Vol. 277, Issue 5, R1481-R1487, November 1999
Regional wall motion and strain of transplanted hearts in
pediatric patients using magnetic resonance tagging
Mary T.
Donofrio1,2,
Bernard J.
Clark1,
Claudio
Ramaciotti1,
Marshall
L.
Jacobs3,
Kenneth E.
Fellows4,
Paul M.
Weinberg1, and
Mark A.
Fogel1
1 Division of Cardiology,
Department of Pediatrics,
3 Department of Surgery, and
4 Department of Radiology, The
University of Pennsylvania Hospital, Philadelphia, Pennsylvania 19104;
and 2 Division of Cardiology, Department of Pediatrics, The
Medical College of Virginia of the Virginia Commonwealth University,
Richmond, Virginia 19298
 |
ABSTRACT |
Abnormal ventricular systolic torsion is
present during histological rejection in adult cardiac transplant
patients. Because biomechanical properties of transplanted hearts in
the baseline state have not been studied in children, pediatric
patients were evaluated to quantify ventricular wall motion and strain.
Eight transplant studies and eight normal controls were evaluated.
Magnetic resonance tagging was performed to determine radial
shortening, twist, and strain in four ventricular anatomic areas at two
short-axis levels. Controls had counterclockwise twist. Six transplant
studies had clockwise twist, six had akinetic regions, and all had
regions of no twist. One demonstrated paradoxical motion of the septum. A comparison between transplant patients and controls revealed strain
to be similar in all regions except one (superior wall at the
atrioventricular valve level) and strain distribution to be different
only in two of eight regions. Pediatric transplant patients demonstrate
regional wall motion abnormalities in the absence of rejection.
Compared with normal controls, the transplanted left ventricle
maintains normal strain in the presence of abnormal twist. This may be
a compensatory mechanism and have clinical implications.
magnetic resonance imaging; spatial modulation of magnetization; pediatric orthotopic cardiac transplant
| |
INTRODUCTION |
WHEN VIEWED FROM APEX TO base, the short axis of the
normal left ventricle contracts in a counterclockwise fashion during systole (5, 10, 16). Hansen et al. (10) found, after implanting
intramyocardial markers in transplanted hearts, that torsional
deformation was sensitive to the contractile state of the myocardium,
and a decrease in the amplitude and rate of this torsional deformation
was seen in patients suffering from acute rejection confirmed by biopsy
(11). Because magnetic resonance imaging (MRI), using spatial
modulation of magnetization (SPAMM) (1, 2, 15, 16), has the ability to
tag the myocardium noninvasively, alterations in regional wall motion,
represented as twist and radial shortening, may be detected similarly
to the alterations in torsion detected by the implanted intramyocardial markers. Strain, another biomechanical parameter that can be determined by MRI, is a unitless measurement of deformation that partially characterizes the contractile state of the myocardium. With
knowledge of the noninvasive nature of MRI, it may be beneficial to
follow transplant patients with this technique to detect ventricular dysfunction and cellular rejection and potentially decrease the need
for endomyocardial biopsy.
With the work of Hansen and colleagues in mind, an understanding of the
biomechanics of the transplanted heart in the absence of rejection is
key to determining the changes that occur during rejection. The biomechanical properties of the
transplanted ventricle may not be the same as a normal ventricle that
is innervated, has not undergone circulatory arrest, is not subjected
to the toxic effects of immunosuppressant drugs, and is not at risk for rejection. We hypothesize that wall motion and strain in this patient
population are different from the normal left ventricle for just these
reasons. Because studies in the pediatric transplant population are
lacking such biomechanical data, we undertook this study to quantify
baseline properties of regional wall motion and strain in these patients.
It is important to note that this "pilot" study, because of its
relatively small number of patients and comparison with young adult
left ventricles, is only the first step in a long line of investigation. It must be viewed as "hypothesis generating."
| |
METHODS |
Patient Population
Twelve patients under 12 years of age who had undergone an orthotopic
heart transplant at The Children's Hospital of Philadelphia were
studied with MRI between August, 1993 and September, 1994. One patient
had two studies, ~1 yr apart. Five patients were excluded due to the
following: inadequate electrocardiographic gating
(n = 2), inadequate sedation
(n = 1), and/or malfunction of the
imaging system (n = 2). MRI
with SPAMM has been approved by the review board at our institution,
and informed consent was obtained from the families of all patients.
