| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 1995;91:1143-1153.)
© 1995 American Heart Association, Inc.
Articles |
Effect of Coronary Artery Reperfusion on Transmural Myocardial Remodeling in Dogs
Presented in part at the 67th Scientific Sessions of the American Heart Association, Atlanta, Ga, November 11, 1993, and presented in abstract form in Circulation 1993;88:I-627.
From the Division of Cardiology, Department of Medicine, University of California, San Diego, La Jolla.
Correspondence to John Ross, Jr, MD, Department of Medicine, 0613B, University of California, San Diego, 9500 Gilman Dr, La Jolla, CA 92093-0613.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background The effects of reperfusion after coronary occlusion on transmural remodeling of the ischemic region early and late after nontransmural infarction must importantly affect the recovery of regional function. Accordingly, analysis of local volume and three-dimensional strain was performed using a finite element method to determine regional remodeling. Systolic and remodeling strains were measured using radiographic imaging of three columns (
1 cm apart) of four to six gold beads implanted across the
left ventricular posterior wall in 6 dogs. Methods and Results After a control study, infarction was produced by 2 to 4 hours of proximal left circumflex coronary artery occlusion followed by reperfusion. Follow-up studies were performed at 2 days, 3 weeks, and 12 weeks with the dogs under anesthesia and in closed-chest conditions. Biplane cineradiography was performed to obtain the three-dimensional coordinates of the beads. At 2 days, end-systolic strains were akinetic with loss of normal transmural gradients of shortening and thickening. Remodeling strains (RS) were determined by use of a nonhomogeneous finite element method by referring the end-diastolic configuration during follow-up studies to its control state at matched end-diastolic pressures and heart rates. Tissue volume at 2 days increased substantially, more at the endocardium (30±7%) than at the epicardium (5±12%, P<.01); the increase was associated with an average RS in the wall-thickening direction of 0.18±0.15 (P<.01) with all other RS near zero. At 12 weeks systolic function partially recovered, with normal wall thickening in the epicardium (radial strain, 0.081±0.056 [control] versus 0.113±0.088 [12 weeks]) but with dysfunction in the endocardium (0.245±0.108 [control] versus 0.111±0.074 [P<.01] [12 weeks]). This inability of the inner wall to recover function may be related to increased transmural torsional shear and negative longitudinal-radial transverse shear in the inner wall. Volume loss occurred at 12 weeks in the endocardium (-36±16%) corresponding to transmural gradients in longitudinal RS and both transverse shear RS. Negative longitudinal RS was greater at the endocardium (-0.20±0.10) than at the epicardium (-0.06±0.05, P<.01).
Conclusions These results indicate the presence of marked subendocardial edema 2 days after reperfusion following 2 to 4 hours of coronary occlusion. At 3 months after reperfusion, however, there was volume loss in the inner wall due to shrinkage along the myofiber direction with reduced transmural function and loss of longitudinal shortening, while both tissue volume and function recovered completely in the outer wall.
Key Words: myocardial contraction • myocardial infarction • edema • radiography
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Coronary artery reperfusion has several favorable effects on left ventricular remodeling and function after acute myocardial infarction. In experimental animals the effects of coronary occlusion followed by reperfusion vary by species,1 but reperfusion before infarction is complete results in salvage of myocardial tissue and variable recovery of regional2 3 4 5 and global6 function, with prevention of infarct expansion.7 Also, clinical studies8 9 10 11 have demonstrated that early reperfusion by thrombolysis after acute coronary artery occlusion reduces infarct size and improves regional wall motion and left ventricular function. In the dog, the outer wall is usually spared by early reperfusion,12 13 and although late recovery of overall regional function in the ischemic zone after coronary artery reperfusion has been demonstrated, the transmural variation of functional recovery is uncertain. It has been speculated that sparing of the outer wall might play a role in remodeling after reperfusion of the myocardium, and recent studies from our laboratory have shown myocardial cell hypertrophy of surviving regions of the infarcted wall after reperfusion, particularly in the outer wall,14 suggesting that a hypertrophic response might contribute to late recovery of regional function in the ischemic zone.
Histological techniques and uniaxial dimension measurements, however, do not directly address the questions of whether regional tissue volumes change, what the time course of such changes is, and whether reperfused myocardium is functional in directions other than the single measured dimension. In this study, we characterized regional transmural deformation in the reperfused myocardium by measuring three-dimensional finite strains. This method can be used to directly address the time course of changes in tissue volume using local material markers as well as to determine the temporal course and directions of both systolic and diastolic (remodeling) strains. To measure regional three-dimensional finite strains, three columns of gold beads were implanted in dogs across the left ventricular posterior wall by use of the technique of Waldman and coworkers.15
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The animals in this study were handled according to the animal welfare regulations of the University of California, San Diego, and the guidelines of the American Association for Accreditation of Laboratory Animal Care, which are in accord with the animal use principles of the American Heart Association. The experimental protocol was approved by the Animal Subjects Committee of the University of California, San Diego.
Animal Preparation
Twenty-six mongrel dogs of both sexes,
weighing between 22.6 and
42.0 kg (mean weight, 31.0 kg), were used in this study. Results
reported are from 6 animals for which complete data were available at
control and after 48 hours, 3 weeks, and 3 months of reperfusion. There
were 3 operative deaths at the initial surgery, and 4 animals died
within the first postoperative week (2 were found dead and 2 had
thromboembolism). Five animals exhibited no dyskinesia during the
coronary occlusion and their data were excluded; 4 died of ventricular
arrhythmia during coronary occlusion or reperfusion, and 4 died or were
euthanized late (infection, instrumentation failure, death of unknown
causes). On the day of the initial surgery, dogs were premedicated with
morphine (10.5 mg/kg IM) and anesthetized with sodium thiamylal (20
mg/kg IV). After endotracheal intubation, anesthesia was maintained
with isoflurane (1% to 2%). Arterial blood gases were measured
repeatedly during the surgery to keep PO2
above 150 mm Hg and PCO2 and pH within the
physiological range. A left thoracotomy was performed at the fifth
intercostal space, and the pericardium was opened. A high-fidelity
micromanometer (Konigsberg P7) and a Tygon fluid-filled catheter
(diameter, 1.27 mm) were inserted through a stab wound in the apex to
measure left ventricular pressure. The zero-pressure reference of the
fluid-filled catheter was taken at the level of the right atrium, and
the micromanometer was calibrated by matching the left ventricular
pressure with the fluid-filled catheter (Statham P-23Db). A 2-cm length
of the proximal left circumflex coronary artery was dissected free to
place a Silastic hydraulic cuff occluder and a Doppler ultrasonic
flowmeter probe around the artery. All visible epicardial coronary
collateral vessels between the left anterior descending artery and the
circumflex artery below the coronary occluder were ligated. Bipolar
pacing electrodes constructed from Teflon-coated stainless steel wire
were sutured to the left atrial appendage.
