 8-11, 1993; the 43rd Annual Scientific Session of the American College of Cardiology, Atlanta, Ga, March 1994; and the 5th Scientific Sessions of the American Society of Echocardiography, San Francisco, Calif, June 1994.
From the Columbia University College of Physicians and Surgeons, Division of Cardiology, New York, NY (A.S.G., Z.S., M.J.S., D.W.L., O.O.A., R.A.R., D.K.B., D.L.K.); the University of Kentucky Medical Center, Division of Cardiology, Lexington (P.M.S.); and Danbury Hospital, Division of Cardiology, Danbury, Conn (A.M.K.).
Correspondence to Aasha S. Gopal, MD, Division of Cardiology, Columbia University, Atchley Pavilion 552, 161 Ft Washington Ave, New York, NY 10032.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background Reliable, serial, noninvasive quantitative estimation of left ventricular ejection fraction is essential for selecting and timing therapeutic interventions in patients with heart disease. Equilibrium radionuclide angiography is widely used for this purpose but has well-recognized limitations. Advantages of echocardiography over equilibrium radionuclide angiography include assessment of wall motion, valvular pathology, and cardiac hemodynamics, in addition to portability, lack of radiation exposure, and substantially lower cost. However, conventional echocardiographic techniques are limited by geometric assumptions, image positioning errors, and use of subjective visual methods. To overcome these limitations, a three-dimensional echocardiographic method was developed. This study compares ejection fraction by three-dimensional echocardiography, quantitative two-dimensional echocardiography, and subjective two-dimensional echocardiographic visual estimation with that by equilibrium radionuclide angiography.
Methods and Results Fifty-one unselected patients with suspected heart disease underwent left ventricular ejection fraction determination by equilibrium radionuclide angiography and three-dimensional echocardiography using an interactive line-of-intersection display and a new algorithm, ventricular surface reconstruction, for volume computation. In 44 patients, ejection fractions were also estimated visually by experienced observers from two-dimensional echocardiography and by quantitative two-dimensional echocardiography using an apical biplane summation-of-disks algorithm. An excellent correlation was obtained between three-dimensional echocardiography and equilibrium radionuclide angiography (r=.94 to .97, SEE=3.64% to 5.35%; limits of agreement, 10.3% to 13.3%) without significant underestimation or overestimation. SEE values and limits of agreement were twofold to threefold lower than corresponding values for all two-dimensional echocardiographic techniques. In addition, interobserver variability was significantly lower for the three-dimensional echocardiographic method (10.2%) than for the apical biplane summation-of-disks method (26.1%) and subjective visual estimation (33.3%).
Conclusions Determination of ejection fraction by three-dimensional echocardiography yields results comparable to those obtained by equilibrium radionuclide angiography and is substantially superior to all two-dimensional echocardiographic methods. Therefore, three-dimensional echocardiography may be used for accurate serial quantification of left ventricular function as an alternative to equilibrium radionuclide angiography.
Key Words: echocardiography • heart function tests • imaging • computers
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Left ventricular (LV) systolic function is the most important prognostic factor in a variety of cardiac disorders, and accurate, serially reproducible, noninvasive methods for measuring left ventricular ejection fraction (LVEF) are essential for the management of patients.1 2 3 Equilibrium radionuclide angiography has been widely used for this purpose because it is noninvasive and does not rely on assumptions of LV geometry.4 5 6 7 8 However, it requires both the injection of radiopharmaceutical agents and time-consuming computer and operator processing of acquired image data. In addition, it has limited resolution for assessing regional ventricular function. Echocardiography has the advantage of complete safety, the immediate provision of clinically interpretable images, and superior ability to assess regional wall motion as well as to provide information about valvular function and cardiac hemodynamics. Although echocardiography has been widely used to measure LVEF, currently available echocardiographic methods are nonetheless subject to important limitations. One-dimensional M-mode methods are based on assumptions that become invalid in patients with abnormally shaped ventricles.9 To achieve further improvement, several algorithms for quantitatively estimating LVEF from two-dimensional (2D) images have been developed and applied.10 11 12 13 14 15 16 17 18 19 20 21 These methods, however, are limited by geometric assumptions regarding the shape of the ventricle and the position and orientation of imaging planes. Errors in correctly positioning the imaging transducer may be caused by the overlying ribs. These limitations can result in significant errors and interobserver variability.22 23 Consequently, subjective visual estimates of LVEF based on 2D echocardiographic images, although inherently observer dependent, are widely used and thought by many to be adequate for routine clinical use.10 24 25 26 Recently, quantitative three-dimensional (3D) echocardiographic methods have been developed to overcome the major limitations of 2D echocardiography and to improve the quantitative accuracy and reliability of echocardiography.27 28 29 30 31 32 33 34 35 36 The availability of additional spatial data in 3D echocardiography makes it possible to apply a new algorithm, ventricular surface reconstruction, for computation of volume37 38 39 40 and hence LVEF. With these, end-diastolic and end-systolic LV volumes have been accurately quantified in animal models41 and in normal human subjects.42 In patients with heart disease, however, only preliminary data exist for estimation of LV volume and LVEF, and none are available comparing the 3D echocardiographic method with other noninvasive methods currently in use. Therefore, this study was undertaken to compare LVEF by 3D echocardiography with LVEF by equilibrium radionuclide angiography in unselected patients. Additionally, we compared 3D echocardiography with 2D echocardiographic subjective visual estimation of LVEF by experienced observers and also with estimation of LVEF by the 2D quantitative apical-biplane summation-of-disks method recommended by the American Society of Echocardiography.43
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Patients
Fifty-two unselected patients presenting to the Nuclear Cardiology Laboratory at Columbia-Presbyterian Medical Center for clinically indicated radionuclide angiographic studies were enrolled in the study. All patients gave informed consent in accordance with Columbia University's Institutional Review Board. One patient was excluded because of technically inadequate echocardiographic images. Hence, 51 patients (32 men, 19 women) 10 to 82 years old (mean, 51.2±16.5 years) and body surface areas ranging from 1.33 to 2.28 m2 (mean, 1.83±0.21 m2) constituted the final study group (Table 1
). Six
patients were in atrial flutter or fibrillation.
Patient diagnoses were as follows: regional wall motion abnormality due
to coronary artery disease, 20; dilated
cardiomyopathy, 10; normal, 8; valvular
heart disease, 6; LV hypertrophy, 4; congenital heart
disease, 1; primary pulmonary hypertension, 1; and cor
pulmonale, 1. All enrolled patients underwent 3D
echocardiography within 2 to 4 hours after
radionuclide angiography. All except the first 7 patients (n=44; 27
men, 17 women; 10 to 82 years old, mean, 50.1±17.3 years) with body
surface areas ranging from 1.33 to 2.28 m2 (mean,
1.82±0.20 m2) (Table 2) also underwent a
conventional 2D echocardiographic examination
immediately after the 3D echocardiographic examination.