Imaging was done within 1.9 ± 1.9 days (range 1-6
days) prior to endomyocardial biopsy in six of the eight studies used
for analysis. One patient did not have a biopsy at the time of
catheterization, and the other underwent catheterization and biopsy 3.7 mo prior to imaging. The patient age was 5.1 ± 3.7 yr (range 18 days-11.2 yr), which was 2.6 ± 1.2 yr (range 6 mo-4 yr)
posttransplant. All patients were clinically well.
Eight volunteers (ages 25 ± 3.6 yr) with normal left ventricular
function were imaged as controls. All were clinically well with no
signs of cardiac dysfunction.
Sedation was achieved with chloral hydrate for children under 3 yr, and
oral pentobarbital sodium and meperidine were used for children between
3 and 7 yr. For cases of inadequate sedation, intravenous
pentobarbital sodium was given.
MRI
Patients were imaged using a Siemans 1.5-Tesla Magnetom
magnetic resonance scanner. MRI used electrocardiographic gated, spin echo, and cine techniques. Older patients were placed in a body coil
and younger patients in a head coil. Initial scans to define the
cardiovascular anatomy consisted of two sets of
T1 axial slices, 4-6 mm
thick, depending on the size of the patient, with the interslice spacing equal to the slice thickness. The acquisitions were interleaved to yield contiguous slices. Three localizers were performed to achieve
the final standardized short-axis views of the left ventricle. The
first localizer imaged through the atrioventricular valve plane was
done to visualize both valve orifices. The second localizer was
positioned perpendicular to a line drawn between the two
atrioventricular valve centers seen on the first localizer and
intersecting the ventricular apex. This yielded an image of the mitral
valve and the left ventricular apex in the same plane. Finally, the
third localizer was positioned to intersect the mitral valve and apex of the left ventricle yielding a four-chamber view of the heart. Two
slices, one one-third of the distance from atrioventricular valve (AVV)
to apex and the second two-thirds the distance from AVVs to apex
(denoted Apex) were chosen to undergo imaging to create two left
ventricular short-axis SPAMM sequences.
Spatial Modulation of Magnetization
SPAMM imaging is performed by a sequence of timed, nonselective
radio-frequency pulses of which the sum is 130° separated by a
series of magnetic field gradient radio-frequency pulses that produce
saturated spins in two sets of parallel stripes perpendicular to each
other. This sequence is followed by a standard gradient echo sequence
to achieve a grid on the myocardium dividing the tissue into cubes.
Slices ranging in thickness between 5 and 8 mm are acquired for 12 equally spaced time periods (phases) beginning at end-diastole and
moving throughout systole. Myocardial wall motion distorts the magnetic
grid, and assessment of motion and deformation is made by tracking the
movement of the intersection points formed by the grid lines. For each
SPAMM sequence, effective repetition time = R-R interval,
inversion time = 16 ms, flip angle = 30°, matrix size = 128 × 256 interpolated to 256 × 256, slice thickness range = 4-7
mm, tag thickness range = 1.5-2 mm, field of view range = 160-280, and number of excitation = 3. There were 30 ms between images.
Analysis
Images were downloaded from the Magnetom to a Sun SPARC station 10 (Sun
Microsystems, Mountain View, CA) for computer analysis. Volumetric
Image Display and Analysis (13), a user-interactive software package,
was used to determine strain and wall motion. The intersection points
formed by the perpendicular SPAMM lines were marked and tracked in
series by moving the labeled points from phase to phase on successive
images. A mathematical technique called DeLaunay triangulation (6, 14)
was then used to connect the points to create nonoverlapping, equally
sized triangles. These triangles were also labeled and followed from
image to image.