Three columns of five gold
beads (diameter, 0.9 mm) were implanted in
the posterior free wall of the left ventricle using the technique
described by Waldman and coworkers.15 Five additional
reference markers (gold beads 1.5 mm in diameter) were sutured to the
epicardium: three directly above the three columns, one at the
bifurcation of the left coronary artery (base reference marker), and
one at the apical dimple (apex reference marker), as shown in Fig
1A
. In 17 dogs one pair of ultrasonic crystals was
implanted circumferentially in the posterior wall near the three
columns of beads for assessment of wall motion abnormality during
coronary occlusion. The pericardium was left open, and all wires and
catheters were exteriorized to the back of the animal. The thorax was
closed and the animal was allowed to recover.
|
Experimental Protocols
Control studies were performed 1 to 2
weeks after
instrumentation surgery, when the animal had recovered fully. On the
day of the study, animals were anesthetized with sodium thiamylal (20
mg/kg IV). After endotracheal intubation, anesthesia was maintained
with halothane (0.8% to 1.0%) during the study. High-speed biplane
cineradiography (120 frames/s) was performed to image the bead
movements during the cardiac cycle. Cine cameras were free running, and
the maximum temporal difference between the anteroposterior and lateral
cameras was 4 milliseconds. Respiration was halted at end-expiration
while simultaneous recordings of ECG, left ventricular pressure, left
ventricular dP/dt, and camera shutter marks were also made. Data were
recorded on an eight-channel chart recorder at the speed of 200 mm/s
(Gould, Brush), and digitized data (sampling interval, 4 milliseconds)
were simultaneously recorded on a microcomputer. In the control study,
data were taken at three different left ventricular
end-diastolic pressures to facilitate matching of
end-diastolic pressure between control and later studies.
End-diastolic pressure was increased by intravenous
infusion with dextran (70% in 0.9% NaCl). At the end of each study, a
calibration phantom was imaged for subsequent reconstruction of
three-dimensional bead coordinates from the biplane views.
Within a week after the control study, circumflex coronary occlusion was performed after analgesia and sedation with acepromazine (0.3 mg/kg IM) and buprenorphine (0.3 mg/kg IM). Before coronary occlusion, a bolus of lidocaine (40 to 50 mg IV) was injected through the left ventricular catheter and then continuously infused at 1 to 2 mg/min to minimize ventricular arrhythmias. Coronary artery occlusion was performed by inflating the Silastic cuff occluder with air or mineral oil. The initial occlusion was maintained for 2 hours while ECG, left ventricular pressure, segment length, and coronary flow were simultaneously monitored. If wall motion in the ischemic region was not severely hypokinetic at 2 hours after coronary occlusion, the occlusion was extended to 3 or 4 hours to induce myocardial infarction. Before reperfusion, a bolus of lidocaine (20 to 40 mg) was administered, and the atrium was paced at a rate of 150 beats per minute for overdrive suppression of ventricular arrhythmias during the first 5 minutes after reperfusion. Two hours after reperfusion, or after a stable condition was achieved, animals were transferred to a recovery cage. Tocainide (600 mg/day PO) was given for the first 3 days.
Follow-up studies were performed at 2 days and at 3, 8, and 12 weeks after reperfusion. Data are presented in this manuscript from 6 animals in which suitable data were available at each time point. In each study, the animals were again anesthetized and ventilated with a halothane mixture. High-speed cineradiography and hemodynamic data acquisition were repeated with left ventricular end-diastolic pressure adjusted as described above to match that of the control study.
The animals were euthanized after 12 weeks of reperfusion. With the dog under general anesthesia, as described above, the chest was opened by means of median sternotomy. The heart was arrested with an injection of saturated KCl or by hypoxia and perfused retrogradely from the aorta with glutaraldehyde (2% in buffer) at a constant intraventricular pressure of 10 to 12 mm Hg. After the heart was fixed, it was excised for morphological analysis. We carefully examined the position of beads using fluoroscopy, and excluded beads in the papillary muscle from the strain analysis.
Morphological Analysis
The fixed left ventricle was cut
transversely perpendicular to
the left ventricular long axis into rings approximately 1 cm thick.
Muscle and scar areas were traced on sheets of clear plastic, and
slices that contained the bead set were measured by planimetry. The
degree of infarction was expressed as the percentage of total left
ventricular ring area that was scar.
A transmural block of tissue was then cut across the ventricular wall from the center of the region in which beads had been implanted. The block was embedded in paraffin, sliced at a thickness of 10 µm, and stained with Milligan trichrome stain for analysis of scar formation. Percent scar area was measured by a point-counting method.
Strain and Tissue Volume Analysis
Corresponding images of the
calibration phantom were used to
compute a perspective transformation, which was then applied to the
corresponding two-dimensional coordinates of implanted myocardial
markers to reconstruct their three-dimensional
coordinates.15 The three-dimensional x-ray coordinates of
the beads were transformed to a right-handed Cartesian cardiac
coordinate system defined by the apex and base reference beads and the
three surface beads above each column.15 16 The
cardiac
coordinate system consists of three mutually orthogonal coordinates
(circumferential, longitudinal, and radial) at the center of the
triangle formed by the surface beads above the three columns. The
lateral and anterior-posterior cineradiographic views of the
end-diastolic and end-systolic frames were separately
projected onto a digitizing pad (Sketchpro, Hewlett-Packard) to obtain
two-dimensional bead coordinates. End-diastole was defined as the time
immediately following the A wave coincident with rapid onset of a
positive left ventricular dP/dt, and end-systole was defined as 20
milliseconds before peak negative dP/dt. Regional myocardial volumes
were determined from tetrahedral volumes defined by four noncoplanar
markers as described previously15 (Fig 1B
).