No change in patient status or medical regimen occurred during the
period of data acquisition.
|
|
3D Echocardiography
Instrumentation
The
features of our 3D echocardiographic system
have been described.34 35 36 The 3D
echocardiographic system (K3 Systems, Inc) is composed
of an acoustic spatial locater (model GP 8-3D, Science Accessories
Corp) and personal computer (model 4DX-33V, Gateway 2000). These
components are linked to a conventional 2D echocardiograph
(model 77020AC, Hewlett-Packard Corp). The conventional 2.5-MHz
ultrasound transducer is fitted with three rigidly mounted sound
emitters that are energized in sequence. The sound waves emitted at a
frequency of 60 kHz are received by an array of four microphones
positioned approximately 0.75 m above the patient. The sound emitters,
overhead microphone array, and their electronics compose the acoustic
spatial locater. The locater associates 3D spatial coordinates with
each echocardiographic image by measuring the time for
the sound to travel between the sound emitters and each overhead
microphone. Distances or ranges are calculated from the elapsed times
and used to compute x, y, and z
cartesian coordinates of each sound emitter and thus the transducer and
its images. The acoustic spatial locater is accurate (
0.75 to 1 mm)
to within 0.1% of the distance from the sound emitters to the
microphones. At the time of each data acquisition, the system performs
an internal check of the range accuracy by calculating the distances
and angles between each of the three sound emitters. If significant
transducer motion occurs during data acquisition, these calculated
values will vary from the actual values. If this variation is more than
1 mm or 1°, the data are automatically discarded. ECG-gated real-time
echocardiographic images are transmitted to the
personal computer, digitized, and stored, together with their spatial
coordinate data, in cine-loop format for future retrieval and
analysis. Each cine loop consists of a sequence of 16 images.
Every other video frame is acquired, beginning with the QRS complex.
The cine loop may be advanced and reversed at variable speed as
well as frame by frame for viewing. Typically, one
diastolic interval and two systolic intervals are captured
in each cine loop at normal heart rates.
Image Acquisition
Using the Line-of-Intersection Display
LV volume is computed from a
series of real-time parasternal
short-axis images acquired with a novel "line-of-intersection"
display as a guide.35 The line-of-intersection display is
created by first obtaining a parasternal long-axis reference image. Any
subsequent real-time short-axis image intersects the reference
long-axis image, creating a single line common to both images, the line
of intersection. This line is computed and displayed in each image.
When the transducer is moved, the relative position of the two images
is changed. The line of intersection is rapidly recomputed and
redisplayed several times per second as each new set of range data is
acquired. The line therefore appears as an interactive moving line in
each image and indicates the position and orientation of that image
with respect to the other image. The operator may press a function key
on the computer keyboard to view the reference and real-time images
alternately and observe their changing relations as shown by the line
of intersection. The display of the line of intersection in the
reference image is used by the operator to guide the positioning of the
real-time short-axis images during acquisition of a data set for volume
computation. By using the display, the operator is able to ensure that
the short-axis images adequately represent the ventricle and
are optimized for endocardial boundary identification. The short-axis
images are spaced appropriately from the inferior surface
of the aortic valve throughout the body of the left ventricle to the
apex. For computation of LV volume, a typical data set is composed of 7
to 10 short-axis images positioned as shown by the lines of
intersection in Fig 1. These short-axis images need not
be parallel or evenly spaced but must not intersect in the region of
interest, the left ventricle. Intersection of images in the region of
interest is avoided by use of the line-of-intersection display. All
images for ventricular reconstruction are acquired during
suspended respiration. Two separate sets of data are acquired for
computation of volume during diastole and systole using
end-diastolic and end-systolic reference images to account
for translation of the ventricle. The time currently required for
acquisition of each data set is typically 6 to 8 minutes, depending on
patient-related factors. Data acquisition time may be prolonged to 10
to 15 minutes in patients who are difficult to image.
|
Surface Reconstruction and Volume Computation
A
polyhedral surface reconstruction algorithm has been
adopted for LV volume computation using the traced endocardial borders
of each short-axis image.42 End-diastolic
(onset of QRS complex) and end-systolic (smallest chamber dimension)
video frames from each acquired cine loop are selected for off-line
boundary tracing. The traced borders are divided into 180 equidistant
boundary points by interpolation. A centroid is automatically defined
for each traced boundary. Consecutive pairs of corresponding points on
adjacent boundaries are connected, forming a pair of triangles that
define the surface between the points. Each surface pair of triangles
is also connected to the centroid of each boundary so as to define a
solid sector or wedge. A total of 180 wedges thus enclose the volume
between the two boundaries. Each wedge is then decomposed into three
tetrahedrons. The volumes of the 540 tetrahedrons are calculated and
summed to yield the volume between the two boundaries. Volumes between
all boundaries are in turn calculated and summed to yield the total
volume of the ventricle. Thus, end-diastolic and
end-systolic volumes are computed from their respective data sets, and
LVEF is calculated from these values. Wire frame computer
reconstructions depicting the traced boundaries are used to ensure
proper orientation of short-axis images and reconstruction of the
ventricle. Fig 2 shows wire frame models of ventricles
with LVEFs of 15% and 66%.
|
Procedure
A single 3D
echocardiographic examination was
performed by the same examiner on all patients. Two observers blinded
to the results of the radionuclide angiographic studies independently
selected end-diastolic and end-systolic video frames and
manually traced the respective endocardial boundaries using specific
border tracing conventions. Tracing of the endocardium was performed on
the white side of the black-and-white boundary. Papillary muscles were
included as part of the cavity volume when they were noted to be
discontinuous with the myocardium. When they were
continuous with the myocardium and the boundary outlining
the papillary muscle was greater than a semicircle, the papillary
muscle was included as part of the cavity volume by continuation of the
endocardial boundary through the muscle. When they were continuous with
the myocardium and the boundary outlining the papillary
muscle was less than a semicircle, the papillary muscles were excluded
from the cavity volume by tracing of the endocardium around the
papillary muscle. The time required for tracing endocardial borders was
approximately 1 minute per boundary, or a total of 8 or 9 minutes for
each data set. Each observer traced each data set twice or more until
two of his or her results were within 10% of each other. These two
results were averaged, taken as the volume for each data set, and used
for calculation of LVEF.
2D Echocardiography
LVEF by Visual Estimation
After the 3D echocardiographic study, a
conventional 2D echocardiographic examination was
performed using the same 2.5-MHz transducer and ultrasound scanner.
Standard parasternal long-axis and short-axis and apical two-chamber,
four-chamber, and long-axis views were recorded on 1/2-in VHS
videotape. LVEF was estimated subjectively from these views
independently by two blinded experienced observers (with >3 years'
experience in interpretation of echocardiograms).
LVEF by
Quantitative 2D Echocardiography
The apical two-chamber and
four-chamber views were used to
calculate end-diastolic and end-systolic volumes according
to the apical biplane summation-of-disks algorithm recommended by the
American Society of Echocardiography using an
average of two or three beats.43 End-diastolic
(onset of QRS complex) and end-systolic (frame just before mitral valve
opening) video frames were selected, digitized, and traced by two
independent observers using a video display analysis system
(Color Workstation Freeland Medical System or Nova Microsonics). The
border tracing convention applied was the same as that used for
analysis of 3D echocardiographic images.