Wall Motion (Twist and Radial Shortening)
Twist and shortening were measured relative to the
centroid of the ventricular cavity at end-diastole that was determined by tracing the endocardial border and calculating the center based on
those borders. The center of each triangle at each phase was then
obtained by computerized counting of pixels. With the use of a
Cartesian coordinate system and the following mathematical formulas,
the distance (radial shortening) and the angle (twist) made by a ray
from the centroid of the cavity to the center of the triangle was
obtained
and
where
Pn + 1
Pn
is the distance between the centroid of the triangle and the centroid
of the ventricle at phase n and
n + 1 (shortening at each phase) and
is the angle the rays make (twist at each phase).
Xn and
Yn and
Xn + 1 and
Yn + 1 are the
coordinates of the center of the triangle at phase
n and n + 1, and
Xc and
Yc are
coordinates of the centroid of the cavity.
Pn
· Pn + 1 is the
vector dot product of the rays at phase
n and n + 1. Radial shortening was defined as the distance moved toward the center of the
ventricular cavity indexed to the diameter of the ventricular cavity
(mm/end-diastolic radius in mm). Twist was defined as the amount of
rotation about the center of the ventricular cavity and was measured in
degrees (a positive value indicates counterclockwise twist and a
negative value clockwise twist when viewed from apex to base). Data
were also displayed graphically, in which end-diastole was denoted by a
point and motion in systole represented by a line following the point
(Fig. 1,
A-C).
View larger version (71K):
[in this window]
[in a new window]
|
Fig. 1.
A: short-axis image using spatial
modulation of magnetization, which divides myocardium into a
magnetically tagged grid. Intersection of grid lines are labeled with
white points and are tracked from phase to phase.
B: points are connected into
triangles. C: centroid of each
triangle is tracked through systole. A point represents the starting
point at end-diastole, and a line represents systolic movement.
D: distortion of triangles (strain) is
represented by color coding.
|
|
Strain
Strain is the unitless measurement of deformation that relates the
changes in the myocardial wall shape in systole relative to
end-diastole. A positive strain value signifies elongation, and a
negative strain represents compression. Strain calculations use the
two-dimensional strain tensor and the method of eigensystems solutions
to determine the deformation of each triangle. Principal strain
(E1; the
parameter used in this study) always represents the largest compressive
(negative) strain. Strain data were obtained by averaging the
strain of all the triangles at each phase and noting the maximal
compressive strain observed with its standard deviation. Strain data
were also displayed in color coded or gray scale form, either alone, or
superimposed on the anatomic magnetic resonance image of the ventricle
(Fig. 1, A,
B, and
D).
Data Interpretation
The myocardium was divided into four areas (ventricular septum,
inferior wall, posterior wall, and superior wall) at each of the two
short-axis levels (AVV and Apex levels) to obtain regional parameters.
Twist was studied by evaluating initial twist (the initial motion in 1 direction) and net twist (the sum of the total motion throughout the
phases in systole). Radial shortening, as well, was analyzed as initial
and net values. Values of
E1 strain were
analyzed in each region at both short-axis levels. Strain heterogeneity, or the distribution of strain within a region, was
evaluated by calculating the standard deviation of strain divided by
the average strain for each region. This ratio was compared among
regions between transplant patients and normal controls.
Statistics
Only data from the first MRI scan in the patient who underwent two
studies were used in the quantitative statistical analysis. Comparisons
between groups were made using Student's
t-test. Comparisons between multiple
groups were done with analysis of variance. All measurements are
recorded as the means ± SD, and differences between values
represent differences in the geometric means. Significance is defined
as P < 0.05.
| |
RESULTS |
Twist Analysis
Qualitative analysis. Normal control
subjects had counterclockwise twist in all regions (Fig.