Tetrahedra
were selected with the criteria of being 2.0 to 4.0 mm in radial height
and 15 to 70 mm3 in volume in the end-diastolic
reference state of the control study, and the same tetrahedra were used
to determine the regional myocardial volumes throughout the study in
each animal. Tetrahedra whose centroids were located 0% to 30% of the
distance from the epicardium were averaged to represent
epicardial volume, and tetrahedra with depths of 31% to 60% and 61%
to 90% were averaged for midwall and endocardial volumes,
respectively.
For calculating systolic and remodeling strains, we
followed a finite
element method that can be used to determine continuous, nonhomogeneous
strain variations across the wall from the same bead
coordinates17 (Fig 1A). In this approach, each
bead was
assigned to a material point on the element and the element was fitted
to the coordinates of the set of markers using a least-squares method,
with a bilinear variation in the plane of the wall and a quadratic
polynomial basis function in the transmural direction. Finally, the
single bilinear-quadratic lagrangian finite element incorporated
coordinate data from the entire bead set, and strains were determined
from the deformation of the finite element between a given reference
and deformed states. The mean results for transmural distributions of
strain were obtained at seven discrete locations spanning the entire
transmural bead set. Normal strains expressed by
E11, E22, and E33
represent extensional deformations along the circumferential,
longitudinal, and radial directions, respectively. Shear strains
(E12, E13, and E23)
represent changes in angle between pairs of line segments in
each of the three respective mutually orthogonal coordinate planes; for
example, if two line segments are initially orthogonal and parallel to
the circumferential and longitudinal directions and the angle between
them becomes acute during a deformation, positive E12 has
occurred. Systolic strains were calculated by relative changes of the
end-systolic configuration (deformed state) from the
end-diastolic configuration (reference state) in the same
cardiac cycle. To determine the change in the end-diastolic
configuration between the control and subsequent studies, remodeling
strains were calculated by choosing end-diastole of the control study
as the reference state and end-diastole of a later study as the
deformed state at the same end-diastolic pressure.
Statistical Analysis
All values are expressed as
mean±SD. Hemodynamic variables were
analyzed with a one-factor repeated-measures ANOVA (P<.05
considered significant) in which values at 2 days, 3 weeks, and 12
weeks were compared with control values. There were no significant
differences in any hemodynamic indexes throughout the study (Table
1). To compare strains and tissue volumes over time
between animals, it was necessary to compare them at the same depth.
Volumes of tetrahedra determined directly from the coordinates of four
beads were grouped according to the percent wall thickness calculated
from the distance of the centroid of the tetrahedron from the
epicardium and the wall thickness at 12 weeks. Strains and volume
changes calculated using the finite element method were grouped at
corresponding Gauss points of the finite elements. The bases of the
elements averaged 95.4% of the total wall thickness (mean, 11.2 mm)
determined at postmortem. Any bead accidentally placed in a papillary
muscle was ignored. Systolic and remodeling strains were analyzed with
a two-factor ANOVA. Strains were compared at three depths: midwall
(MID=50% of wall thickness), subendocardium (ENDO=97% of wall
thickness), and subepicardium (EPI=3% of wall thickness). Systolic
strains at 2 days, 3 weeks, and 12 weeks were compared with those at
control. Here, average or mean transmural values were compared as well
as strains at the three depths. The significance of differences of
remodeling strains from zero was also analyzed with t tests.
Bonferroni corrections were applied to account for multiple
comparisons.
|
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Systolic Strains
Control
End-systolic circumferential strain (E11) under control conditions in the region to be rendered ischemic was negative, indicating shortening, as shown in Fig 2
from a
representative animal. Longitudinal strain (E22)
was also negative (Fig 2), and there was a significant
transmural
gradient for this component, showing greater shortening with depth
(Table 2: mean, -0.041 at EPI versus -0.087 at
ENDO).
Average radial strain (E33) was positive (Fig 2) and
also
exhibited a significant transmural gradient, with greater thickening
toward the inner wall (Table 2: mean, 0.081 at EPI versus 0.245
at
ENDO). Mean transmural in-plane or torsional shear (E12)
and both transverse shear strains (E13 and E23)
were small and positive (Table 2). E12 showed a
significant
transmural gradient, becoming more positive with depth (Table
2: mean,
-0.005 at EPI versus 0.040 at ENDO), but there were no significant
transmural gradients in either transverse shear strains, both with
means of 0.04.
|
|
Two Days After Reperfusion
Compared
with control, overall function as indicated by the three
normal strains (E11, E22, and
E33) at 2 days after reperfusion was severely depressed,
with loss of the transmural gradients of E22 and
E33, as shown in 1 animal (Fig 2). Mean values of
transmural E11, E22, and
E33 were close to zero, indicating akinesis (Table
2).
Although there were no significant changes in mean shear strains,
significant transmural gradients were observed in both in-plane and
transverse shear strains. E12 and E13 showed
more positive shear with depth, and E23 was more negative
with depth. Both transverse shear strains indicated substantially
smaller or more negative values than control, particularly in the inner
wall (Table 2: 0.021 at 2 days versus 0.061 at control for
E13 at ENDO; -0.036 at 2 days versus 0.033 at control for
E23 at ENDO).
Three Weeks After Reperfusion
Normal strains indicated partial recovery at 3 weeks after
reperfusion, and compared with 2 days there was significant shortening
and wall thickening (Table 2). Mean E11
(-0.033) and
E33 (0.082) were still significantly depressed compared
with control, with loss of the transmural gradient in E33.
E11 in the outer half of the wall was depressed (Table
2:
-0.012 at 3 weeks versus -0.059 during control at EPI;
-0.016 at 3
weeks versus -0.084 during control at MID), while E33 in
the inner half had recovered only partially (0.082 at 3 weeks versus
0.177 during control at MID; 0.083 at 3 weeks versus 0.245 during
control at ENDO). Average E22 (-0.039) recovered to near
the control level but was still significantly depressed near the
endocardium (-0.052 at 3 weeks versus -0.087 at control). Mean
in-plane shear, E12, increased significantly (0.053
at 3 weeks versus 0.016 at control), with loss of the transmural
gradient. Mean E13 and E23 showed no
significant changes, but near the endocardium both transverse shears
were smaller or more negative, similar to observations at 2 days (Table
2: 0.017 at 3 weeks versus 0.061 at control for E13
and
-0.011 at 3 weeks versus 0.033 at control for E23). At 3
weeks after reperfusion, shear in the longitudinal-radial plane
(E23) continued to exhibit a negative transmural
gradient.
