(Circulation. 1995;92:926-934.)
© 1995 American Heart Association, Inc.
Articles |
Impact of Left Ventricular Unloading After Late Reperfusion of Canine Anterior Myocardial Infarction on Remodeling and Function Using Isosorbide-5-Mononitrate
From the Cardiology Division of the Department of Medicine, University of Alberta, Edmonton, Canada.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background Late reperfusion during acute myocardial infarction results in delayed recovery of ventricular function and less remodeling, whereas ventricular unloading with nitrates improves function and attenuates remodeling. Whether late reperfusion combined with prolonged unloading with isosorbide-5-mononitrate (ISMN) might produce greater functional recovery and less remodeling than late reperfusion alone is not known.
Methods and Results In vivo left ventricular function
and topography (echocardiograms), postmortem topography (planimetry),
and collagen (hydroxyproline) were measured in dogs that were
randomized to reperfusion 2 hours after left anterior descending
coronary artery ligation, and ISMN (n=12) or placebo (n=12) was
given as 25 mg IV over 4 hours followed by 50 mg PO QID for 6 weeks.
Compared with placebo, the ISMN group had similar heart rate but lower
left atrial pressure, mean arterial pressure, and
rate-pressure products. Although in vivo baseline remodeling and
functional parameters were similar in the two groups, by 6
weeks the ISMN group had smaller (P
.05) infarct and
noninfarct segment lengths, ventricular volumes, and mass;
less (P<.001) asynergy; and greater (P<.001)
ejection fraction. More important, by 2 days, ejection fraction was
18% greater (P<.025) and asynergy 26% less
(P<.05) with ISMN. At 6 weeks, ISMN showed less
(P.05) scar size, scar collagen, cavity dilation,
noninfarct wall thickness, and apical bulging than placebo. In another
4 dogs, acute ISMN produced less improvement in function and remodeling
than prolonged ISMN.
Conclusions Late reperfusion of acute anterior myocardial infarction combined with prolonged ISMN unloading results in greater and earlier recovery of ventricular function and less remodeling than late reperfusion alone.
Key Words: remodeling • reperfusion • hypertrophy • proteins
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Late coronary artery reperfusion in acute myocardial infarction is not consistently associated with prompt improvement of regional left ventricular systolic function because of myocardial stunning and reperfusion injury.1 Thus, reperfusion after 2 hours of coronary occlusion in dogs results in partial recovery of regional function over 2 to 4 weeks.2 3 4 Late reperfusion is also associated with a thicker infarct wall and less expansion5 and with hypertrophy of spared myocardium in the infarct wall,6 which might contribute to the recovery of function. Short-term left ventricular unloading with intravenous nitrate infusion in nonreperfused experimental or clinical infarction within 6 or more hours of coronary occlusion produces prompt improvement of ventricular function,7 8 improved myocardial perfusion,9 10 decreased infarct size,7 8 9 10 11 decreased infarct expansion,8 12 and decreased ventricular dilatation.8 Long-term unloading with nitrate to decrease left ventricular preload, afterload, chamber size, and wall stress also limits progressive postinfarction remodeling.13 14 15 16 Recent studies suggest that prolonged therapy with isosorbide-5-mononitrate (ISMN) is associated with less tolerance in angina patients,17 18 fewer arrhythmias in patients after infarction,19 and less postinfarction "hypertrophy" in dogs.20 Short-term intravenous ISMN also proved safe and improved hemodynamics in patients with nonreperfused acute infarction.21
The aim of this study was to determine whether reperfusion after 2 hours of coronary occlusion combined with prolonged left ventricular unloading with ISMN in the dog results in more prompt and greater recruitment of regional ventricular function and limitation of remodeling during postinfarct healing than reperfusion alone.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Experimental Preparation
The experiments were approved by the institutional animal welfare committee and conformed to the guiding principles of the American Physiological Society. Thirty healthy mongrel dogs (16 to 29 kg) of either sex were randomly allocated to ISMN (n=15) and placebo (n=15) groups and chronically instrumented through a left lateral thoracotomy under general anesthesia (sodium pentobarbital, 30 mg/kg IV), as described previously.10 12 16 Indwelling catheters for serial hemodynamics were inserted into the external jugular vein, internal carotid artery, and left atrium and filled with heparinized saline, and their ends were exteriorized behind the neck. The left anterior descending coronary artery was tied between the first and second diagonal branches with a silk ligature. Metal beads were sutured onto the anterior, lateral, and posterior epicardial surfaces of the mid left ventricular wall for consistent echocardiographic imaging. Two hours after ligation, reperfusion was established by removal of the ligature. One dog in each group died at the time of reperfusion, so that 28 surviving dogs went on to receive ISMN (n=14) or placebo (n=14) therapy. Beginning immediately after reperfusion, these dogs were given either intravenous infusions of ISMN, 25 mg in 25 mL solution over 4 hours (6.25 mg/h), or matching placebo in 25 mL solution over 4 hours via a Harvard pump. After the pericardium and chest were closed, penicillin (1 million units) and streptomycin (1 g) were given intramuscularly. On the following day, the ISMN group was started on 50-mg capsules QD for 6 weeks, and the placebo group was given matching placebo capsules QD for 6 weeks. The dogs had free access to fluids and did not receive treatment for heart failure. At 6 weeks, the 24 surviving dogs were anesthetized, and the hearts were arrested in diastole with an overdose of intravenous potassium chloride, excised, washed in normal saline solution, and weighed.
Measurements During Healing
As described
previously,8 16 serial two-dimensional
echocardiograms (Toshiba SSH-65A; 3.5-MHz transducer, 0.5-in VHS
videotape), ECGs (Gould pen recorder), and
hemodynamics (Statham P23Db for left atrial and
arterial pressures) were recorded in conscious dogs
standing in a jacket before therapy, at 2 days, and weekly during
therapy and again 24 hours after therapy was stopped. The indwelling
catheters were flushed every 3 to 4 days with heparinized saline to
maintain patency. Hemodynamics and ECGs were also
recorded under anesthesia before and after surgery.
Echocardiographic views8 16 included the
parasternal long-axis views, five short-axis views at mitral, chordal,
midpapillary, low papillary, and apical levels, and apical four- and
two-chamber views.