2,
left). Although there was some
diversity in the transplant patients patterns of wall motion, there
were a few unifying themes. Six of eight studies had clockwise twist of
either the superior, inferior, and/or posterior wall (the
two studies that did not have a wall with clockwise
twist were from patient 1, who
underwent 2 MRI scans 1 yr apart). All studies had regions of no twist
(i.e., direct linear movement toward the left ventricular
centroid). Six studies had akinetic regions (posterior wall in 1, ventricular septum in 2, inferior wall in 2, and ventricular septum and
inferior wall in 1), and one had paradoxical wall
motion of the ventricular septum (Table 1 and Fig. 3,
left). The patient who was imaged
twice had minor twist abnormalities in his first study (regions of no
twist at the posterior wall) and similar findings 1 yr later (Table 1, patients 1A and
1B).
View larger version (63K):
[in this window]
[in a new window]
|
Fig. 2.
Left: normal left
ventricular (LV) twist. Note that point represents
starting point at end-diastole and lines represent
systolic movement. Twist is in counterclockwise direction.
Right: normal LV strain. Note strain
color code in which greatest negative strain (compression) is black and
greatest positive strain (elongation) is white.
|
|
View larger version (67K):
[in this window]
[in a new window]
|
Fig. 3.
Left: twist
abnormalities in a postcardiac transplant patient. Note area of
clockwise twist (CW), counterclockwise twist (CCW), and no twist (NTw).
There is a region that is akinetic (Ak) and another with paradoxical
wall motion (Pd). Right: strain in
same patient. Arrow points to a region of increased compressive
strain.
|
|
Quantitative analysis.
COMPARISON WITHIN PATIENT SUBTYPES.
Transplant patients revealed no significant difference in initial or
net twist among the four regions at the AVV and no differences in
initial twist at the Apex. There was a difference in net twist (P < 0.02) among the four regions at
the Apex (net twist was smaller at the inferior wall). Normal controls
had no significant difference in initial or net twist among the four
regions at both short-axis levels. Radial shortening in transplant
patients demonstrated that initial and net shortening were both
different among the four regions at the AVV
(P < 0.01 for initial, and
P < 0.05 for net shortening) and
Apex (P < 0.05 for both initial and
net shortening). For both short-axis levels, the ventricular septum had
less initial and net shortening than the other regions. Normal controls
had differences only in initial shortening at the AVV
(P < 0.05). In these patients, the
superior wall had a higher initial shortening than the other regions.
COMPARISON BETWEEN PATIENT SUBTYPES.
When comparing transplant patients to the normal subjects, initial
twist was different at the ventricular septum of the AVV
(P < 0.05) A trend was noted for
greater initial and net twist in normal subjects than transplant patients in all regions except the superior wall of the Apex (and at
the apical septal wall for net twist only) (Fig.
4, A
and B). This greater net
twist in normal subjects reached statistical significance at the apical
inferior wall (P < 0.02). The
regional pattern of shortening was similar between groups at either
short-axis level with two exceptions:
1) transplant patients had greater initial shortening at the apical posterior wall than normal controls, and 2) transplant patients had
greater net shortening at the apical superior wall than normal controls
(P < 0.05; Fig.
5, A and
B).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 4.
Net twist at atrioventricular valve (AVV;
A) and apex (Apex;
B) levels in transplant patients and
normal controls. A trend was noted for greater net twist in normal
subjects than transplant patients in all regions except
apical superior and septal walls. This greater net twist in normal
subjects reached statistical significance in net twist at apical
inferlior wall (* P < 0.02).
VS, ventricular septum; IW, inferior wall; PW, posterior wall; SW,
superior wall. Error bars, SD.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 5.
Net shortening at AVV (A) and Apex
(B) levels in transplant patients
and normal controls. Regional pattern of shortening was similar between
groups at either short-axis level with the exception that transplant
patients had greater net shortening at apical superior wall than normal
controls (* P < 0.05). Error
bars, SD.
|
|
Strain analysis.
COMPARISON WITHIN PATIENT SUBTYPES. No
statistical difference in
E1 strain at the
AVV and Apex was found within the four myocardial regions of both the
transplant patients and normal subjects. Heterogeneity of strain within
each region was not statistically different among the four regions at
the AVV and Apex levels in the transplanted hearts or in the normal subjects.
COMPARISON BETWEEN PATIENT SUBTYPES.