Twelve Weeks After Reperfusion
At 12
weeks after reperfusion (Fig 2 and Table 2),
average
circumferential shortening (E11) was not significantly
different from control, but longitudinal function (E22) was
akinetic, shortening having been lost in that direction that had
partially recovered at 3 weeks. Compared with control, both in-plane
normal strains (E11 and E22) showed normal
function near the epicardium but significantly depressed function in
the inner half of the wall (-0.035 at 12 weeks versus -0.095
during
control at ENDO for E11; 0.010 at 12 weeks versus -0.087
during control at ENDO for E22). Circumferential shortening
(E11), which had recovered near the endocardium after 3
weeks, was again abnormal. Average radial strain (E33) also
decreased significantly (0.098 at 12 weeks versus 0.169 at control)
with loss of normal wall thickening near the endocardium and at midwall
(0.111 at 12 weeks versus 0.245 during control at ENDO), but
E33 near the epicardium completely recovered. Mean
E23 decreased (-0.020 at 12 weeks versus 0.040 during
control), being more negative in the inner half of the wall (-0.019 at
12 weeks versus 0.042 during control at MID; -0.060 at 12 weeks versus
0.033 during control at ENDO), resulting in a negative transmural
gradient at 12 weeks (0.018 at EPI versus -0.019 at MID and -0.060
at
ENDO) similar to that at both 2 days and 3 weeks after
reperfusion.
Remodeling: Tissue Volume Changes
Tissue volume changes were
determined in two ways from the
three-dimensional bead positions (see "Methods"): by use of local
tetrahedral volumes or of the finite element method. Average results
from both methods are given in Table 3. Transmural
trends using the finite element method, shown in Fig 3,
were also used because of the added capability to interpolate at more
depths.
|
|
At 2 days after reperfusion, regional myocardial volumes were
increased
transmurally with a significant transmural gradient. Average increases
in myocardial volumes were 18% using the finite element method (Table
3, top) and 22% using the tetrahedral method (Table
3, bottom).
Increases in regional volumes were significantly greater with depth
(Table 3, top: 5.4% increase at EPI versus 19.0% at MID and
29.5% at
ENDO).
At 3 weeks, marked decreases in volume were observed in the
subendocardium with a significant transmural gradient (2.0% decrease
at EPI versus 14.6% at MID and 33.0% at ENDO). At 12 weeks, changes
of regional volumes showed trends very similar to those at 3 weeks,
with a similar transmural gradient (2.0% increase at EPI versus 14.4%
decrease at MID and 36.3% decrease at ENDO) (see also Fig 3).
Further
decreases in volumes were observed in the subendocardium, while
subepicardial volumes were preserved.
Remodeling: Finite Strains
Remodeling strains were calculated
from finite elements fitted to
end-diastolic marker coordinates at control (reference
configuration) and at 2 days, 3 weeks, and 12 weeks after reperfusion.
Two-dimensional projections of the marker positions and fitted finite
elements in one experiment are shown (Fig 4). At day 2,
wall thickness was increased, with an increase in myocardial volume. In
contrast, at week 12, endocardial and midwall tissue shrinkage was
observed.
|
Remodeling strains at 2 days and 12 weeks after reperfusion
are shown
for a representative animal in Fig 5 and for all
animals in Table 4.
|
|
Two Days After Reperfusion
Average circumferential (E11) and longitudinal
(E22) remodeling strains were not significantly different
from zero and were transmurally uniform at 2 days after reperfusion
(Fig 5, Table 4). Average radial strain
(E33), which
indicates changes in local wall thickness, was positive and
significantly different from zero (Table 4: mean
E33, 0.185). E33 also indicated an
increase in thickening with depth, resulting in a large transmural
gradient (0.039 at EPI versus 0.186 at MID and 0.330 at ENDO). Mean
in-plane and transverse shear strains were small and not significantly
different from zero. However, remodeling shear in the
circumferential-radial plane (E13) developed a significant
transmural gradient (-0.009 at EPI versus 0.015 at MID and 0.055 at
ENDO).
Three Weeks After Reperfusion
In contrast to
2 days after reperfusion, both E11 and
E22 became negative and were significantly different from
zero (Table 4: mean E11, -0.114; mean
E22, -0.115) at 3 weeks after reperfusion,
indicating shrinkage in both directions. E22 became more
negative at all depths compared with 2 days after reperfusion and
developed a large transmural gradient (-0.052 at EPI versus -0.118
at
MID and -0.163 at ENDO). E33 was positive and
significantly different from zero (mean E33, 0.159).
Compared with 2 days after reperfusion, there were significant
increases in epicardial E33 at 3 weeks after reperfusion
(0.174 versus 0.039 at 2 days) and decreases in endocardial
E33 (0.190 versus 0.330 at 2 days), with a loss of the
transmural gradient. E12 became uniformly negative across
the wall and on average was significantly different from zero (mean
E12, -0.030). Although there were no significant
differences from zero in mean transverse shear strains, E23
developed a very large transmural gradient, changing sign near midwall
(-0.064 at EPI versus 0.033 at MID and 0.145 at ENDO).
Twelve Weeks After Reperfusion
Transmural trends in
one experiment are shown in Fig 5. Although
the circumferential shrinkage (E11) observed at 3 weeks
after reperfusion was less marked by 12 weeks, mean E22 was
still significantly different from zero, indicating additional
longitudinal shrinkage (Table 4: mean E22,
-0.144),
and was more negative at midwall and the subendocardium, with a
significant transmural gradient similar to but even larger than that
observed at 3 weeks after reperfusion (-0.065 at EPI versus -0.154
at
MID and -0.197 at ENDO). E33 indicated substantial
thickening (mean E33, 0.155) with no transmural
gradient, but a tendency for further increases in epicardial
E33 (0.228) and decreases in endocardial E33
(0.121) was observed at 12 weeks after reperfusion compared with 3
weeks.