Equilibrium
Radionuclide Angiography
Equilibrium radionuclide angiography was done
by the modified in
vivo method of labeling autologous red blood cells.44 The
patient was injected with 1.5 mg stannous pyrophosphate (TechneScan
PYP, Mallinckrodt, Inc). Twenty minutes later, 3 mL of the patient's
blood was withdrawn into a heparinized syringe containing 30 mCi
99mTc pertechnetate (Mallinckrodt, Inc). After a 10-minute
incubation period, the blood was reinjected into the patient, resulting
in distribution of the tracer throughout the blood pool. Imaging was
performed in the "best septal" left anterior oblique (30° to
40°) view with a digital Anger camera (Picker Dynacamera) with a
slant-hole collimator positioned at a caudal angulation of 30° and
interfaced with a Medasys Pinnacle computer system. Acquisition was
synchronized with the R wave of the ECG. Beat rejection windows were
defined by use of a histogram of RR intervals. Images were obtained in
a 64x64-pixel matrix format with 250 000 counts per image in 28
frames redisplayed in a cine-loop format. Background subtraction was
performed with a curvilinear area of interest lateral to the left
ventricle. End-diastolic and end-systolic frames were
defined by use of the time-activity curve. LVEF was calculated by use
of standard software by a single experienced observer within regions
manually drawn around the left ventricle in diastole and
systole.
Statistical Analysis
The demographic data obtained for the
entire cohort of patients
(n=51) are described in terms of ranges, means, and SDs. The 3D
echocardiographic method was compared with radionuclide
angiography by simple linear regression and Pearson's correlation
coefficient for each observer. The regression equations for each were
compared with the line of identity by use of the F test.45
An analysis of the limits of agreement between 3D
echocardiography and radionuclide angiography was
also performed as described by Altman and Bland.46 For
this analysis, the difference between the two measurements was
plotted against their mean. Then, the bias (mean difference between
radionuclide angiography and echocardiography) and
the limits of agreement (2 SD of the difference) were determined.
For the 44 patients who also underwent 2D echocardiography, linear regression and Pearson's correlation coefficient were determined for each observer. A limits-of-agreement analysis46 between radionuclide angiography and each echocardiographic method was also performed. The results of both observers were averaged in the limits-of-agreement analysis to simplify the comparison between the three echocardiographic methods (3D echocardiography, quantitative 2D echocardiography, and subjective visual estimation). The nonparametric sign test was used to compare the magnitudes of the difference between radionuclide angiographic and echocardiographic LVEF for each echocardiographic method.47 Interobserver variability was calculated as the SD of the differences between the two operators expressed as a percent of the average value. The significance of the difference in interobserver variability among the three echocardiographic techniques was tested with the F test for homogeneity of variances.48
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Comparison of LVEF by 3D Echocardiography With Radionuclide Angiography
LVEF estimation by 3D echocardiography was compared with equilibrium radionuclide angiography in 51 patients (Table 1
). In this group, the mean LVEF was 46.4±18.7%
(range, 9% to
75%). Table 3 summarizes the statistical comparison of
the two methods. The SEEs were 5.35% and 3.64% and the limits of
agreement were 13.3% and 10.3% for each observer, respectively. The
regressions for each observer were significantly different from the
line of identity by the F test over the entire range of LVEFs (9% to
75%). However, over an LVEF range of 30% to 65%, the regression
equations describing the relation between 3D
echocardiography and multigated radionuclide
angiography did not differ from the line of identity by the F test. Fig
3 depicts the linear regression and limits-of-agreement
plots for each observer. There was no significant systematic error by
the 3D echocardiographic method for either observer.
Interobserver variability of LVEF by 3D
echocardiography was 10.2%.
|
|
Comparison of LVEF by 3D Echocardiography and
2D Echocardiography With Equilibrium
Radionuclide Angiography
LVEF estimations by 3D echocardiography and
by
both subjective and quantitative 2D echocardiographic
methods were compared with that by equilibrium radionuclide angiography
in all but the first 7 patients (n=44) (Table 2). The
mean LVEF was
46.7±18.9% (range, 9% to 75%). Table 4 summarizes
the results of the statistical comparison. Fig 4 depicts
the linear regression plots for each method and observer. The
regression equations for each technique and observer were found to be
significantly different from the line of identity by the F test over
the entire range (9% to 75%) of LVEFs but not significantly different
from the line of identity over an LVEF range of 30% to 65%.
Limits-of-agreement analysis was performed using the averaged
values of both observers for each method (Fig 5). The
limits of agreement for 3D echocardiography by each
observer were 13.3% and 9.4%. These values are approximately 1.5 to 2
times lower than the corresponding values for subjective visual
estimation and quantitative 2D echocardiography
(Table 4). Significant systematic underestimation or
overestimation of
LVEF did not occur by the 3D echocardiographic method.
However, underestimation was detected in the values of observer 1 by
the quantitative technique (bias, +6.84) and observer 2 (experienced)
by visual estimation (bias, +8.30). Interobserver variability of 3D
echocardiography (10.3%) was significantly lower
(P<.01) by the F test for homogeneity of
variances48 than interobserver variability obtained by
quantitative 2D echocardiography (26.0%) and
visual estimation by experienced observers (33.8%). The magnitude of
the difference between 3D echocardiography and
radionuclide angiography was significantly less than that of
quantitative 2D echocardiography
(P<.05) and visual estimation by experienced observers
(P<.002) as determined by the nonparametric
sign test.47 The sign test did not reveal any significant
differences among the conventional echocardiographic
techniques.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The comparison of 3D echocardiography with equilibrium radionuclide angiography in this study showed the two methods to be comparable in their results. 3D echocardiography yields an LVEF in unselected patients that correlates very well (r=.94 to .97) with that from equilibrium radionuclide angiography with a small SEE (4% to 5%). The limits-of-agreement analysis shows no significant systematic underestimation or overestimation (+0.57% to +1.37%) by 3D echocardiography and a very small SD of the differences between the two methods (5.14% to 6.65%). These data indicate that 95% of the time, the results of the two studies will be within 10% to 13% of each other. We judge that this small variation is acceptable from a clinical point of view and conclude that 3D echocardiography represents an acceptable alternative to equilibrium radionuclide angiography for measurement of LVEF.