Postmortem Measurement of Scar Size, Geometry, and
Collagen
As described previously,16 risk region was
measured
on postmortem coronary arteriograms recorded on whole-heart
and transverse-section radiographs (five sections 1 to 1.5 cm thick).
Outlines of weighed left ventricular rings, risk regions,
and infarct scars made on plastic overlays were planimetered
(Hewlett-Packard 9835A computer and 9874A digitizer interfaced with a
VAX 750 computer) for derivation of infarct size and topographic
parameters, including "thinning" ratio (average
thickness of infarcted wall to average thickness of the normal wall)
and "expansion" index (ratio of endocardial lengths of
infarct-containing to non–infarct-containing segments demarcated by
papillary muscles), and average topographic short-axis maps were made
for each group. Epicardial and endocardial left ventricular
contours from whole-heart radiographs were digitized to measure the
area and depth of the apical bulge and construct average topographic
long-axis maps. Histopathological studies for infarction and collagen
were done on a 5-mm slice from the ring in the middle of the infarct
zone, and triplicate 5-µm sections were stained with hematoxylin and
eosin, Mallory's stain, or Masson's trichrome. Myocardial
hydroxyproline (milligrams per gram dry tissue weight), a marker for
collagen, was measured in transmural samples (100 to 200 mg) from the
center and border regions of the infarct scar and center of the
nonoccluded bed.
Analysis of Echocardiograms
As described
previously,16 22 23 coded
echocardiograms were analyzed double-blind on video playback by
two independent observers (B.I.J. and M.I.K.) for in vivo functional
and topographic parameters. Briefly, endocardial and
epicardial outlines of the left ventricular images at end
diastole and end systole were traced with a light pen
(Diasonics CardioRevue Center) and copied on plastic overlays. Anatomic
landmarks such as papillary muscles were indicated on the tracings.
Asynergy, defined as akinesis (no systolic inward motion and
thickening), dyskinesis (systolic outward motion and thinning), or both
were marked on each endocardial diastolic outline.
Circumferential extents on each short-axis outline were then digitized
(Hewlett-Packard 9878A and 9835A) and used to compute total endocardial
surface area of asynergy. Outlines from five short-axis and two
long-axis views were used to compute volumes by means of the modified
Simpson rule. Global ejection fraction was calculated as
(end-diastolic volume minus end-systolic
volume)/end-diastolic volume. Interobserver error was <5%
in marking asynergy, segment length, wall thickness, and areas of
outlines, in agreement with previous
studies.16 22 23
Topographic measurements were made on end-diastolic
outlines of papillary short-axis images, and expansion index (ratio of
the lengths of the asynergy-containing and the non–asynergy-containing
segments) and thinning ratio (ratio of the average thicknesses of the
asynergic and nonasynergic zones) were computed. Regional area ejection
fraction was calculated at the same level as (end-diastolic
area minus end-systolic area)/end-diastolic area. Left
ventricular aneurysm was defined as the presence of
diastolic bulge with further bulging and thinning in
systole. Left ventricular mass was calculated from the
volume of myocardium (difference in volumes of epicardial
and endocardial shells at end diastole) multiplied by an
assumed specific gravity of 1.05 g/mL. Echogenicity in the asynergic
zone was scored semiquantitatively by consensus: none (0), mild (1),
moderate (2), or marked (3).
Acute ISMN and Chronic Placebo Groups
To verify whether
chronic nitrate was required, acute
intravenous ISMN plus chronic placebo was given to another
eight dogs, and the same in vivo and postmortem measurements were made
as in the randomized dogs.
Statistics
Data at different steps were coded and analyzed in
blind
fashion. The statistical tests used were ANOVA for the significance of
difference within and between groups, 
2 test for
the significance of difference in event frequency between groups, and
repeated-measures ANOVA for comparing serial data within groups.
Results are presented as mean±SEM. Statistical significance
was set at P<.05.
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Of the 28 dogs that were randomized to therapy, the protocols were completed in 24, 12 in each group. Four dogs (2 placebo, 2 ISMN) that died in their cages within 1 week, presumably from arrhythmias, were excluded from analysis. The mean interval between occlusion and death was 42.7 days, with no difference between the groups.
Effect on Hemodynamics
The hemodynamic results for the 24
time intervals
over 6 weeks are summarized in Fig 1
. Between 2 days and
6 weeks, recordings were made 4 hours (range, 3 to 5 hours)
after the 8 AM dose. Heart rate, mean left atrial
pressure, and mean left arterial pressure were similar
before surgery and during the 2 hours of occlusion in the two groups.
In both groups, heart rate increased after occlusion, remained
unchanged over the acute phase of reperfusion, decreased
(P.005) by 2 days, and showed no further change over the 6
weeks. Left atrial pressure increased after occlusion
(P<.001) in both groups and decreased over the first 5
hours of reperfusion (P.05). However, atrial pressure
dropped to lower levels with ISMN and remained persistently depressed
between 2 days and 6 weeks, whereas that with placebo increased between
2 days and 6 weeks. Mean arterial pressures were similar in
the two groups after occlusion but were lower with ISMN over most of
the 6 weeks. Compared with placebo, mean arterial pressure
with ISMN was 8 mm Hg, or 7.5% less, at 15 minutes after reperfusion;
10 mm Hg, or 9.4% less, at 5 hours; and 14 mm Hg, or 14.4% less, at
6 weeks. The double product of heart rate and mean blood pressure
(in beats per minutexmm Hgx102) between 15 minutes and
6
weeks after reperfusion did not show a statistically significant
decrease with placebo (151±11 versus 126±15, P<.1)
but
decreased significantly with ISMN (138±13 versus 103±10,
P<.025). Furthermore, the "triple" product of
heart rate, mean blood pressure, and mean left atrial pressure (in
beats per minutexmm Hg2x103) between 15
minutes and 6 weeks after reperfusion showed no change with placebo
(144±16 versus 187±27, P=NS) but decreased with
ISMN
(133±23 versus 98±15, P<.005) and was lower with
ISMN
than placebo at 6 weeks (98±15 versus 187±27,
P<.005).