On average, E12 was significantly different from zero
(mean
E12, -0.041) and was uniformly negative across the
wall, with values similar to those at 3 weeks. Transverse shear strains
at 12 weeks after reperfusion showed trends similar to those at 3
weeks. Although E13 was not significantly different from
zero, a more positive epicardial E13 and negative
endocardial E13 resulted in a significant transmural
gradient (0.110 at EPI versus 0.073 at MID and -0.055 at ENDO). On
average E23 was positive and significantly different from
zero (mean E23, 0.068), and it maintained the large
transmural gradient observed at 3 weeks after reperfusion (-0.022 at
EPI versus 0.063 at MID and 0.172 at ENDO). By computing principal
remodeling strains and associated principal directions from the average
remodeling strain components (Table 4), we found maximum
shrinkage in
the subendocardium, as indicated by the most negative principal value,
of -0.274 with an orientation of 85° from circumferential. The
relation between a principal stretch and strain is calculated from the
equation 
=(2E+1)1/2; this corresponds to a stretch
ratio
of 0.672 or a shrinkage of 32.8%, occurring predominantly in the plane
of the myofibers and directed longitudinally.
Histology
Global scar area in myocardial slices containing
the beads was
from 4.5% to 14.2% (average, 7.9±3.5%). Histological examination
indicated that scar formation was least in the subepicardium and
greatest in the subendocardium: 4.9% in the subepicardium, 13.7% in
the midmyocardium, and 44.1% in the subendocardium.
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The present study demonstrated for the first time that function and tissue volume recover to normal values in the outer infarcted reperfused myocardium while the inner wall of the infarcted area undergoes remodeling, which, after initial marked edema, is associated with tissue loss and reduced systolic function. This remodeling process is clearly different from that recently observed in a pig model of coronary occlusion without reperfusion.18 In that model the wall thinned, and there was extensive scar (80% at the endocardium) and little return of function in either the outer or inner wall. In the present study, 2 days after myocardial infarction and reperfusion in the dog, myocardial tissue volume was substantially increased and systolic function was absent. After 12 weeks, epicardial tissue had thickened while almost normal systolic function was regained. The endocardium remodeled in a unique fashion: there was significant tissue loss, with fibrosis and shrinkage in the plane of the myofibers, and the wall thickened somewhat. Substantial shearing deformation was also involved in the remodeling process, including uniform negative torsional shear, an overall increase in longitudinal-radial shear, and large gradients in both transverse shear remodeling strains, especially in the longitudinal radial plane. Therefore, previous uniaxial measurements could substantially underestimate the extent of remodeling. Moreover, complex deformations involving both torsion and bending may be involved in the remodeling process. Endocardial function remained depressed because of a loss of the normal transmural gradients of systolic shortening and thickening and the development of negative systolic shear in the longitudinal radial plane after 12 weeks of reperfusion.
Tissue Remodeling
Forty-eight hours after reperfusion,
regional myocardial volumes
were increased at all sites, with a large transmural gradient
increasing more toward the subendocardium. Cell swelling, vascular
endothelial damage, and interstitial edema have been observed in
reperfused myocardium in previous
studies.19 20 21 22 23
Whalen et
al22 demonstrated that tissue water in the ischemic
myocardium rapidly increased after reperfusion and that tissue edema
was primarily the result of cell swelling. Jennings et
al24 demonstrated a transmural difference of tissue water
in the reperfused myocardium, there being more water in the endocardial
tissue, a finding similar to ours. During coronary occlusion,
endocardial ischemia is usually more severe than epicardial
ischemia.2 25 26 A likely mechanism for
the nonuniform
tissue edema is that more severe ischemia in the subendocardium causes
more severe cell damage after reperfusion in that region.
Reconfiguration of the ischemic myocardium with reperfusion has not been examined in detail. In previous studies using uniaxial dimension gauges, it was demonstrated that a significant increase in end-diastolic wall thickness in the ischemic zone was observed immediately after reperfusion, whereas the change in end-diastolic segment length with reperfusion was small.2 3 14 Villarreal et al27 assessed transmural myocardial deformation during acute ischemia by measuring three-dimensional finite strains and demonstrated a significant passive reconfiguration that varied transmurally with lengthening in the epicardial tangent plane, and wall thinning increased from epicardium toward endocardium. In contrast, a much different reconfiguration with tissue edema occurred after reperfusion compared with acute ischemia in the present study. The remodeling strains showed that greater increases in radial lengths with depth occurred without significant changes in circumferential and longitudinal lengths at 2 days after reperfusion, indicating greater increases in end-diastolic wall thickness with depth. This transmural trend in radial strain was similar to that of the regional myocardial volume changes, so the local increases in myocardial volumes were caused primarily by corresponding increases in local wall thickness. May-Newman et al28 found large radial strains with a transmural gradient (ENDO>EPI), as well as a positive transverse shear strain (E13), due to perfusion without ischemia in an unloaded isolated heart model, results similar to those in the present study. Why the tissue edema with reperfusion develops predominantly in the radial direction after reperfusion is uncertain. One possible mechanism is a change in the collagen matrix with ischemia. Some studies have demonstrated breakage of the collagen network after ischemia.29 30
Theroux et al2 demonstrated progressive decreases in end-diastolic length in ischemic segments after reperfusion, compatible with tissue loss, and some improvement in shortening. However, subendocardial tissue loss with reperfusion seems to be less than that with permanent coronary occlusion.31 Similar late subendocardial tissue loss with reperfusion was observed in a study by Lavallee et al.4 In sustained coronary occlusion in the pig, tissue volume loss in the inner wall averaged 60%, nearly twice the amount observed in the present study.18 Sasayama et al32 reported disproportional late shrinkage between the subendocardial segment and the wall thickness after chronic circumflex artery occlusion, suggesting the possibility of compensatory hypertrophy of residual myocytes in the ischemic zone. Experimentally, cellular hypertrophy was shown in the noninfarcted and marginal regions in rats with transmural infarction,33 34 and Kambayashi et al14 recently demonstrated larger cross-sectional areas of cells in reperfused than in nonischemic myocardium, concluding that hypertrophy of surviving regions of the infarcted wall, particularly the outer wall, might play a role in the late partial recovery of regional function. In the present study, marked reductions of regional myocardial volumes in the subendocardium, compatible with subendocardial tissue shrinkage, were observed from 3 through 12 weeks after reperfusion. There was no significant change of subepicardial tissue volume at 12 weeks. Thus, tissue volumes in this reperfusion model were substantially preserved compared with those in the model of permanent coronary occlusion.18 However, the significant shrinkage in the fiber plane in the subepicardium probably indicates that there was a combination of tissue loss with significant hypertrophy.