The comparison of each of the conventional 2D echocardiographic techniques with equilibrium radionuclide angiography revealed no significant differences among them, although significant differences from 3D echocardiography were noted. The 2D echocardiographic correlations were not as high and the SEEs were larger than those of 3D echocardiography. The limits-of-agreement analysis showed underestimation by 2D echocardiography in several instances and significantly larger SDs of the differences. These data indicate that 95% of the time, the results of the 2D echocardiographic methods and equilibrium radionuclide angiography will be within 19% to 21% of each other, limits almost double those of 3D echocardiography. Interobserver variability of the 2D echocardiographic techniques was two to three times greater than that of 3D echocardiography. From these results, we conclude that determination of LVEF by 3D echocardiography is clearly superior to subjective visual estimation and to the quantitative 2D echocardiographic methods recommended by the American Society of Echocardiography. As a matter of interest, we also compared 3D echocardiography with a method incorporating both a short-axis and an apical view (5/6 area-length method) described by Wyatt et al.49 The results showed a similar superiority of 3D echocardiography over this method (r=.77, SEE=11.6 mL; bias, 2.16%; limits of agreement, 25.1%; interobserver variability, 25.2%). We also conclude that when 2D echocardiography is used to determine LVEF, there is no added benefit from quantitative analysis compared with subjective visual estimation.
The regression equations describing the relation between each of the echocardiographic techniques and equilibrium radionuclide angiography differed significantly from the line of identity for the entire range of systolic function. This relation was not particular to 3D echocardiography but was found for conventional echocardiographic methods as well. Because of this result and previous reports of variability of the radionuclide LVEF at the high and low ends of the spectrum of systolic function,5 8 50 51 52 the extremes of the LVEF range (0% to 30% and 65% to 75%) were excluded and the analysis was repeated. The results of the repeated analysis showed that the regression equations for 3D and 2D echocardiography versus equilibrium radionuclide angiography were not statistically different from the line of identity for the LVEF range 30% to 65%, thus confirming that both methods were in fact measuring the same parameter.
Comparison to Previous Studies
To date, previous human
studies comparing LVEF by 3D
echocardiography with accepted clinical modalities
have not been published. Only preliminary data involving comparisons
with invasive cineventriculography
exist.53 Martin and Bashein33 have measured
stroke volume with 3D transesophageal ultrasonic
scanning. Siu et al41 demonstrated in an in vivo,
open-chest animal preparation that LVEF can be measured accurately
(r=.96, SEE=5.64%) by 3D
echocardiography. The results obtained in this
study are consistent with the above studies as well as with
results obtained in our previous in vivo work, in which a twofold to
threefold improvement was obtained by 3D
echocardiography over conventional 2D
echocardiography in the measurement of cardiac
dimensions22 23 and calculation of LV
volume.42
The correlation coefficients and SEEs obtained by comparison of the quantitative 2D echocardiographic technique with equilibrium radionuclide scintigraphy in our study are consistent with those obtained by previous investigators.12 17 18 20 21 24 Despite extensive boundary tracing and off-line analysis, these methods have been shown to provide little or no improvement over visual estimates of global LVEF10 24 or estimates based on wall-motion scores.26 Although use of subjective visual estimation of global LVEF is widespread, only a few studies have formally examined the reproducibility of this approach. In those studies, reasonably high correlation coefficients were obtained. However, 95% CIs have varied from 10.6% to 24.6%.10 24 In a study comparing subjective and quantitative 2D echocardiographic estimation of LVEF with radionuclide angiography, Amico et al24 report only a moderate correlation between two observers: r=.77, SEE=9.7%; mean percentage difference, 23±37%. Their conclusion was that the LVEF estimate by quantitative echocardiography was worse than subjective visual estimation.24 In our study, there was no significant difference between the two methods.
Limitations of Conventional 2D
Echocardiography
Conventional echocardiographic systems allow the
operator to obtain tomographic images in an almost infinite variety of
positions and orientations using the hand-held transducer. However,
they have no means of localizing these images in 3D space or measuring
the spatial relation between consecutively acquired images for volume
computation. Hence, assumptions about ventricular geometry
and the position of the imaging plane are necessary to compute LVEF by
these techniques. The operator performing the
echocardiographic examination must rely solely on image
content and knowledge of cardiac anatomy to position images for
volume computation. This reliance on observer skill leads to
significant interobserver variability and error in positioning and
measuring images that adequately represent the left
ventricle.22 23
The apical biplane summation-of-disks method recommended by the American Society of Echocardiography43 is based on visualizing the left ventricle adequately from the apex. However, an adequate apical echocardiographic window is unavailable in a large percentage of patients because of the interposition of ribs.24 As a result, >90% of apical views are foreshortened,19 yielding an inaccurate linear dimension of the long axis of the ventricle, a critical measurement used in this method. The method further assumes but does not verify an orthogonal relation between the apical two- and four-chamber views. In a previous study using the line-of-intersection display to assess the positioning of standard echocardiographic views, the apical four-chamber and two-chamber views were optimally positioned in the same study with respect to displacement and angulation in only 12% of examinations.22 These limitations also led to high SEEs and interobserver variability in the measurement of LV dimensions.23 Thus, despite careful off-line analysis, quantitative 2D echocardiographic methods do not offer any advantage over subjective visual estimation of LVEF.10 24
Subjective visual assessment by conventional echocardiography permits only a qualitative descriptive estimation of LVEF, although specific numerical values are commonly reported as a result. This approach may be adequate for the rapid assessment and management of patients in the acute setting. However, it is not sufficient for serial evaluation of patients, in particular those with regurgitant valvular lesions, in whom cardiac size and systolic function are critical for the proper timing of therapeutic interventions. High interobserver variability in subjective estimation of LVEF by conventional echocardiography often results in requests for duplicate testing by alternative imaging modalities, study review, and reinterpretation.
Advantages and Limitations of 3D
Echocardiography
The advantages of 3D echocardiography derive
from the recording of new spatial data not previously
available. These spatial data define the position and orientation of
each echocardiographic image, and thus each visualized
anatomic point, in an independent 3D spatial coordinate system. The
availability of these data makes it possible to create the
line-of-intersection display of the 3D spatial relation of a series of
2D images. The line-of-intersection display enables the operator to
more easily and accurately position a real-time image with respect to
its nonvisualized z dimension orthogonal to the
x-y plane of the image. If one can see the line of
intersection in a reference image, errors in positioning real-time
images are recognized and thus eliminated or minimized. In addition,
the need to avoid intersecting LV cross-sectional images dictated by
the polyhedral surface reconstruction algorithm may be fulfilled by use
of this display method during acquisition. Conventional measurements
are then obtained more easily, rapidly, and accurately.
The 3D data set also makes it possible to compute 3D parameters such as ventricular volume, surface area, mass, and LVEF without the use of geometric assumptions.42 54 55 These computations are achieved by use of a ventricular surface reconstruction algorithm that uses more data than conventional methods and, by its nature, accounts for the variation of abnormally shaped ventricles. This algorithm also makes it possible to compute not only global parameters but also regional parameters of abnormal wall motion, such as regional subtended volume, mass, surface area, and regional ejection fraction.55 From these regional parameters, new secondary parameters can be derived. An example is an index of infarct expansion or aneurysm formation based on the ratio of the regional infarct surface area to subtended volume.56
The cost of adding 3D capability to existing 2D scanners is relatively low. Its use is practical because of the high accuracy and reproducibility of the clinical data produced and because these data can be obtained at a lower cost than by equilibrium radionuclide angiography. The spatial locater may be attached to the transducer without altering the instrument or affecting its use as a 2D scanner, thus preserving and extending the useful life of a major capital investment. Operation of the 3D system by a sequence of computer screen menus is easily learned by trained sonographers. Use of the line-of-intersection display to guide image positioning is a new skill that most experienced sonographers acquire with a few hours of practice.