Comparing the transplant patients with normal controls,
E1 strain was not
different in any region of the AVV or Apex (Figs. 3,
right, and
6, A and
B) with the exception of larger
E1 strains in the
control group at the AVV superior wall
(P < 0.05). Although, in
general, heterogeneity of strain was not different between transplant
patients and normal controls at either short-axis level, significantly
larger heterogeneity of strain was noted in transplant patients at the
ventricular septum and superior wall of the AVV
(P < 0.05; Fig.
7, A and
B).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 6.
Principal
(E1) strain in
transplant patients and normal controls at AVV
(A) and Apex
(B) levels.
E1 strain was not
different in any region of AVV or Apex with the exception of larger
E1 strains in
control group than transplant patients at AVV superior wall
(* P < 0.05). Error bars,
SD.
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 7.
Heterogeneity of
E1 strain in
transplant patients and normal controls at AVV
(A) and Apex
(B). On
y-axis is coefficient of variation (SD
of E1 strains
within a given region/average
E1 strain in that
region). Although in general, heterogeneity of strain was not different
between transplant patients and normal controls at either short-axis
level, significantly larger heterogeneity of strain was noted in
transplant patients at VS and SW of AVV
(* P < 0.05). Error bars,
SD.
|
|
Analysis of Abnormal Twist Regions in Transplant Patients
Regions of abnormal and normal twist were selected
qualitatively (graphic data) and confirmed quantitatively (a
significant difference in twist between the 2 regions in degrees; Fig.
8A) for
the expressed purpose of comparing strain and shortening between these
two regions. It was prespecified that if the qualitative analysis did
not agree with the quantitative data, then we would not proceed with
the comparison. All qualitative analyses did agree with the
quantitative ones. Although there were difference in initial (AVV and
Apex, P < 0.05) and net shortening
(AVV P < 0.05, Apex
P < 0.01; Fig.
8B), there was no difference in
strain or strain heterogeneity between these abnormal and normal twist regions (Fig. 9,
A and
B).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 8.
Net twist (A) and shortening
(B) in qualitative abnormal (Abn)
vs. normal (NL) twist regions in transplant patients. Note differences
in both twist and shortening at both AVV and Apex
(* P < 0.02, ** P < 0.001). Error bars,
SD.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 9.
Strain (A) and heterogeneity of
strain (B) in qualitative Abn vs. NL
twist regions in transplant patients. Note that despite selected Abn
twist regions, there was no difference in strain or strain
heterogeneity at AVV or Apex. Error bars, SD.
|
|
Cardiac Catheterization and Biopsy Results (Table
2)
Hemodynamic assessment was generally normal in all patients. Peak
systolic pulmonary pressures were noted to be in the normal range
(11-36 mmHg) with the exception of patient
1B, whose peak systolic pulmonary pressure of 39 mmHg
was present when his systemic peak systolic pressure was 129 mmHg.
Cardiac index was also noted to be normal in all patients with one
exception; patient 3, whose cardiac index was 3.3 l · min
1 · m2
by oxygen saturation data (using the Fick equation), had a cardiac index of 3.7 l · min1 · m2
by the thermodilution technique.
Those patients who had coronary angiography had no evidence of
stenosis. Seven patients underwent right ventricular biopsy. Six
specimens revealed Grade 0, and one revealed Grade 1A histology according to the criteria published by Billingham et al. (Table 2) (4).
Electrocardiogram Findings
All patients were in sinus rhythm, and two had evidence
of a second P wave. Six patients had a narrow QRS complex, and two had
patterns consistent with a right bundle branch block. One of the six
with a narrow QRS complex had a pattern suggesting a right ventricular
conduction delay (normal QRS duration and a right bundle pattern).
There was no correlation noted between the depolarization and twisting
patterns. No patient had a history of significant ectopy.
| |
DISCUSSION |
Hansen and colleagues (11) reported in 1987 that compared with
baseline, left ventricular torsional deformation amplitude and rate
were altered in adult patients with ventricular dysfunction and
histological evidence of rejection. Torsional deformation was
determined by placing intramyocardial radiopaque markers in the
transplanted hearts and viewing them with fluoroscopy. Compared with
their prerejection data, the torsional deformation in the maximally
deforming segment (denoted
max) decreased by 25% during acute rejection and myocardial necrosis. This was associated with a
significant decrease in the torsional rate
(dmax/dt).