Importantly, blood pressure and left atrial pressure increased slightly
24 hours after ISMN was stopped (P.05) but not placebo
(Fig 1).
|
In Vivo Changes in Ventricular Wall
Stretch
Serial echocardiograms at the papillary level from awake dogs
in
ISMN and placebo groups are shown in Fig 2. The 6-week
echocardiographic data in Figs 2 through
5 refer to
those recorded 24 hours after ISMN or placebo was stopped. In Fig
3A through 3C, there was less stretching of the
infarcted wall over the 6 weeks after reperfusion in the ISMN than in
the placebo group. Thus, anterior segment lengths in ISMN and placebo
groups were similar at the preocclusion baseline (9.64 versus 9.59 cm,
P=NS) but shorter with ISMN by 2 days (8.7±0.2 versus
10.2±0.2 cm, P<.0005) and remained shorter at 6 weeks
(9.1±0.2 versus 10.5±0.3 cm, P<.0025). The posterior
segment lengths in ISMN and placebo groups were similar at baseline
(4.2 versus 4.5 cm, P=NS) and did not change significantly
by 6 weeks with placebo (4.6 versus 4.5 cm, P=NS) or ISMN
(4.2 versus 4.2 cm, P=NS). The expansion index was smaller
with ISMN than placebo (P.05), the respective values being
2.2 versus 2.4 at 1 week, 2.3 versus 2.5 at 3 weeks, and 2.2 versus 2.4
at 6 weeks.
|
|
|
|
In Vivo Changes in Ventricular Wall
Thickness
Wall thicknesses over the 6 weeks after reperfusion were
less with
ISMN than with placebo (Fig 3D
through 3F). Anterior wall
thicknesses
in ISMN and placebo groups were similar at the preocclusion baseline
(10.0 versus 10.1 mm, P=NS) but were smaller with ISMN over
the 6 weeks, values at 2 days, 3 weeks, and 6 weeks being 13.1±0.2
versus 14.7±0.2 mm (P<.0005), 11.8 versus 13.4 mm
(P<.0005), and 10.8 versus 11.6 mm (P<.025),
respectively. Posterior wall thicknesses were similar in the two groups
at the preocclusion baseline (10.1 versus 10.0 mm, P=NS) and
increased over the 6 weeks with placebo (11.9 versus 10.1 mm,
P<.0005) but remained unchanged with ISMN (10.2 versus 10.0
mm, P=NS), so that the thickness at 6 weeks was smaller with
ISMN (10.2±0.1 versus 11.9±0.1 mm, P<.0005). There
was no
overall difference in thinning ratio over the 6 weeks in the two
groups. However, the ratios at 2 days were greater than baseline with
both placebo (1.43±0.03 versus 1.0±0, P<.0005) and
ISMN
(1.32±0.02 versus 1.0±0, P<.0005) and declined over
the
subsequent 6 weeks in both groups.
In Vivo Changes in Ventricular Area and
Volume
The end-diastolic and end-systolic areas after
reperfusion were slightly smaller with ISMN than placebo (Fig
4A and 4B), but the overall difference did not
achieve
statistical significance (P=NS). The volumes (Fig
4C and 4D)
at the preocclusion baseline were similar for ISMN and placebo groups
at both end diastole (71 versus 78 mL, P=NS) and
end systole (26 versus 30 mL, P=NS). However,
diastolic volume over the 6 weeks was smaller with ISMN,
values at 3 and 6 weeks being 64±3 versus 73±3 mL
(P<.025) and 64±3 versus 74±6 mL (P<.1),
respectively. Systolic volume was also smaller with ISMN, values at 2
days, 3 weeks, and 6 weeks being 30±3 versus 39±3 mL
(P<.025), 28±2 versus 37±2 mL (P<.0025),
and
26±2 versus 38±3 mL (P<.0025), respectively. In
addition,
the diastolic endocardial surface areas were similar at the
preocclusion baseline in the two groups (74 versus 76
cm2, P=NS) but slightly smaller with
ISMN, values at 2 days, 3 weeks, and 6 weeks being 68 versus 73
cm2 (P<.1), 69 versus 74 cm2
(P<.1), and 68 versus 78 cm2
(P<.025), respectively.
In Vivo Changes in Ventricular Mass
Left ventricular mass
(Fig 4E) was similar in placebo
and ISMN groups at the preocclusion baseline (111 versus 110 g,
P=NS) and increased significantly by 2 days with placebo
(119 versus 111 g, P<.01) but not with ISMN (113 versus 110
g, P=NS). Although the mass was similar in the two groups at
2 days (113 versus 119 g, P=NS), it was persistently smaller
over the remaining 6 weeks with ISMN, values at 6 weeks being 104
versus 120 g, respectively (P<.005). Importantly, the mass
at 6 weeks was smaller than baseline with ISMN (104 versus 110 g,
P<.01) but greater than baseline with placebo (120 versus
111 g, P<.01).
In Vivo Changes in Regional and Global Dysfunction
The
angular extent of regional asynergy at the papillary level
increased from a baseline value of zero in both groups but was
persistently smaller after reperfusion with ISMN than placebo (Fig
5A). Thus, regional asynergy was lower with ISMN than
placebo at 2 days (21.9±2.5% versus 27.8±1.9%,
P<.05),
3 weeks (19.8±1.9% versus 25.0±1.6%, P<.025), and 6
weeks (17.8±1.4% versus 27.6±2.1%, P<.0025).
Furthermore, total asynergy, as a percentage of endocardial
end-diastolic surface area, increased from a baseline value
of zero in both groups but remained persistently lower after
reperfusion with ISMN than placebo (Fig 5B). Total asynergy
values for
the two groups were, respectively, 13.9±1.5% versus 17.4±0.9%
(P<.05) at 2 days, 12.4±0.6% versus 15.8±0.9%
(P<.05) at 3 weeks, and 11.0±0.8% versus 17.8±1.1%
(P<.0005) at 6 weeks. The area ejection fraction at the
papillary level decreased from the preocclusion baseline in both groups
but was persistently greater after reperfusion with ISMN than placebo
(Fig 5C). Thus, area ejection fraction did not differ
significantly
before occlusion (53±2% versus 59±3%, P=NS) but
was
significantly greater with ISMN at 2 days (42±2% versus 35±3%,
P<.05), 3 weeks (52±2% versus 40±2%,
P<.0005), and 6 weeks (49±2% versus 41±2%,
P<.005). In addition, the volume ejection fraction
decreased from the preocclusion baseline in both groups but was greater
after reperfusion with ISMN than placebo (Fig 5D). Thus, volume
ejection fraction in the groups did not differ before occlusion
(66±3% versus 61±2%, P=NS) but was greater with
ISMN at
2 days (53±2% versus 43±2%, P<.0025), 3 weeks
(57±2%
versus 49±.5%, P<.0025), and 6 weeks (59±1% versus
49±1%, P<.0005).