A limitation at present is the additional time required to perform the 3D study of the left ventricle after the standard 2D echocardiographic examination is complete. At the present time, 6 to 8 minutes is typically required to collect a single data set. We anticipate that the 3D echocardiographic study will be incorporated into a single examination protocol, so that eventually all data acquisition will require no more time than a 2D study currently does.
The need for manual endocardial boundary tracing also represents a limitation of 3D echocardiography, as it does for all other imaging modalities. Manual tracing can result in technical variation and is labor intensive. It is a step that is not easily automated because of the comparatively low density and resolution of ultrasound data, the presence of acoustical noise and artifacts, and the great variety of contours to be traced. Boundary tracing, however, is very amenable to performance by a trained sonographer. At the present time, about 1 minute is required to retrieve, view, and trace each boundary. We anticipate that the benefits to be obtained by boundary tracing will eventually be so great that this method will be considered essential and the most effective approach to interpretation of most abnormal echocardiograms. Physicians interpreting 3D echocardiographic studies may check and verify the quality of image placement and boundary tracing and either approve or revise the work of the sonographer. Therefore, boundary tracing by sonographers may actually reduce the time required for study interpretation by the physician.
3D Echocardiography: An Alternative to
Equilibrium Radionuclide Angiography
Echocardiography already provides
significant
information about wall motion abnormalities, valvular
morphology, and hemodynamics not available from the
equilibrium radionuclide angiogram. The results of this study suggest
that 3D echocardiography may be used as well for
measurement of LVEF as an alternative to equilibrium radionuclide
angiography in patients with satisfactory
echocardiographic windows. If this is done, substantial
savings may result. The hardware and software necessary for 3D
echocardiographic studies can be added to existing 2D
echocardiographic machines at relatively low cost, and
this allows the provision of more accurate measurements of LVEF
contemporaneously with the acquisition of standard 2D
echocardiographic information about regional wall
motion, cardiac valvular function, and intracardiac
hemodynamics. As an additional benefit, 3D
echocardiographic studies may be repeated frequently
for serial patient follow-up without the hazards of radiation exposure
posed by radionuclide angiography.
Study Limitations and Sources of Error
In our previous in
vitro investigations, our 3D
echocardiographic system proved to be highly accurate
and did not introduce new measurement errors.36 The
acoustic spatial locater error or variation is approximately 0.1% of
the measured distances.57 The mean system error in
measuring volumes in vitro is approximately 0.4% of true
volume.36
The in vivo method of data acquisition used in this study is to place and hold the imaging transducer in a desired location, acquire a single set of spatial coordinate data, and subsequently acquire a series of 16 video images over the next second. Hence, a single spatial data set is assigned to each image. If respiratory or whole-body motion occurs during data acquisition, subsequent data will be misregistered with respect to data acquired before the motion. Our discrete method of data acquisition is also subject to potential error due to transducer motion occurring during the 1-second acquisition of 16 video frames. During the acquisition, two ventricular systoles and the diastolic interval between them are captured when heart rates are normal. To ensure that transducer motion does not introduce significant error, we have demonstrated no significant difference in computed ventricular volumes using the first end-diastolic image of each data set compared with the second end-diastolic image of the data set.42 From these results, we infer that although transducer motion during image acquisition by the discrete method is theoretically possible, in practice it does not introduce significant error into the results of volume computation. The polyhedral surface reconstruction algorithm slightly underestimates ventricular volume because it approximates the ventricular surface by a series of short chords that are connected to form triangles. Because the ventricular surface is usually convex, the volume lying between the true surface and the surface approximated by the triangles is omitted, resulting in a slight underestimation of volume. However, when 7 to 10 cross sections are used to represent the adult ventricle, according to the work of Weiss et al,58 significant underestimation is not introduced.
The need to avoid images that intersect within the LV cavity dictated by the current version of the polyhedral surface reconstruction algorithm is viewed by some as a limitation. However, the availability of the interactive line-of-intersection display to the sonographer enables accurate positioning of nonintersecting short-axis images. This feature of the 3D echocardiographic system is viewed by the authors as an advantage and not a limitation, since it allows significant preprocessing of data before data acquisition. This leads to rapid, accurate data acquisition and eliminates repetitive data review and unnecessary boundary tracing.
Boundary tracing error remains a significant potential source of error with all imaging methods and by far outweighs other errors mentioned above. This can be minimized by careful adherence to well-defined rules for tracing. In our experience, tracing endocardial boundaries on the "white" side of the black-and-white boundary has produced the best results. Another source of error is difficult visualization of the tip of the ventricular apex due to an overlying rib. However, use of the second parasternal long-axis reference image over the apex and the line-of-intersection display greatly helps its accurate localization. Variable inclusion of the papillary muscle as part of the cavity introduces another source of error to ventricular volume computation.
Equilibrium radionuclide angiography is a method that is free of geometric assumptions and has been compared with contrast ventriculography.5 7 8 Despite its limitations and in the absence of true measures of LVEF, it is widely used by clinicians and thus serves as an acceptable method for comparison. Factors that contribute to the intrinsic variability and errors of radionuclide angiography are well recognized.5 8 50 51 52 In particular, Wackers et al8 report that the mean variability of absolute ejection fraction for repeat studies in patients with normal LVEFs is significantly greater than that in patients with abnormal LVEFs. Technical factors such as defining regions of interest and background radioactivity zones have been shown to affect normal more than abnormal LVEFs.52 Factors leading to overestimation and underestimation of LVEF by equilibrium radionuclide angiography are discussed in detail by Zaret and Berger.50 In particular, overestimation of LVEF may occur because of an inappropriately high background correction, low end-systolic counts relative to background, selection of a large end-diastolic region of interest, and mispositioning of the detector primarily over relatively normal myocardium. Underestimation of LVEF may occur as a result of inclusion of the left atrium and/or ascending aorta in the LV region of interest during ventricular systole and poor separation of the left and right ventricles, particularly in the lower septum. In patients with poor LVEFs due to anterior infarction and aneurysm formation, underestimation of LVEF may occur because of differences in count detection that are physically dependent on the distance from the collimator. The abnormal anterior segment is closest to the detector and thus attenuates the contribution of normally contracting basal segments that are farther away from the detector. These factors may have contributed in part to our finding that the regression equation describing the relation between LVEFs obtained by 3D echocardiography and equilibrium radionuclide angiography for the full range of LVEF was significantly different from the line of identity.
Conclusions
We conclude that determination of LVEF by 3D
echocardiography yields results that are comparable
to those obtained by equilibrium radionuclide angiography. We also
conclude that 3D echocardiographic determination of
LVEF is superior to all 2D echocardiographic
methods.
|
Acknowledgments |
|---|
This study was supported in part by Investigatorship and Grant-in-Aid Awards to Dr Gopal from the American Heart Association, New York Affiliate.
|
Footnotes |
|---|
Dr King has a financial interest in the three-dimensional echocardiographic system described in this article.