Left ventricular end-diastolic volume, stroke volume, ejection
fraction, peak left ventricular filling rate, and diastolic recoil did
not change during the rejection episode. These findings are the driving force behind this initial pilot study. Because the biomechanics of the
transplanted left ventricle have numerous reasons why they should not
be the same as those of the normal left ventricle (see introduction),
this study attempted to define twist, radial shortening, and strain in
the pediatric transplanted left ventricle in the baseline state. The
findings should be viewed as hypothesis generating because of the small
number of patients and comparisons to young adult left ventricles. If
these data are confirmed by larger numbers of patients and pediatric
controls, it will add to our understanding of cardiac transplantation
and form the basis of a potential way to more definitively determine
rejection noninvasively.
Twist and Strain
MRI using SPAMM to assess twist in normal left ventricles was first
described by Young et al. (16). This technique goes further than the
ones previously described (3, 10, 11), because regional events may be
easily examined in addition to global parameters. A recent study showed
that regional area strain correlates with regional oxygen consumption
in certain instances (9). Measurements of myocardial strain have been
determined in the normal adult left ventricle using MRI and SPAMM (16). In studies from our institution (7, 8), SPAMM has been used to quantify
strain in a number of congenital heart lesions. In this study, we found
that even in the baseline state, pediatric transplant patients have
significant abnormalities. This information gives us insight into the
biomechanical properties of the transplanted heart. Once twist, radial
shortening, and strain are determined in pre- versus postrejection
states, correlation with histological rejection can be attempted.
Our results
This study was designed to evaluate the biomechanics of
the transplanted left ventricle in a nonrejection state and then
compare the results with normal subjects. Data revealed that all
patients had evidence of markedly abnormal twist. Despite this finding, E1 strain and
heterogeneity of strain were similar within all regions at the AVV and
Apex levels and not different from the normal controls. These results
demonstrate that the transplanted left ventricle seems to maintain
normal strain in the presence of abnormal twist.
It is unclear what factors may be causing these alterations in twist.
It has been noted in transplanted primate hearts with rejection that in
the same heart, there are regions of both myocardial necrosis and
regions that are relatively spared from disease (12). Focal areas of
fibrosis may play a role in causing abnormal wall motion in the
nonrejected state, but this remains unknown.
The catheterization and biopsy data were normal in the patients without
clinical evidence of rejection, and ventricular depolarization abnormalities seen on electrocardiogram did not correlate with twisting
patterns. We speculate that the regional wall motion abnormalities may
be a compensatory mechanism to maintain normal strain and hemodynamic
parameters in the presence of characteristics unique to the
transplanted heart such as loss of sympathetic control, focal fibrosis
from the period of ischemia prior to transplantation, effects
of immunosuppressant drugs, or sequelae from past episodes of cellular rejection.
From our experience, we have found that the transplanted left
ventricle, in general, have a thicker myocardium and smaller cavity
volume relative to the myocardial mass than normal individuals (compare
Figs. 2 and 3). However, one would expect this concentric, symmetrical myocardial hypertrophy to have a "global" effect on cardiac mechanics, not the regional effect seen in our study. Nevertheless, this is a possibility and should be kept in mind when
considering our results.
Limitations
Comparisons of regional strain and wall motion in this
study were made to adults with normal left ventricular function. We have imaged two pediatric patients with vascular ring anomalies (ages
1.1 and 1.3 yr) to determine whether there are important differences
that occur with age. Their twist data appear qualitatively to be the
same as the adult controls. Another limitation of this study is that
correlation of the findings with endomyocardial biopsy is imprecise in
that the sample is taken from the right ventricular septal surface and
may not reflect the overall histological findings in the left
ventricle. Finally, this study involves a small number of patients. A
multicenter study would be ideal to enable quantification of the
differences in strain and regional wall motion in transplant patients
to determine whether these biomechanical properties change with time
and/or episodes of rejection.