In Vivo Changes in Echogenicity and Frequency of Aneurysm
and Arrhythmia
Echogenicity of the asynergic zone increased in both
groups after
reperfusion, and the frequency was similar over the 6 weeks. However,
there was a trend for a more rapid resolution of the area of increased
echogenicity in the ISMN group over the 6 weeks. Thus, the scores were
zero at baseline in both groups but were significantly lower
(P<.0005) with ISMN at 2 days (1.83±0.17 versus
2.75±0.13), 3 weeks (1.17±0.21 versus 2.36±0.15), and 6
weeks
(0.33±0.14 versus 1.82±0.18). Left ventricular apical
aneurysm was less frequent with ISMN than placebo, values being
0 of 12 versus 11 of 12 (2=8.4,
P<.05) at 2 weeks and 1 of 12 versus 10 of 12
(2=5.4, P<.05) at 6 weeks. The
frequency of arrhythmia was also slightly less
(P<.1) with ISMN than placebo over the 6 weeks.
Effect of Acute ISMN on In Vivo
Parameters
Four of the eight dogs in the acute ISMN group died: two
within 1
hour of reperfusion and two more within 2 weeks. Data on the four dogs
that survived the 6 weeks are summarized in Table 1. The
hemodynamics for acute ISMN over the first 5 hours
after reperfusion were similar to those for "chronic" ISMN, with
evidence of acute unloading compared with placebo. Thus, values at 30
minutes and 5 hours after reperfusion were, respectively, as follows:
heart rate, 139 versus 139 beats per minute (P=NS); left
atrial pressure, 10 versus 6 mm Hg (P<.01); and mean
arterial pressure, 112 versus 98 mm Hg
(P<.001). However, these parameters for acute
ISMN between 2 days and 6 weeks were similar to those of the placebo
group (Table 1). The effects of acute ISMN on functional and
remodeling
parameters were less than with prolonged ISMN (Table 1).
Thus, when acute ISMN was compared with chronic placebo,
diastolic and systolic volumes were slightly smaller
(P=NS), anterior segment length and expansion index were
significantly smaller (P.05), and ejection fraction was
significantly greater (P.05). When acute ISMN was compared
with chronic ISMN, anterior segment length, expansion index, and
ejection fraction were similar (P=NS), systolic volume and
total asynergy slightly greater (P=NS), regional asynergy
greater (P.05) at 2 days and 6 weeks, and anterior and
posterior wall thickness greater (P.05) at 6 weeks.
|
Effect on Scar Size and Collagen Deposition
At 6 weeks,
infarct scar size as percent risk and left
ventricular mass were smaller (P<.05) in the
ISMN than placebo and acute ISMN groups (Table 2).
However, there was no statistically significant difference
(P=NS) in mass of the noninfarcted left ventricle for ISMN
compared with placebo (85±2 versus 90±3 g) or acute ISMN
(85±2
versus 80±3 g). Scar histology was similar among the groups.
Myocardial hydroxyproline content (in milligrams per gram dry weight)
was not significantly different in the normal zones of ISMN and placebo
groups (3.9±0.5 versus 4.4±0.4, P=NS) but was
slightly
lower with ISMN in the infarct scar center (16.6±2.6 versus
30.3±6.1,
P<.05) and border (8.9±1.4 versus 17.0±4.5,
P<.1). Values in the acute ISMN group did not differ
significantly from those of the chronic placebo group. As reported
previously,10 15 22 23 a
significant gradient in
hydroxyproline content from infarct center to border and normal regions
(P<.01) was found in all three groups.
|
Postmortem Topography
Average long-axis contours at 6 weeks
revealed less apical bulging
and thickness of the apical scar in ISMN than placebo groups (Fig
6) and less bulging with "chronic" than acute ISMN
(Table 3). In addition, average short-axis maps of the
left ventricular transverse sections at 6 weeks showed
smaller infarcts, greater scar wall thickness, less noninfarct wall
thickness, and less cavity area with chronic ISMN than placebo (Table
3). The beneficial effect was smaller with acute than chronic
ISMN.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Since myocardial reperfusion in the clinical setting is often achieved late (>6 hours)1 24 and late reperfusion limits remodeling even if myocardium is not salvaged,5 25 26 adjunctive therapy to promote recovery of function in late reperfusion would be desirable.2 3 4 6 This study represents the first systematic characterization of in vivo changes in left ventricular geometry and function during healing over 6 weeks after late reperfusion of anterior infarction and the effect of prolonged unloading beginning with the onset of reperfusion and continued throughout healing.
There were three new major findings. First, remodeling after late
reperfusion is associated with infarct wall thickening and minimal
infarct expansion in the first 2 days, gradual resolution of infarct
wall thickening over the next 6 weeks, progressive expansion and
hypertrophy of the noninfarcted wall, and some cavity
dilation. Second, application of early and prolonged left
ventricular unloading attenuated these progressive
remodeling changes. Thus, comparing the ISMN with the placebo group at
6 weeks, the infarct wall thickness was 7% smaller
(P<.05), the infarct-containing segment was 15% shorter
(P<.005), the non–infarct-containing segment was 7%
shorter (P=NS), end-diastolic volume was 16%
smaller, end-systolic volume was 43% smaller (P<.025), the
noninfarct wall thickness was 15% smaller (P<.0025), left
ventricular mass was 15% smaller (P<.0025),
asynergy was 62% lower (P<.001), and ejection fraction was
17% greater (P<.001). Although infarct wall thickness at 6
weeks was smaller with ISMN than placebo, it was still greater than at
baseline (P.05) in both ISMN (10.8 versus 10.0 mm) and
placebo (11.6 versus 10.1 mm) groups. Third, late reperfusion was
associated with incomplete recovery of function by 6 weeks, but
adjunctive ventricular unloading with ISMN resulted in
earlier, greater, and persistent recovery of function.