Received November 2, 1994; revision received January 17, 1995; accepted February 8, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. McGhie AI, Willerson JT, Corbett JR. Radionuclide assessment of ventricular function and risk stratification after myocardial infarc-tion. Circulation. 1991;84(suppl I):I-167-I-176.
2. Cohn PF, Gorlin R, Cohn L, Colins JJ. Left ventricular ejection fraction as a prognostic guide in surgical management of coronary and valvular heart disease. Am J Cardiol. 1974;34:136-141. [Medline] [Order article via Infotrieve]
3. Cintron G, Johnson G, Francis G, Cobb F, Cohn JN, for the V-HeFT VA Cooperative Studies Group. Prognostic significance of serial changes in left ventricular ejection fraction in patients with congestive heart failure. Circulation. 1993;87(suppl VI):VI-17-VI-23.
4. Strauss HW, Zaret BL, Hurley PJ, Natarajan TK, Pitt B. A scintiphotographic method for measuring left ventricular ejection fraction in man without cardiac catheterization. Am J Cardiol. 1971;28:575-580. [Medline] [Order article via Infotrieve]
5.
Burow RD, Strauss HW, Singleton R, Pond M, Rehn T,
Bailey IK, Griffity LC, Nickoloff E, Pitt B. Analysis of
left ventricular function from multiple gated acquisition
cardiac blood pool imaging: comparison to contrast angiography.
Circulation. 1977;56:1024-1028.
6.
Green MV, Brody WR, Douglas MA, Borer JS, Ostrow HG,
Line BR, Bacharach SL, Johnston GS. Ejection fraction by count
rate from gated images. J Nucl Med. 1978;19:880-883.
7.
Folland ED, Hamilton GW, Larson SM, Kennedy JW,
Williams DL, Ritchie JL. The radionuclide ejection fraction: a
comparison of three radionuclide techniques with contrast
angiography. J Nucl Med. 1977;18:1159-1166.
8. Wackers FJT, Berger HJ, Johnstone DE, Goldman L, Reduto LA, Langou RA, Gottschalk A, Zaret BL. Multiple gated cardiac blood pool imaging for left ventricular ejection fraction: validation of the technique and assessment of variability. Am J Cardiol. 1979;43:1159-1166. [Medline] [Order article via Infotrieve]
9. Teichholz LE, Krevlen T, Herman MV, Gorlin R. Problems in echocardiographic volume determinations: echocardiographic correlations in the presence or absence of asynergy. Am J Cardiol. 1976;37:7-11. [Medline] [Order article via Infotrieve]
10. Stamm RB, Carabello BR, Mayers DL, Martin RP. Two-dimensional echocardiographic measurement of left ventricular ejection fraction: prospective analysis of what constitutes an adequate determination. Am Heart J. 1982;104:136-144.[Medline] [Order article via Infotrieve]
11.
Carr KW, Engler RL, Forsythe JR, Johnson AD, Gosink B.
Measurement of left ventricular ejection fraction by
mechanical cross-sectional
echocardiography.
Circulation. 1979;59:1196-1206.
12.
Folland ED, Parisi AF, Moynihan PF, Jones DR, Feldman
CL, Tow DE. Assessment of left ventricular ejection
fraction and volumes by real-time, two-dimensional
echocardiography: a comparison of cineangiographic
and radionuclide techniques. Circulation. 1979;60:760-766.
13.
Gueret P, Meerbaum S, Wyatt HL, Uchiyama T, Lang TW,
Corday E. Two-dimensional echocardiographic
quantitation of left ventricular volumes and ejection
fraction. Circulation. 1980;62:1308-1318.
14. Feigenbaum H. Echocardiography. 4th ed. Philadelphia, Pa: Lea & Febiger; 1986:153-155.
15.
Eaton LW, Maughan WL, Shoukas AD, Weiss JL.
Accurate volume determination in the isolated ejecting canine
left ventricle by two-dimensional
echocardiography.
Circulation. 1979;60:320-326.
16.
Schiller NB, Acquatella H, Ports TA, Drew D, Goerke J,
Ringertz H, Silverman NH, Brundage B, Botvinick EH, Boswell R, Carlsson
E, Parmley WW. Left ventricular volume from paired
biplane two-dimensional echocardiography.
Circulation. 1979;60:547-555.
17.
Starling MR, Crawford MH, Sorensen SG, Levi B, Richards
KL, O'Rourke RA. Comparative accuracy of apical biplane
cross-sectional echocardiography and gated
radionuclide angiography for estimating left ventricular
size and performance. Circulation. 1981;63:1075-1084.
18.
Quinones MA, Waggoner AD, Reduto LA, Nelson JG, Young
JB, Winters WL Jr, Ribeiro LG, Miller RR. A new, simplified and
accurate method for determining ejection fraction with two-dimensional
echocardiography.
Circulation. 1981;64:744-753.
19.
Erbel R, Schweizer P, Lambertz H, Henn G, Meyer
J, Krebs W, Effert S. Echoventriculography: a
simultaneous analysis of two-dimensional
echocardiography and
cineventriculography.
Circulation. 1983;67:205-215.
20. Albin G, Rahko P. Comparison of echocardiographic quantitation of left ventricular ejection fraction to radionuclide angiography in patients with regional wall motion abnormalities. Am J Cardiol. 1990;65:1031-1032. [Medline] [Order article via Infotrieve]
21. Ren J, Kotler MN, DePace NL, Mintz GS, Iskandrian AS, Hakki A, Panidis IP, Kimbiris D, Segal BL. Comparison of left ventricular ejection fraction and volumes by two-dimensional echocardiography, radionuclide angiography, and cineangiography. J Cardiovasc Ultrasonogr. 1983;3:213-222.
22. King DL, Harrison MR, King DL Jr, Gopal AS, Kwan OL, DeMaria AN. Ultrasound beam orientation during standard two-dimensional imaging: assessment by three-dimensional echocardiography. J Am Soc Echocardiogr. 1992;6:569-576.
23. King DL, Harrison MR, King DL Jr, Gopal AS, Martin RP, DeMaria AN. Improved reproducibility of left atrial and left ventricular measurements by guided three-dimensional echocardiography. J Am Coll Cardiol. 1992;20:1238-1245. [Abstract]
24. Amico AF, Lichtenberg GS, Reisner SA, Stone CK, Schwartz RG, Meltzer RS. Superiority of visual versus computerized echocardiographic estimation of radionuclide left ventricular ejection fraction. Am Heart J. 1989;118:1259-1265. [Medline] [Order article via Infotrieve]
25. Rich S, Sheikh A, Gallastegni S, Kondos GT, Mason T, Lam W. Determination of left ventricular ejection fraction by visual estimation during real time two-dimensional echocardiography. Am Heart J. 1982;104:603-606. [Medline] [Order article via Infotrieve]
26. Rifkin RD, Koito H. Comparison with radionuclide angiography of two new geometric and four nongeometric models for echocardiographic estimation of left ventricular ejection fraction using segmental wall motion scoring. Am J Cardiol. 1990;65:1485-1490. [Medline] [Order article via Infotrieve]
27. Handschumacher MD, Lethor JP, Siu SC, Mele D, Rivera JM, Picard MH, Weyman AE, Levine RA. A new integrated system for three-dimensional echocardiographic reconstruction: development and validation for ventricular volume with application in human subjects. J Am Coll Cardiol. 1993;21:743-753. [Abstract]
28. Geiser EA, Ariet M, Conetta DA, Lupkiewicz SM, Christie LG, Conti CR. Dynamic three-dimensional echocardiographic reconstruction of the intact human left ventricle: technique and initial observation in patients. Am Heart J. 1982;103:1056-1065. [Medline] [Order article via Infotrieve]
29.