Analysis of remodeling strains gives more detailed information about the process of tissue shrinkage and hypertrophy with reperfusion after coronary occlusion. In a model of volume overload,35 uniform increases in myocardial mass were observed across the wall, and end-diastolic growth or remodeling strains obtained in the hypertrophic state with respect to the control state showed positive strains in both the circumferential and longitudinal components with small radial growth strains. These results indicated that remodeling due to volume overload was predominantly parallel to the epicardial tangent plane. In contrast, quite a different growth or remodeling process with ischemia and reperfusion was found in the present study. In-plane radial strains, on average negative, and radial strains, on average positive, were observed transmurally at 12 weeks after reperfusion, accompanied by positive transverse shear (E23). Longitudinal strain was more negative than circumferential strain, and showed a large transmural gradient, with much greater shrinkage at the endocardium. The general pattern of in-plane shrinkage and tissue loss was similar to that observed in the pig without reperfusion,18 but the shrinkage direction was substantially different, being much more longitudinal. Whether this is due to species differences or the presence of viable myocytes and extracellular matrix after reperfusion is not possible to determine. However, the latter explanation seems more likely because there were still myocytes in the subendocardial tissue and because the contraction direction corresponded much more to the local architecture seen at control than it does in the pig without reperfusion. Moreover, the modest transmural increase in wall thickness is unique to this model.
Recovery of Systolic Function
The control measurements in the
present study showed greater
systolic shortening and thickening in the inner layers, compatible with
results of previous
studies,36 37 38 39 and
normal transmural
strain gradients were observed, as in previous
studies.15 40
A number of studies have demonstrated postischemic dysfunction after coronary reperfusion in various experimental settings, as discussed earlier, and it has generally been accepted that in the dog model, reperfusion after 20 minutes to 3 hours of coronary occlusion can salvage subepicardial tissue despite subendocardial infarction.1 In the present study at 2 days after reperfusion, there was no active shortening or thickening in the ischemic zone and there was loss of normal transmural strain gradients, in agreement with the echocardiographic data of Ellis et al41 ; these investigators reported that active wall thickening assessed by serial two-dimensional echocardiography was not observed until 72 hours of reperfusion after 2 hours of coronary occlusion.
Late partial recovery of systolic function after reperfusion was observed in the ischemic zone, as described previously,2 3 4 5 14 42 43 and even after permanent coronary occlusion some recovery of function has been seen in the ischemic region.31 32 In the present study, transmural systolic function recovered partially by 12 weeks after reperfusion, but the degree of recovery of systolic function in each strain component was not uniform; the circumferential and radial strains showed gradually progressive recovery but wall thickening in the inner wall remained depressed compared with control. Longitudinal strain never recovered and at 12 weeks was essentially akinetic across the wall. The loss of longitudinal shortening in the inner wall implies a loss of fiber shortening, which in turn probably reduces the normal longitudinal-radial shear. Because this shearing deformation contributes to normal wall thickening,44 its reduction may be responsible for decreased wall thickening in the inner wall. Fibrosis and scarring may constrain myocardial sheets from moving relative to one another,45 causing this abnormal shearing. In combination with loss of fiber shortening, this abnormal shear may provide an explanation for the severe loss of endocardial wall thickening and regional function.
Limitations
Technical limitations of three-dimensional strain
analysis
have been discussed previously.15 17 46
The resolution of
the radiographic technique is one of the potentially limiting factors.
The spatial resolution for identifying the marker centroids is about
0.2 mm in our system. With tissue shrinkage, the beads, particularly in
the subendocardium, would be moving closer together, and overlapping of
beads could make their identification difficult. To minimize this cause
of error, we carefully chose the best position in both
anterior-posterior and lateral projections in which all beads were
identifiable.
The finite element method was used to calculate transmural distributions of finite deformation because transmural myocardial deformation is not homogeneous and ischemia might be expected to have a significant influence on the strain variation within a tetrahedron. Root-mean–square fitting errors in the fitted coordinates showed ranges in both end-diastolic and end-systolic configurations similar to those reported in previous studies,17 indicating that fitting errors in this study were at an acceptable level and that transmural distributions of strains were accurately computed from the coordinate data using the finite element method. When determining remodeling strains, we compared diastolic configurations between the control state and the subsequent studies at matched end-diastolic pressures in each animal to avoid apparent remodeling that was actually due to changes in loading conditions alone. It is generally accepted that the ischemic region usually becomes more stiff with scar formation and that regional diastolic properties after infarction might be different from control. However, we did not measure absolute left ventricular volumes, and how much the diastolic pressure–volume relation was changed with ischemia in our study is unknown.
|
Acknowledgments |
|---|
This study was supported by National Institutes of Health research grant HL-17682, Ischemic Heart Disease Specialized Center of Research (SCOR), awarded by the National Heart, Lung, and Blood Institute. We thank Denice J. Brannigan, Farid Abdel-Wahhab, Barry Peters, and Richard Pavalec for their valuable technical assistance. We also thank Kazuya Murata and Gregory Eising for assistance with surgery and Noa Katrin and Cheryl Bugsch for the preparation of the manuscript.
Received August 24, 1994; accepted October 9, 1994.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Ross J Jr. Mechanical consequences of regional myocardial ischemia. In: Fozzard HA, Haber A, Jennings RB, Katz AM, Morgan HE, eds. The Heart and Cardiovascular System: Scientific Foundations. 2nd ed. New York, NY: Raven Press Publishers; 1991;2:1997-2020.
2. Theroux P, Ross J Jr, Franklin D, Kemper WS, Sasayama S. Coronary arterial reperfusion, III: early and late effects on regional myocardial function and dimension. Am J Cardiol. 1976;38:599-606. [Medline] [Order article via Infotrieve]
3.