Merits and Limitations of the Model
Since the widespread
application of reperfusion therapy over the
past decade has increased the number of survivors with small,
predominantly nontransmural myocardial infarction, the dog model seems
appropriate because it bears similarity to the patient who has survived
infarction after reperfusion and has good collateral reserve. The model
is also well suited for studying in vivo changes in
ventricular geometry and function during postinfarct
healing because of the ease in obtaining serial two-dimensional
echocardiographic images.16 22 23 The
reperfused infarcts in this study were smaller than nonreperfused
infarcts, which average about 19% of left ventricular mass
at 1 day and shrink to 10% at 6 weeks in this
model.16 22 23 They also showed less
remodeling compared
with that seen with nonreperfused infarcts in our previous
studies.22 23 Thus, there was confirmation of
preserved
infarct wall thickness in postmortem topographic maps and minimal
apical bulging on postmortem radiographs at 6 weeks in this study,
which contrasts with the marked thinning and bulging of the infarct
wall in nonreperfused infarcts in previous
studies.22 23
In addition, postmortem data confirmed less cavity dilation and less
noninfarct wall hypertrophy with ventricular
unloading in the ISMN group.
Mechanisms
The early increase in infarct wall thickness after
late
reperfusion was presumably due to a combination of edema, cell
swelling, and hemorrhage, so that gradual resolution over 6
weeks might be expected. The infarct wall thicknesses in our placebo
group were similar to those measured with ultrasonic crystals by
Kambayashi et al6 at 2 days (14.7 versus 14.3 mm), 1 week
(13.8 versus 13.7 mm), and 3 weeks (13.0 versus 13.1 mm) after
reperfusion done 1.5 to 2 hours during inferior infarction
in dogs. However, those investigators6 stopped
measurements at 3 weeks. In our study, infarct wall thickness in the
placebo group continued to decrease progressively to 11.6 mm at 6
weeks. Consistent with the trend found by Kambayashi et al,
noninfarct wall thickness in our placebo group increased during
healing, from 10.1 mm at baseline to 11.9 mm at 6 weeks, indicating
hypertrophy. Importantly, left ventricular
unloading with ISMN prevented the increase in noninfarct wall thickness
and left ventricular mass by 6 weeks and led to more rapid
resolution of the infarct wall thickening compared with placebo. This
latter effect might have been due, in part, to vasodilatory effects of
ISMN, such as increased collateral inflow and drainage via
coronary veins and lymphatics, an explanation that has been
postulated for the increased "washout" of creatine kinase
isoenzymes and injurious metabolites after reperfusion. The more rapid
decrease in infarct zone echogenicity in our ISMN group might be a
reflection of this washout, in addition to the slightly smaller infarct
size and less infarct collagen. The fact that unloading resulted in a
smaller increase in ventricular mass suggests that the
trigger for hypertrophy in our placebo group was related to
the greater load, wall stress, and wall stretch. That
hemodynamic load was greater with placebo compared with
ISMN is supported by the facts that mean left atrial pressure (an index
of preload) was greater by 19% at 2 days and 62% at 6 weeks
(P<.01), mean arterial pressure (an index of
afterload) was greater by 6% at 1 week and 14% at 6 weeks
(P<.05), and the triple product (an arbitrary index of
hemodynamic load and myocardial work) was greater by
8% at 2 days and 91% at 6 weeks (P<.005). By virtue of
the Laplace law, the greater expansion of the infarct zone, noninfarct
zone, cavity areas, and endocardial surface areas in the placebo group
would be expected to cause greater wall stretch and stress and greater
myocyte stretch. The latter is known to trigger upregulation of
contractile and noncontractile protein gene
expression.27 28
In this study, the persistent left ventricular dysfunction in the placebo group was most likely due to stunning, reperfusion injury, and incomplete salvage.1 26 Total regional asynergy with placebo remained unchanged over 6 weeks (17% at 2 days, 16% at 3 weeks, and 18% at 6 weeks), but corresponding volume ejection fraction (43% at 2 days, 50% at 3 weeks, and 49% at 6 weeks) improved to 82% of the baseline value by 3 weeks, and infarct wall thickness at 6 weeks was still greater than baseline (11.6 versus 10.0 mm, P<.001). In contrast, recovery of function with ISMN was associated with a suppression of reactive or adaptive hypertrophy and was most likely the result of decreased ventricular load, stress, and stretch. The prompt and persistent improvement in volume ejection fraction with ISMN compared with placebo, by 53% at 2 days (80% of baseline) and 59% at 6 weeks (89% of baseline), suggests that the major mechanism for the improved function is via the sequence of early and prolonged left ventricular unloading, attenuation of early ventricular remodeling, decreased ventricular size, and decreased wall stress, although decreased scar size (from 6% to 3%) may have contributed. Other potentially beneficial effects of ISMN that may have played a role include improvement in collateral flow during healing, increased washout of harmful metabolites of late reperfusion, decreased oxygen free-radical activity, increased delivery of nitric oxide (NO) to the reperfused infarct zone during healing, and preservation of the collagen matrix integrity. Our data with acute ISMN alone suggest that prolonged ISMN is more beneficial.
There are several reasons why ventricular unloading with ISMN might protect the supporting myocardial collagen matrix and contribute to preservation of ventricular geometry and function after late reperfusion. First, since increased regional wall stresses created by regional bulging in systole29 30 and diastole22 23 31 after infarction disrupt the collagen network,22 23 decreased bulging with unloading might prevent disruption. Similarly, decreased chamber dilation with unloading might lessen mechanical disruption of the matrix in noninfarct zones. Second, since ischemia damages the matrix,32 33 anti-ischemic effects of ISMN might alleviate the damage. Third, since chamber dilation causes upregulation of collagenases and other metalloproteinases that disrupt the matrix,34 decreased chamber size with ISMN might downregulate these enzymes and block matrix disruption. Fourth, since reperfusion ischemic injury disrupts collagen matrix and its mechanical coupling function, thereby contributing to myocardial stunning,33 ISMN might prevent these via its unloading, anti-ischemic, and NO-donor effects. Fifth, since metabolites of late reperfusion can potentially damage the matrix, their washout induced by ISMN might be protective. Sixth, since release of NO is impaired after ischemia-reperfusion35 and ISMN is an NO donor, it can potentially increase the level of NO36 and protect both matrix and myocyte. The latter was shown by intracoronary infusion of the cysteine-containing NO donor SPM-5185 after late reperfusion of canine anterior infarction, which resulted in decreased myocardial infarct size, neutrophil-dependent endothelial cell damage, and segmental systolic and diastolic ventricular dysfunction.