McPherson DD, Skorton DJ, Kodiyalam S, Petree L, Noel
MP, Kieso R, Kerber RE, Collins SM, Chandran KB. Finite element
analysis of myocardial diastolic function using
three-dimensional echocardiographic reconstructions:
application of a new method for study of acute ischemia in
dogs. Circ Res. 1987;60:674-682.
30. Raichlen JS, Trivedi SS, Herman GT, St John Sutton MG, Reichek N. Dynamic three-dimensional reconstruction of the left ventricle from two-dimensional echocardiograms. J Am Coll Cardiol. 1986;8:364-370. [Abstract]
31. Ghosh A, Nanda NC, Maurer G. Three-dimensional reconstruction of echocardiographic images using the rotation method. Ultrasound Med Biol. 1982;8:655-661. [Medline] [Order article via Infotrieve]
32. Linker DT, Moritz WE, Pearlman AS. A new three-dimensional echocardiographic method of right ventricular volume measurement: in vitro validation. J Am Coll Cardiol. 1983;8:101-106.
33. Martin RW, Bashein G. Measurement of stroke volume with three-dimensional transesophageal ultrasonic scanning. Anesthesiology. 1989;70:470-476. [Medline] [Order article via Infotrieve]
34. King DL, Al-Banna SJ, Larach DL. A new, three-dimensional random scanner for ultrasonic/computergraphic imaging of the heart. In: White DN, Barnes R, eds. Ultrasound in Medicine. New York, NY: Plenum Press; 1976;2:363-366.
35. King DL, King DL Jr, Shao MYC. Three-dimensional spatial registration and interactive display of position and orientation of real-time ultrasound images. J Ultrasound Med. 1990;9:525-532. [Abstract]
36. King DL, King DL Jr, Shao MYC. Evaluation of in vitro measurement accuracy of a three-dimensional ultrasound scanner. J Ultrasound Med. 1991;10:77-82. [Abstract]
37. Cook LT, Cook PN, Lee KR, Batnitsky S, Wong BYS, Fritz SL, Ophir J, Dwyer SJ III, Bigongiari LR, Templeton AW. An algorithm for volume estimation based on polyhedral approximation. IEEE Trans Biomed Eng. 1980;27:493-500. [Medline] [Order article via Infotrieve]
38. Gopal AS, King DL, Katz J, Boxt LM, King DL Jr, Shao MYC. Three-dimensional echocardiographic volume computation by polyhedral surface reconstruction: in vitro validation and comparison to magnetic resonance imaging. J Am Soc Echocardiogr. 1992;5:115-124. [Medline] [Order article via Infotrieve]
39. Schroder KM, Sapin PM, King DL, Smith MD, DeMaria AN. Three-dimensional echocardiographic volume computation: in vitro comparison to standard two-dimensional echocardiography. J Am Soc Echocardiogr. 1993;6:467-475. [Medline] [Order article via Infotrieve]
40. Sapin PM, Schroeder KM, Smith MD, De Maria AN, King DL. Three-dimensional echocardiographic measurement of left ventricular volume in vitro: comparison to two-dimensional echocardiography and cineventriculography. J Am Coll Cardiol. 1993;22:1530-1537. [Abstract]
41. Siu SC, Rivera M, Guerrero L, Handschumacher MD, Lethor JP, Weyman AE, Levine RA, Picard MH. Three-dimensional echocardiography: in vivo validation for left ventricular volume and function. Circulation. 1993;88(pt 1):1715-1723.
42. Gopal AS, Keller AM, Rigling R, King DL Jr, King DL. Left ventricular volume and endocardial surface area by three-dimensional echocardiography: comparison to two-dimensional echocardiography and magnetic resonance imaging in normal subjects. J Am Coll Cardiol. 1993;22:258-270. [Abstract]
43. American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-dimensional Echocardiograms: Schiller NB, Shah PM, Crawford M, DeMaria AN, Devereaux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I, Silverman NH, Tajik J. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. J Am Soc Echocardiogr. 1989;2:356-367.
44.
Calahan RJ, Frowlich JW, McKusick KA, Leppo J, Strauss
WH. A modified method for the in vivo labelling of red blood
cells with Tc-99m: concise communication. J
Nucl Med. 1982;23:315-318.
45. Neter J, Wasserman W. Applied Linear Statistical Models. Homewood, Ill: Richard D Irwin, Inc; 1974:140-143.
46. Altman DG, Bland JM. Measurement in medicine: the analysis of method comparison studies. Statistician. 1983;32:307-317.
47. Zar JH. Biostatistical Analysis. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall; 1984:386-387.
48. Zar JH. Biostatistical Analysis. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall; 1984:181-183.
49. Wyatt HL, Haendchen RV, Meerbaum S, Corday E. Assessment of quantitative methods for two-dimensional echocardiography. Am J Cardiol. 1983;52:396-401. [Medline] [Order article via Infotrieve]
50. Zaret BL, Berger HJ. First-pass and equilibrium radionuclide angiocardiography for evaluating ventricular performance. In: Simoons ML, Reiber JHC, eds. Nuclear Imaging in Clinical Cardiology. Boston, Mass: Martinus Nijhoff Publishers; 1984:140-141.
51. Kaul S, Boucher CA, Okada RD, Newell JB, Strauss HW, Pohost GM. Sources of variability in the radionuclide angiographic assessment of ejection fraction: a comparison of first-pass and gated equilibrium techniques. Am J Cardiol. 1984;53:823-828. [Medline] [Order article via Infotrieve]
52. Williams DL, Hamilton GW. The effect of errors in determining left ventricular ejection fraction from radionuclide counting data. In: Sorenson JA, ed. Nuclear Cardiology: Selected Computer Aspects. New York, NY: Society of Nuclear Medicine; 1978:107-117.