Bush LR, Buja M, Samowitz W, Rude RE, Wathen M, Tilton GD,
Willerson JT. Recovery of left ventricular segmental function after
long-term reperfusion following temporary coronary occlusion in
conscious dogs: comparison of 2- and 4-hour occlusions.
Circ Res. 1983;53:248-263.
4.
Lavallee M, Cox D, Patrick TA, Vatner SF. Salvage of
myocardial function by coronary artery reperfusion 1, 2, and 3 hours
after occlusion in conscious dogs. Circ Res. 1983;53:235-247.
5. Kloner RA, Ellis SG, Lange R, Braunwald E. Studies of experimental coronary artery reperfusion: effect on infarct size, myocardial function, biochemistry, ultrastructure, and microvascular damage. Circulation. 1983;68(suppl I):I-8-I-15.
6.
Ono S, Bhargava V, Ono S, Mao L, Hagan G, Rockman HA, Ross J
Jr. In vivo assessment of left ventricular remodeling after myocardial
infarction by digital video-contrast angiography in the rat.
Cardiovasc Res. 1994;28:349-357.
7.
Hochman JS, Chao H. Limitation of myocardial infarct
expansion by reperfusion independent of myocardial salvage.
Circulation. 1987;75:299-306.
8. ISAM Study Group. A prospective trial of intravenous streptokinase in acute myocardial infarction (ISAM). N Engl J Med. 1986;314:1465-1471. [Abstract]
9.
Martin GV, Sheehan FH, Stadius M, Maynard C, Davis KB,
Ritchie JL, Kennedy JW. Intravenous streptokinase for acute myocardial
infarction: effects on global and regional systolic function.
Circulation. 1988;78:258-266.
10. White HD, Norris RM, Brown MA, Takayama M, Maslowski A, Bass NM, Ormiston JA, Whitlock T. Effect of intravenous streptokinase on left ventricular function and early survival after acute myocardial infarction. N Engl J Med. 1987;317:850-855. [Abstract]
11. Van de Werf F, Arnold AER. Intravenous tissue plasminogen activator and size of infarct, left ventricular function, and survival in acute myocardial infarction. Br Med J. 1988;197:1374-1379.
12. Ginks WR, Sybers HD, Maroko PR, Covell JW, Sobel BE, Ross J Jr. Coronary artery reperfusion, II: reduction of myocardial infarct size at one week after the coronary occlusion. J Clin Invest. 1972;51:2717-2723.
13. Reimer KA, Jennings RB. The `wave front phenomenon' of myocardial ischemic cell death, II: transmural progression of necrosis within the framework of ischemic bed size (myocardium at risk) and collateral flow. Lab Invest. 1979;40:633-644. [Medline] [Order article via Infotrieve]
14.
Kambayashi M, Miura T, Oh BH, Murata K, Rockman HA, Parra G,
Ross J Jr. Myocardial cell hypertrophy with reperfusion in dogs.
Circulation. 1992;86:1935-1944.
15.
Waldman LK, Fung YC, Covell JW. Transmural myocardial
deformation in the canine left ventricle: normal in vivo
three-dimensional finite strains. Circ Res. 1985;57:152-163.
16. Meier GD, Ziskin MC, Santamer WP, Bo AA. Kinematics of the locating heart. IEEE Trans Biomed Eng. 1980;27:319-329. [Medline] [Order article via Infotrieve]
17. McCulloch AD, Omens JH. Non-homogeneous analysis of three-dimensional transmural finite deformation in canine ventricular myocardium. J Biomech. 1991;24:1-10. [Medline] [Order article via Infotrieve]
18. Holmes JW, Yamashita H, Waldman LK, Covell JW. Scar remodeling and transmural deformation following infarction in the pig. Circulation. In press.
19. Kloner RA, Ganote CE, Whalen DA, Jennings RB. Effect of a transient period of ischemia on myocardial cells, II: fine structure during the first few minutes of reflow. Am J Pathol. 1974;74:399-422. [Medline] [Order article via Infotrieve]
20.
Kloner RA, DeBoer LWV, Darsee JR, Ingwall JS, Braunwald E.
Recovery from prolonged abnormalities of canine myocardium salvaged
from ischemic necrosis by coronary reperfusion. Proc Natl Acad
Sci U S A. 1981;78:7152-7156.
21. Shen AC, Jennings RB. Myocardial calcium and magnesium in acute ischemic injury. Am J Pathol. 1972;67:417-440. [Medline] [Order article via Infotrieve]
22. Whalen DA, Hamilton DG, Ganote CE, Jennings RB. Effect of a transient period of ischemia on myocardial cells, I: effects on cell volume regulation. Am J Pathol. 1974;74:381-398. [Medline] [Order article via Infotrieve]
23. Willerson JT, Scales F, Mukherjee A, Platt M, Templeton GH, Fink GS, Buja LM. Abnormal myocardial fluid retention as an early manifestation of ischemic injury. Am J Pathol. 1977;87:159-188. [Abstract]
24.
Jennings RB, Schaper J, Hill ML, Steenbergen C Jr, Reimer KA.
Effect of reperfusion late in the phase of reversible ischemic injury:
changes in cell volume, electrolytes, metabolites, and ultrastructure.
Circ Res. 1985;56:262-278.
25.
Gallagher KP, Matsuzaki M, Koziol JA, Kemper WS, Ross J Jr.
Regional myocardial perfusion and wall thickening during ischemia in
conscious dogs. Am J Physiol. 1984;247:H727-H738.
26.
Bolli R, Patel BS, Hartley CJ, Thornby JI, Jeroudi MO, Roberts
R. Nonuniform transmural recovery of contractile function in stunned
myocardium. Am J Physiol. 1989;257:H375-H385.
27.
Villarreal FJ, Lew WYW, Waldman LK, Covell JW. Transmural
myocardial deformation in the ischemic canine left ventricle.
Circ Res. 1991;68:368-381.
28.
May-Newman K, Omens JH, Pavelec RS, McCulloch AD.
Three-dimensional transmural mechanical interaction between the
coronary vasculature and passive myocardium in the dog.
Circ Res. 1994;74:1166-1178.