The antihypertrophy effect of ISMN might also involve a nonhemodynamic mechanism via inhibition of myocyte and/or interstitium growth.37 38 Although the finding that prolonged ISMN (30 mg BID over 16 weeks) after DC myocardial damage in the dog prevented the increase in ventricular mass in the absence of a decrease in blood pressure or ventricular volume suggested this mechanism, preload was reduced in that study.20 Previous studies12 15 suggest that the finding of slightly higher infarct collagen content in our placebo group is more likely a reflection of the slightly larger infarct size than inhibition of infarct collagen by ISMN, since nitrates did not decrease infarct collagen or damage the noninfarct collagen matrix in nonreperfused infarcts but appeared instead to increase infarct collagen12 and increase mechanical resistance of the infarcted ventricle to distension.15 Infarct collagen might also be higher with placebo because the reperfused infarct wall is thicker, so tissue collapse26 might be less during the time interval when fibroblast activity increases and collagen deposition plateaus,39 40 so that there is more volume for collagen deposition.
Implications
The overall results of this study indicate that
even the small
anterior reperfused infarct is associated with persistent
ventricular dysfunction and progressive remodeling with
ventricular dilation and hypertrophy and that
these effects can be attenuated by prolonged unloading with ISMN.
Although chronic nitrate use, especially in high and/or multiple
regular doses, is limited by development of tolerance in one or more of
its actions, ISMN in this study, given after reperfusion by initial
intravenous infusion and followed by a once-daily dose,
produced effective unloading over 6 weeks. The fact that serial
hemodynamics and echocardiograms were recorded 4
hours after the daily oral medication might explain some of the
findings, since plasma concentration of ISMN peaks around that
time.17 18 21 We did not determine
whether
ventricular unloading by purely mechanical means might have
produced similar benefits, so that the other biological effects of
nitrates (eg, antiplatelet, antithrombotic, spasmolytic,
flow-promoting, antiproliferative, and antineutrophil) might explain
some of the benefits. Several small preliminary clinical studies tested
prolonged nitrate therapy over 6 weeks after infarction with or without
thrombolysis and reported attenuation of
ventricular dilation and dysfunction.13 14
Although results of the ISIS-441 42 and
GISSI-343 trials did not support the routine use of
prolonged nitrates in the populations studied, these were treatment and
mortality trials, not reperfusion trials. It is possible that modest
differences in remodeling and function with small reperfused anterior
infarcts, as seen in our model, might not translate into significant
morbidity and mortality as with large reperfused anterior infarcts.
Conclusions
Prolonged ventricular unloading after late
reperfusion
during anterior myocardial infarction is effective in limiting
ventricular remodeling and produces early and persistent
recovery of left ventricular function in the dog model.
Further studies are needed to test the efficacy of nitrate unloading
after late reperfusion in the clinical setting.
|
Acknowledgments |
|---|
This study was supported in part by grants from the Medical Research Council of Canada and the Canadian Heart and Stroke Foundation, Ottawa, Ontario. This work was performed during the tenure of Dr Jugdutt as scientist of the Alberta Heritage Foundation for Medical Research. We are grateful to Schwarz Pharma AG for the free supply of isosorbide-5-mononitrate and placebo. We also thank Michael Joljart, Ruwan Liyanage, and Vijayan Menon, BSc, for assistance with computing and Catherine Graham for assistance with typing.
|
Footnotes |
|---|
Reprint requests to Dr B.I. Jugdutt, 2C2.43 Walter Mackenzie Health Sciences Centre, Division of Cardiology, Department of Medicine, University of Alberta, Edmonton, Alta, Canada T6G 2R7.
Received December 27, 1994; revision received February 1, 1995; accepted February 20, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Braunwald E. Myocardial reperfusion, limitation of infarct size, reduction of left ventricular dysfunction, and improved survival: should the paradigm be expanded? Circulation. 1989;79:441-442.
2. Théroux P, Ross J Jr, Franklin D, Kemper WS, Sasayama S. Coronary arterial reperfusion, III: early and late effects on regional myocardial function and dimensions in conscious dogs. Am J Cardiol. 1976;38:599-606. [Medline] [Order article via Infotrieve]
3.
Lavallée 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.
4.
Bush LR, Buja LM, 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.
5.
Hochman JS, Choo H. Limitation of myocardial
expansion by reperfusion independent of myocardial salvage.
Circulation. 1987;75:299-306.
6.
Kambayashi M, Miura T, Oh B-H, Murata K, Rockman HA,
Parra G, Ross J Jr. Myocardial cell hypertrophy
after myocardial infarction with reperfusion in dogs.
Circulation. 1992;86:1935-1944.
7.
Jugdutt BI, Sussex BA, Warnica JW, Rossall RE.
Persistent reduction in left ventricular asynergy in
patients with acute myocardial infarction by intravenous
infusion of nitroglycerin.
Circulation. 1983;68:1264-1273.
8.
Jugdutt BI, Warnica JW. Intravenous
nitroglycerin therapy to limit myocardial infarct size,
expansion and complications: effect of timing, dosage and infarct
location. Circulation. 1988;78:906-919.
9.
Jugdutt BI, Becker LC, Hutchins GM, Bulkley BH, Reid
PR, Kallman CH. Effect of intravenous
nitroglycerin on collateral blood flow and infarct size
in the conscious dog. Circulation. 1981;63:17-28.
10.
Jugdutt BI. Myocardial salvage by
intravenous nitroglycerin in conscious
dogs: loss of beneficial effect with marked
nitroglycerin-induced hypotension.