Perspectives
Although it is tempting to place the transplanted heart
in the same category as a normal heart, there are multiple reasons, as
previously stated, why this should not be so. Our data, which suggest
alterations of twist in the presence of normal strain in the
transplanted pediatric left ventricle when compared with the normal
adult left ventricle, add another layer of complexity to the whole
cardiac transplantation philosophy. If our data can be replicated in
comparison with the normal pediatric left ventricle, they would imply
that even in a transplanted heart that is not undergoing rejection,
twist is uncoupled from strain. This would add to our understanding of
the biomechanics of these ventricles and has implications for using
twist as a marker during rejection (obviously, the altered baseline
twist must be considered).
It must be remembered that our data are only "snapshots in time"
in the life of a transplanted left ventricle. Although strain measurements did not appear altered, it may be useful to investigate these parameters throughout the life of the transplanted left ventricle
and certainly during rejection episodes. The noninvasive nature of MRI
makes it ideal to evaluate these biomechanical properties in the hopes
of finding a noninvasive surrogate to cardiac catheterization and
biopsy for transplant rejection.
The transplanted heart in the baseline state has abnormal twist yet
maintains similar strain patterns compared with the normal left
ventricle. Further evaluation is needed to determine if there are
clinically important sequelae that result from alterations in twist and
whether these abnormalities progress or resolve with time. If changes
in left ventricular regional strain and wall motion determined by MRI
and SPAMM are able to predict those patients with histological
rejection, this will limit the need for cardiac catheterization and
endomyocardial biopsy, procedures that have a small but significant risk.
| |
ACKNOWLEDGEMENTS |
Present addresses: C. Ramaciotti, Dallas Children's Hospital,
Univ. of Texas, Dept. of Pediatrics, Div. of Cardiology, 5323 Harry
Hines Blvd., Dallas, TX 75235; M. L. Jacobs, St. Christopher's Hospital for Children, Dept. of Surgery, Div. of Cardiothoracic Surgery, Front and Erie St., Philadelphia, PA 19134; M. A. Fogel, Wyeth-Ayerst Research, c/o 51 Levering Circle, Bala Cynwyd, PA 19004;
K. Fellows, Dept. of Radiology, and B. Clark and P. Weinberg, Dept. of
Pediatrics, Div. of Cardiology, Children's Hospital of Philadelphia,
34th and Civic Center Blvd., Philadelphia, PA 19104.
| |
FOOTNOTES |
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests, other correspondence, and present
location: M. T. Donofrio, Division of Pediatric Cardiology, Medical
College of Virginia-Virginia Commonwealth Univ., Box 980342, Richmond,
VA 23298 (E-mail: mdonofri{at}hsc.vcu.edu).
Received 2 June 1998; accepted in final form 15 July 1999.
| |
REFERENCES |
1.
Axel, L.,
and
L. Dougherty.
MR imaging of motion with spatial modulation of magnetization.
Radiology
171:
841-845,
1989[Abstract/Free Full Text].
2.
Axel, L.,
and
L. Dougherty.
Heart wall motion: improved method of spatial modulation of magnetization for MR imaging.
Radiology
172:
349-350,
1989[Abstract/Free Full Text].
3.
Beyar, R.,
F. C. P. Yin,
M. Hausknecht,
M. L. Weisfeldt,
and
D. A. Kass.
Dependence of left ventricular twist-radial shortening relation on cardiac cycle phase.
Am. J. Physiol.
257 (Heart Circ. Physiol. 26):
H1119-H1126,
1989[Abstract/Free Full Text].
4.
Billingham, M. E.,
N. R. B. Cary,
M. E. Hammond,
J. Kemnitz,
C. Marboe,
H. A. McCallister,
D. C. Snovan,
G. L Winters,
and
A. Zerbe.
A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: heart rejection study group.
J. Heart Transplant.
9:
587-93,
1990[Medline].
5.