29. Sato S, Ashraf M, Millard RW, Fujiwara H, Schwartz A. Connective tissue changes in early ischemia on porcine myocardium: an ultrastructure study. J Mol Cell Cardiol. 1983;15:261-267. [Medline] [Order article via Infotrieve]
30. Zhao M, Zhang H, Robinson TF, Factor SM, Sonnenblick EH, Eng C. Profound structural alterations of the extracellular collagen matrix in postischemic dysfunction ('stunned') but viable myocardium. J Am Coll Cardiol. 1987;10:1322-1334. [Abstract]
31.
Theroux P, Ross J Jr, Franklin D, Covell JW, Bloor CM,
Sasayama S. Regional myocardial function and dimensions early and late
after myocardial infarction in the unanesthetized dog.
Circ Res. 1977;40:158-165.
32. Sasayama S, Gallagher KP, Kemper WS, Franklin D, Ross J Jr. Regional left ventricular wall thickness early and late after coronary occlusion in the conscious dog. Am J Physiol. 1981;240:H293-H299.
33. Anversa P, Loud AV, Levicky V, Guideri G. Left ventricular failure induced by myocardial infarction, II: tissue morphometry. Am J Physiol. 1985;248:H883-H889.
34.
Olivetti G, Capasso JM, Meggs LG, Sonnenblick EH, Anversa P.
Cellular basis of chronic ventricular remodeling after myocardial
infarction. Circ Res. 1991;68:856-869.
35.
Omens JH, Covell JW. Transmural distribution of myocardial
tissue growth induced by volume-overload hypertrophy in the dog.
Circulation. 1991;84:1235-1245.
36.
Gallagher KP, Osakada G, Hess OM, Koziol JA, Kemper WS, Ross J
Jr. Subepicardial segmental function during coronary stenosis and the
role of myocardial fiber orientation. Circ Res. 1982;50:352-359.
37. Gallagher KP, Osaka G, Matsuzaki M, Miller M, Kemper WS, Ross J Jr. Nonuniformity of inner and outer systolic wall thickening in conscious dogs. Am J Physiol. 1985;259:H241-H248.
38.
LeWinter MM, Kent RS, Kroener JM, Carew TE, Covell JW.
Regional differences in myocardial performance in the left ventricle of
the dog. Circ Res. 1975;37:191-199.
39.
Weintraub WS, Hattori S, Agawal JB, Boden-Heimer VS, Banka VS,
Helfant RH. The relationship between myocardial blood flow and
contraction by myocardial layer in the canine left ventricle during
ischemia. Circ Res. 1981;48:430-438.
40. Prinzen FW, Arts T, Van der Vusse GJ, Reneman RS. Gradients in fiber shortening and metabolism across the ischemic left ventricular wall. Am J Physiol. 1986;250:H255-H264.
41. Ellis SG, Henschke CI, Sandor T, Wynne J, Braunwald E, Kloner RA. Time course of functional and biochemical recovery of myocardium salvaged by reperfusion. J Am Coll Cardiol. 1983; 1:1047-1055.
42.
Baugman KL, Maroko PR, Vatner SF. Effects of coronary artery
reperfusion on myocardial infarct size and survival in conscious dogs.
Circulation. 1981;63:317-323.
43.
Bolli R, Zhu WX, Thornby JI, O'Neill PG, Roberts R. Time
course and determinants of recovery of function after reversible
ischemia in conscious dogs. Am J Physiol. 1988;254:H102-H114.
44. LeGrice IJ, Takayama Y, Covell JW. Transverse shear along myocardial cleavage planes provides a mechanism for normal systolic wall thickening. FASEB J. 1993;88:A3392. Abstract.
45. LeGrice IJ. A Finite Element Model of Myocardial Structure: Implications for Electrical Activation in the Heart. Auckland, New Zealand: University of Auckland; 1992. Thesis.
46. Waldman LK, Nosan D, Villarreal FJ, Covell JW. Relation between transmural deformation and local myofiber direction. Circ Res. 1988;252:550-562.
This article has been cited by other articles:
 |
|
 |
|
  M. Becker, C. Ocklenburg, E. Altiok, A. Futing, J. Balzer, G. Krombach, M. Lysyansky, H. Kuhl, R. Krings, M. Kelm, et al. Impact of infarct transmurality on layer-specific impairment of myocardial function: a myocardial deformation imaging study Eur. Heart J., June 2, 2009; 30(12): 1467 - 1476. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Gupta and K. J. Grande-Allen Effects of static and cyclic loading in regulating extracellular matrix synthesis by cardiovascular cells Cardiovasc Res, December 1, 2006; 72(3): 375 - 383. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Lindahl, R. C. Chambers, J. Papakrivopoulou, S. J. Dawson, M. C. Jacobsen, J. E. Bishop, and G. J. Laurent Activation of Fibroblast Procollagen alpha 1(I) Transcription by Mechanical Strain Is Transforming Growth Factor-beta -dependent and Involves Increased Binding of CCAAT-binding Factor (CBF/NF-Y) at the Proximal Promoter J. Biol. Chem., February 15, 2002; 277(8): 6153 - 6161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Sinusas, X. Papademetris, R. T. Constable, D. P. Dione, M. D. Slade, P. Shi, and J. S. Duncan Quantification of 3-D regional myocardial deformation: shape-based analysis of magnetic resonance images Am J Physiol Heart Circ Physiol, August 1, 2001; 281(2): H698 - H714. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. B. Hillenbrand, R. J. Kim, M. A. Parker, D. S. Fieno, and R. M. Judd Early Assessment of Myocardial Salvage by Contrast-Enhanced Magnetic Resonance Imaging Circulation, October 3, 2000; 102(14): 1678 - 1683. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Bishop and G. Lindahl Regulation of cardiovascular collagen synthesis by mechanical load Cardiovasc Res, April 1, 1999; 42(1): 27 - 44. [Full Text] [PDF]
|
|
|
|
|
|
P. Croisille, C. C. Moore, R. M. Judd, J. A. C. Lima, M. Arai, E. R. McVeigh, L. C. Becker, and E. A. Zerhouni Differentiation of Viable and Nonviable Myocardium by the Use of Three-Dimensional Tagged MRI in 2-Day-Old Reperfused Canine Infarcts Circulation, January 19, 1999; 99(2): 284 - 291. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||