Circulation. 1983;68:673-684.
11.
Flaherty JT, Becker LC, Bulkley BH, Weiss JL,
Gerstenblith G, Kallman CH, Silverman KJ, Wei JY, Pitt B, Weisfeldt ML.
A randomized prospective trial of intravenous
nitroglycerin in patients with acute myocardial
infarction. Circulation. 1983;68:576-588.
12.
Jugdutt BI. Delayed effects of early
infarct-limiting therapies on healing after myocardial
infarction. Circulation. 1985;72:907-914.
13. Michorowski BL, Tymchak WJ, Jugdutt BI. Improved left ventricular function and topography by prolonged nitroglycerin therapy after acute myocardial infarction. Circulation. 1987;76(suppl IV):IV-128. Abstract.
14. Jugdutt BI, Tymchak WJ, Humen DP, Gulamhusein S, Tang SB. Effect of thrombolysis and prolonged captopril and nitroglycerin on infarct size and remodeling in transmural myocardial infarction. J Am Coll Cardiol. 1992;19:205A. Abstract.
15. Jugdutt BI. Effect of nitroglycerin and ibuprofen on left ventricular topography and rupture threshold during healing after myocardial infarction in the dog. Can J Physiol Pharmacol. 1988;66:385-395. [Medline] [Order article via Infotrieve]
16.
Jugdutt BI, Khan MI. Effect of prolonged nitrate
therapy on left ventricular remodeling after canine acute
myocardial infarction. Circulation. 1994;89:2297-2307.
17. Thadani U, Hamilton SF, Olson E, Anderson JL, Prasad R, Voyles W, Doyle R, Kirsten E, Teague SM. Duration of effects and tolerance of slow-release isosorbide-5-mononitrate for angina pectoris. Am J Cardiol. 1987;59:756-762.[Medline] [Order article via Infotrieve]
18. Chrysant SG, Glasser SP, Bittar N, Shahidi FE, Danisa K, Ibrahim R, Watts LE, Garutti RJ, Ferraresi R, Casareto R. Efficacy and safety of extended-release isosorbide mononitrate for stable effort angina pectoris. Am J Cardiol. 1993;72:1249-1256. [Medline] [Order article via Infotrieve]
19. Pipilis A, Flather M, Collins R, Coats A, Conway M, Appleby P, Sleight P. Hemodynamic effects of captopril and isosorbide mononitrate started early in acute myocardial infarction: a randomized placebo-controlled study. J Am Coll Cardiol. 1993;21:73-79. [Abstract]
20. McDonald KM, Francis GS, Matthews J, Hunter D, Cohn JN. Long-term oral nitrate therapy prevents chronic ventricular remodeling in the dog. J Am Coll Cardiol. 1993;21:514-522. [Abstract]
21.
Rezakovi
DE, Goldner V, Batinic Z, Weiss M,
Stalec J, Pavicic L. Intravenous
isosorbide-5-mononitrate in the treatment of acute myocardial
infarction. Am J Cardiol. 1990;65:50J-56J. [Medline]
[Order article via Infotrieve]
22. Jugdutt BI, Khan MI. Impact of infarct transmurality on remodeling and function during healing after anterior myocardial infarction in the dog. Can J Physiol Pharmacol. 1992;70:949-958. [Medline] [Order article via Infotrieve]
23. Jugdutt BI, Schwarz-Michorowski BL, Khan MI. Effect of long-term captopril therapy on left ventricular remodeling and function during healing and canine myocardial infarction. J Am Coll Cardiol. 1992;19:713-721. [Abstract]
24.
Pfeffer MA, Braunwald E. Ventricular
remodeling after myocardial infarction: experimental observations and
clinical implications. Circulation. 1990;81:1161-1172.
25. Hale SL, Kloner RA. Left ventricular topographic alterations in the completely healed rat infarct caused by early and late coronary artery reperfusion. Am Heart J. 1988;116:1508-1513. [Medline] [Order article via Infotrieve]
26.
Boyle MP, Weisman HF. Limitation of infarct
expansion and ventricular remodeling by late reperfusion:
study of time course and mechanism in a rat model.
Circulation. 1993;88:2872-2883.
27.
Schneider MD, Parker TG. Cardiac myocytes as
targets for the action of growth factors.
Circulation. 1990;81:1443-1456.
28. Chien KR, Knowlton KU, Zhu H, Chien S. Regulation of cardiac gene expression during myocardial growth and hypertrophy: molecular studies of an adaptive physiologic response. FASEB J. 1991;5:3037-3046. [Abstract]
29. Tennant R, Wiggers CJ. The effect of coronary occlusion on myocardial contraction. Am J Physiol. 1935;112:351-356.
30. Force T, Kemper A, Leavitt M, Parisi AF. Acute reduction in functional infarct expansion with late coronary reperfusion: assessment with quantitative two-dimensional echocardiography. J Am Coll Cardiol. 1988;11:192-200. [Abstract]
31. Jugdutt BI. Identification of patients prone to infarct expansion by the degree of regional shape distortion on an early two-dimensional echocardiogram after myocardial infarction. Clin Cardiol. 1990;13:28-40.[Medline] [Order article via Infotrieve]
32.
Fujiwara H, Ashraf M, Sato S, Millard R.
Transmural cellular damage and blood flow distribution in early
ischemia in pig heart. Circ Res. 1982;51:683-693.
33. Zhao M, Zhang H, Robinson TF, Factor SM, Sonnenblick EH, Eng C. Profound structural alterations of the extracellular collagen matrix in postischemic dysfunctional ("stunned") but viable myocardium. J Am Coll Cardiol. 1987;10:1322-1334. [Abstract]
34. Armstrong PA, Moe GW, Howard RJ, Grima EA, Cruz TF. Structural remodelling in heart failure: gelatinase induction. Can J Cardiol. 1994;10:214-220. [Medline] [Order article via Infotrieve]
35. Nichols WW, Mehta JL, Donnelly WH, Lawson D, Thompson L, ter Riet M. Reduction in coronary vasodilator reserve following coronary occlusion and reperfusion in anesthetized dog: role of endothelium-derived relaxing factor, myocardial neutrophil infiltration and prostaglandins. J Mol Cell Cardiol. 1988;20:943-954. [Medline] [Order article via Infotrieve]
36.