53. Sapin PM, Schroeder KM, Gopal AS, Smith MD, DeMaria AN, King DL. Comparison of two and three-dimensional echocardiography with cineventriculography for measurement of left ventricular volume in patients. J Am Coll Cardiol. 1994;24:1054-1063. [Abstract]
54. Gopal AS, Keller AM, Shen Z, Sapin PM, Schroeder KM, King DL Jr, King DL. Three-dimensional echocardiography: in vitro and in vivo validation of left ventricular mass and comparison to conventional echocardiographic methods. J Am Coll Cardiol. 1994;24:504-513. [Abstract]
55. King DL, Gopal AS, King DL Jr, Shao MY. Three-dimensional echocardiography: in vitro validation for quantitative measurement of total and `infarct' surface area. J Am Soc Echocardiogr. 1993;6:69-76. [Medline] [Order article via Infotrieve]
56. King DL, Gopal AS, Schröder KM, Sapin PM, King DL Jr, Shen Z. Ratio of infarct subtended volume to surface area by 3D echocardiography: in vivo measurement of infarct expansion and aneurysm formation. In: Proceedings Computers in Cardiology 1993. Los Alamitos, Calif: IEEE Computer Society Press; 1994:17-20.
57. Moritz WE, Shreve PL, Mace LE. Analysis of an ultrasonic spatial locating system. IEEE Trans Instrum Meas. 1976;25:43-50.
58.
Weiss JL, Eaton LW, Kallman CH, Maughan WL.
Accuracy of volume determination by two-dimensional
echocardiography: defining requirements under
controlled conditions in the ejecting canine left ventricle.
Circulation. 1982;67:889-895.
This article has been cited by other articles:
 |
|
 |
|
  J. L. Gutierrez-Chico, J. L. Zamorano, E. Prieto-Moriche, R. A. Hernandez-Antolin, M. Bravo-Amaro, L. Perez de Isla, M. Sanmartin-Fernandez, J. A. Baz-Alonso, and A. Iniguez-Romo Real-time three-dimensional echocardiography in aortic stenosis: a novel, simple, and reliable method to improve accuracy in area calculation Eur. Heart J., May 2, 2008; 29(10): 1296 - 1306. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. De Ferrari, A. Sanzo, A. Bertoletti, G. Specchia, E. Vanoli, and P. J. Schwartz Baroreflex Sensitivity Predicts Long-Term Cardiovascular Mortality After Myocardial Infarction Even in Patients With Preserved Left Ventricular Function J. Am. Coll. Cardiol., December 11, 2007; 50(24): 2285 - 2290. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Dewey, M. Muller, S. Eddicks, D. Schnapauff, F. Teige, W. Rutsch, A. C. Borges, and B. Hamm Evaluation of Global and Regional Left Ventricular Function With 16-Slice Computed Tomography, Biplane Cineventriculography, and Two-Dimensional Transthoracic Echocardiography: Comparison With Magnetic Resonance Imaging J. Am. Coll. Cardiol., November 21, 2006; 48(10): 2034 - 2044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P. Nikitin, C. Constantin, P. H. Loh, J. Ghosh, E. I. Lukaschuk, A. Bennett, S. Hurren, F. Alamgir, A. L. Clark, and J. G.F. Cleland New generation 3-dimensional echocardiography for left ventricular volumetric and functional measurements: Comparison with cardiac magnetic resonance Eur J Echocardiogr, October 1, 2006; 7(5): 365 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. D. Jacobs, I. S. Salgo, S. Goonewardena, L. Weinert, P. Coon, D. Bardo, O. Gerard, P. Allain, J. L. Zamorano, L. P. de Isla, et al. Rapid online quantification of left ventricular volume from real-time three-dimensional echocardiographic data Eur. Heart J., February 2, 2006; 27(4): 460 - 468. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Jenkins, K. Bricknell, L. Hanekom, and T. H. Marwick Reproducibility and accuracy of echocardiographic measurements of left ventricular parameters using real-time three-dimensional echocardiography J. Am. Coll. Cardiol., August 18, 2004; 44(4): 878 - 886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. P. Kuhl, M. Schreckenberg, D. Rulands, M. Katoh, W. Schafer, G. Schummers, A. Bucker, P. Hanrath, and A. Franke High-resolution transthoracic real-time three-dimensional echocardiography: Quantitation of cardiac volumes and function using semi-automatic border detection and comparison with cardiac magnetic resonance imaging J. Am. Coll. Cardiol., June 2, 2004; 43(11): 2083 - 2090. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. King, L. El-Khoury Coffin, and M. S. Maurer Myocardial contraction fraction: a volumetric index of myocardial shortening by freehand three-dimensional echocardiography J. Am. Coll. Cardiol., July 17, 2002; 40(2): 325 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ahmad, T. Xie, M. McCulloch, G. Abreo, and M. Runge Real-time three-dimensional dobutamine stress echocardiography in assessment of ischemia: comparison with two-dimensional dobutamine stress echocardiography J. Am. Coll. Cardiol., April 1, 2001; 37(5): 1303 - 1309. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Chuang, M. G. Hibberd, C. J. Salton, R. A. Beaudin, M. F. Riley, R. A. Parker, P. S. Douglas, and W. J. Manning Importance of imaging method over imaging modality in noninvasive determination of left ventricular volumes and ejection fraction: Assessment by two- and three-dimensional echocardiography and magnetic resonance imaging J. Am. Coll. Cardiol., February 1, 2000; 35(2): 477 - 484. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. De Simone, G. Glombitza, C. F. Vahl, J.o. Albers, H. P. Meinzer, and S. Hagl Three-dimensional color doppler: a clinical study in patients with mitral regurgitation J. Am. Coll. Cardiol., May 1, 1999; 33(6): 1646 - 1654. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. De Simone, G. Glombitza, C.F. Vahl, H.P. Meinzer, and S. Hagl Three-dimensional Doppler. Techniques and clinical applications Eur. Heart J., April 2, 1999; 20(8): 619 - 627. [Abstract] [PDF]
|
|
|
|
 |
|
 R. De Simone, G. Glombitza, C.-F. Vahl, J. Albers, H.-P. Meinzer, and S. Hagl Assessment of mitral regurgitant jets by three-dimensional color Doppler Ann. Thorac. Surg., February 1, 1999; 67(2): 494 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Buck, P. Hunold, K. U. Wentz, W. Tkalec, H. J. Nesser, and R. Erbel Tomographic Three-dimensional Echocardiographic Determination of Chamber Size and Systolic Function in Patients With Left Ventricular Aneurysm : Comparison to Magnetic Resonance Imaging, Cineventriculography, and Two-dimensional Echocardiography Circulation, December 16, 1997; 96(12): 4286 - 4297. [Abstract] [Full Text]
|
|
|
|
|
|
L. Gepstein, G. Hayam, S. Shpun, and S. A. Ben-Haim Hemodynamic Evaluation of the Heart With a Nonfluoroscopic Electromechanical Mapping Technique Circulation, November 18, 1997; 96(10): 3672 - 3680. [Abstract] [Full Text]
|
|
|
|
|
|
Y. F.M. Nosir, P. M. Fioretti, W. B. Vletter, E. Boersma, A. Salustri, J. T. Postma, A. E.M. Reijs, F. J. Ten Cate, and J. R.T.C. Roelandt Accurate Measurement of Left Ventricular Ejection Fraction by Three-dimensional Echocardiography: A Comparison With Radionuclide Angiography Circulation, August 1, 1996; 94(3): 460 - 466. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||