Buchalter, M. B.,
J. L. Weiss,
W. J. Rogers,
and
E. A. Zerhouni.
Non-invasive quantification of left ventricular rotational deformation in normal humans using magnetic resonance imaging myocardial tagging.
Circulation
81:
1236-1244,
1990[Abstract/Free Full Text].
6.
DeFloriani, L.
Surface representation based on triangular grids.
Visual Computer
3:
27-50,
1987.
7.
Fogel, M. A.,
K. B. Gupta,
P. M. Weinberg,
and
E. A. Hoffman.
Regional wall motion and strain analysis across stages of Fontan reconstruction by magnetic resonance tagging.
Am. J. Physiol.
269 (Heart Circ. Physiol. 38):
H1132-H1152,
1995[Abstract/Free Full Text].
8.
Fogel, M. A.,
P. M. Weinberg,
K. E. Fellows,
and
E. A. Hoffman.
A study in ventricular-ventricular interaction. Single right ventricle compared with systemic right ventricle in a dual-chamber circulation.
Circulation
92:
219-230,
1995[Abstract/Free Full Text].
9.
Goto, Y.,
S. Futaki,
O. Kawaguchi,
K. Hata,
and
T. Takasago.
Coupling between regional myocardial oxygen consumption and contraction under altered preload and afterload.
J. Am. Coll. Cardiol.
21:
1522-1531,
1993[Abstract].
10.
Hansen, D. E.,
G. T. Daughters,
E. L. Alderman,
N. B. Ingels,
and
D. C. Miller.
Torsional deformation of the left ventricular midwall in human hearts with intramyocardial markers: regional heterogeneity and sensitivity to the inotropic effects of abrupt rate switches.
Circ. Res.
62:
941-952,
1988[Abstract/Free Full Text].
11.
Hansen, D. E.,
G. T. Daughters,
E. L. Alderman,
E. B. Stinson,
J. C. Baldwin,
and
D. C. Miller.
Effect of acute human cardiac allograft rejection on left ventricular systolic torsion and diastolic recoil measured by intramyocardial markers.
Circulation
76:
998-1008,
1987[Abstract/Free Full Text].
12.
Hauerich, A.,
W. C. Scott,
K. D. Dawkins,
M. E. Billingham,
and
S. W. Jamieson.
Asymmetric pattern of rejection following orthotopic cardiac transplantation in primates.
Heart
4:
280-285,
1984.
13.
Hoffman, E. A.,
D. Gnanaprakasam,
K. B. Gupta,
J. D. Hoford,
S. D. Kugelmass,
and
R. S. Kulawiec.
VIDA: an environment for multidimensional image display and analysis.
Proc. SPIE
1660:
694-711,
1992.
14.
Lee, D. T.,
and
B. J. Schachter.
Two algorithms for constructing a DeLaunay triangulation.
Int. J. Computer Info. Sci.
9:
219-242,
1980.
15.
Young, A. A.,
L. Axel,
L. Dougherty,
D. K. Bogen,
and
C. S. Parenteau.
Validation of tagging with MR imaging to estimate material deformation.
Radiology
188:
101-108,
1993[Abstract/Free Full Text].
16.
Young, A. A.,
H. Imai,
C. N. Chang,
and
L. Axel.
Two dimensional left ventricular deformation during systole using magnetic resonance imaging with spatial modulation of magnetization.
Circulation
89:
740-752,
1989[Abstract/Free Full Text].
Am J Physiol Regul Integr Compar Physiol 277(5):R1481-R1487
0002-9513/99 $5.00
Copyright © 1999 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
![&THgr; = cos<SUP>−1</SUP> [(<IT>P</IT><SUB><IT>n</IT></SUB> − <IT>P</IT><SUB><IT>n+1</IT></SUB>) / (‖<A><AC><IT>P</IT><SUB><IT>n</IT></SUB></AC><AC>&cjs1166;</AC></A>‖ ‖<A><AC><IT>P</IT><SUB><IT>n+1</IT></SUB>‖</AC><AC>&cjs1166;</AC></A>)]](/content/vol277/issue5/fulltext/R1481/img003.gif)