Lefer DJ, Nakanishi K, Johnston WE, Vinten-Johansen J.
Antineutrophil and myocardial protecting actions of a novel
nitric oxide donor after acute myocardial ischemia and
reperfusion in dogs. Circulation. 1993;88:2337-2350.
37. Garg UC, Hassid A. Nitric oxide-generating vasodilators and 8-bromocyclic guanosine monophosphate inhibit mitogenesis and proliferation of cultured rat vascular smooth muscle cells. J Clin Invest. 1989;83:1774-1777.
38. Itoh H, Pratt RE, Dzau VJ. Atrial natriuretic polypeptide inhibits hypertrophy of vascular smooth muscle cells. J Clin Invest. 1990;86:1690-1697.
39. Jugdutt BI, Amy RWM. Healing after myocardial infarction in the dog: changes in infarct hydroxyproline and topography. J Am Coll Cardiol. 1986;7:91-102. [Abstract]
40. Jugdutt BI. Left ventricular rupture threshold during the healing phase after myocardial infarction in the dog. Can J Physiol Pharmacol. 1987;65:307-316. [Medline] [Order article via Infotrieve]
41. ISIS-4 Collaborative Group. Fourth International Study of Infarct Survival: protocol for a large simple study of the effects of oral mononitrate, of oral captopril, and of intravenous magnesium. Am J Cardiol. 1991;68:87D-100D. [Medline] [Order article via Infotrieve]
42. ISIS Collaborative Group. ISIS-4: randomization of oral isosorbide mononitrate in over 50,000 patients with suspected acute myocardial infarction. Circulation. 1993;88(suppl I):I-394. Abstract.
43. Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico. GISSI-3: effects of lisinopril and transdermal glyceryl trinitrate singly and together on 6-week mortality and ventricular function after acute myocardial infarction. Lancet. 1994;343:1115-1122.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  G. Ertl and S. Frantz Healing after myocardial infarction Cardiovasc Res, April 1, 2005; 66(1): 22 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Ertl and S. Frantz Wound model of myocardial infarction Am J Physiol Heart Circ Physiol, March 1, 2005; 288(3): H981 - H983. [Full Text] [PDF]
|
|
|
|
 |
|
 B. I. Jugdutt and V. Menon Upregulation of Angiotensin II Type 2 Receptor and Limitation of Myocardial Stunning by Angiotensin II Type 1 Receptor Blockers during Reperfused Myocardial Infarction in the Rat Journal of Cardiovascular Pharmacology and Therapeutics, September 1, 2003; 8(3): 217 - 226. [Abstract] [PDF]
|
|
|
|
 |
|
 B. Meyns, J. Stolinski, V. Leunens, E. Verbeken, and W. Flameng Left ventricular support by Catheter-Mountedaxial flow pump reduces infarct size J. Am. Coll. Cardiol., April 2, 2003; 41(7): 1087 - 1095. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. I. Jugdutt and V. Menon Beneficial Effects of Therapy on the Progression of Structural Remodeling During Healing After Reperfused and Nonreperfused Myocardial Infarction: Different Effects on Different Parameters Journal of Cardiovascular Pharmacology and Therapeutics, June 1, 2002; 7(2): 95 - 107. [Abstract] [PDF]
|
|
|
|
|
|
B. I. Jugdutt Monocytosis and adverse left ventricular remodeling after reperfused myocardial infarction J. Am. Coll. Cardiol., January 16, 2002; 39(2): 247 - 250. [Full Text] [PDF]
|
|
|
|
 |
|
 B. I Jugdutt and M. Balghith Enhanced regional AT2-receptor and PKC{varepsilon} expression during cardioprotection induced by AT1-receptor blockade after reperfused myocardial infarction Journal of Renin-Angiotensin-Aldosterone System, June 1, 2001; 2(2): 134 - 140. [Abstract] [PDF]
|
|
|
|
|
|
Yi Xu, V. Menon, and B. I Jugdutt Cardioprotection after angiotensin II type 1 blockade involves angiotensin II type 2 receptor expression and activation of protein kinase C-{varepsilon} in acutely reperfused myocardial infarction in the dog: Effect of UP269-6 and losartan on AT1- and AT2-receptor expression and IP3 receptor and PKC{varepsilon} proteins Journal of Renin-Angiotensin-Aldosterone System, June 1, 2000; 1(2): 184 - 195. [Abstract] [PDF]
|
|
|
|
|
|
B. I. Jugdutt, Yi Xu, M. Balghith, R. Moudgil, and V. Menon Cardioprotection Induced by AT1R Blockade After Reperfused Myocardial Infarction: Association With Regional Increase in AT2R, IP3R and PKC{varepsilon} Proteins and cGMP Journal of Cardiovascular Pharmacology and Therapeutics, January 1, 2000; 5(4): 301 - 311. [Abstract] [PDF]
|
|
|
|
 |
|
 U. Elkayam, J. V. Johnson, A. Shotan, S. Bokhari, A. Solodky, M. Canetti, O. R. Wani, and I. S. Karaalp Double-Blind, Placebo-Controlled Study to Evaluate the Effect of Organic Nitrates in Patients With Chronic Heart Failure Treated With Angiotensin-Converting Enzyme Inhibition Circulation, May 25, 1999; 99(20): 2652 - 2657. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Mahmarian, L. A. Moye, D. A. Chinoy, R. F. Sequeira, G. B. Habib, W. J. Henry, A. Jain, B. R. Chaitman, C. S. W. Weng, H. Morales-Ballejo, et al. Transdermal Nitroglycerin Patch Therapy Improves Left Ventricular Function and Prevents Remodeling After Acute Myocardial Infarction : Results of a Multicenter Prospective Randomized, Double-Blind, Placebo-Controlled Trial Circulation, May 26, 1998; 97(20): 2017 - 2024. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. McDonald, C. Chu, G. Francis, D. Judd, W. Carlyle, C. Toher, K. Hauer, and M. Hartman The effect of delayed reperfusion following infarction in the rat on structural changes in viable myocardium Cardiovasc Res, December 1, 1997; 36(3): 347 - 353. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||