(Circulation. 1995;92:372-380.)
© 1995 American Heart Association, Inc.
Articles |
Left Ventricular Function After Extended Hypothermic Preservation of the Heart Is Dependent on Functional Coronary Capillarity
From The University of Arizona Health Sciences Center, Department of Surgery, and The University of Arizona Heart Center.
Correspondence to Lorraine H. Manciet, PhD, The University of Arizona Health Sciences Center, Department of Surgery, Tucson, AZ 85724.
 |
Abstract |
|---|
 Top Abstract IntroductionMethods Results Discussion References |
|---|
Background A growing body of knowledge has led to the hypothesis that injury to the microcirculation during hypothermic myocardial preservation may result in decreased contractility of hearts upon reperfusion.
Methods and Results To test this hypothesis, we examined the
relationship between no-reflow and left ventricular
function after hypothermic cardiac preservation after reperfusion with
solutions containing dilute whole blood (DWB) or washed red blood cells
(K2RBC). Rat hearts were arrested with high-potassium cardioplegia,
then flushed and stored for 6 hours in low-potassium cardioplegia
at 4°C. Hearts were reperfused at a constant flow rate (4 mL/min)
with K2RBC for 60 minutes (group 1, n=5) or DWB for 7 minutes followed
by 53 minutes of K2RBC (group 2, n=5). Left ventricular
developed pressure (LVDP) was measured with an
intraventricular balloon. Immediately after
functional assessment, hearts were perfused with an india ink solution
to mark flow, then glutaraldehyde. Morphometric
techniques were used to determine the degree of capillary compression
[
(c)], perfused capillary
number per fiber area
[QA(0)P], and perfused capillary surface area
per fiber volume [SV(c,f)P]. Capillaries were
moderately compressed in both groups after reperfusion (group 1,
19±1%; group 2, 20±1%). QA(0)P and
SV(c,f)P were highly correlated with
(c) in hearts reperfused with K2RBC
(r=.92 and
r=.92; P<.01). Although statistically
significant, the correlation was not as strong in DWB-reperfused hearts
(r=.66 and r=.67; P<.05). LVDP was
correlated to QA(0)P and
SV(c,f)P (r=.86 and
r=.87, respectively) for groups 1 and 2.
Conclusions The weaker correlation between capillary perfusion and capillary compression in DWB-reperfused hearts suggests that factors other than compression contribute to no-reflow after hypothermic preservation. Regardless of the composition of the reperfusate, recovery of left ventricular function after hypothermic ischemia is directly related to coronary capillary perfusion upon reperfusion.
Key Words: capillaries • reperfusion • hypothermia
|
Introduction |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Successful preservation of donor hearts for transplantation requires hypothermic arrest and storage. Current protocols are sufficient to protect isolated donor hearts for 4 to 6 hours.1 2 3 4 However, hearts cannot be preserved longer than this time because, although metabolism has been slowed by arrest and hypothermia, alterations take place that compromise functional recovery upon reperfusion.3 4 In addition to the damage sustained during the preservation period, reperfusion of the coronary microvasculature is incomplete upon transplantation. This impaired ability to fully perfuse the microcirculation limits the delivery of oxygen and substrates to the myocardial tissue while reducing the removal of toxic metabolites produced during the preservation period.5 6 7 8 9 10 Consequently, tissue injury that occurs upon reperfusion may exceed that incurred during the ischemic, or preservation, period.8 9 10
The failure of the microvasculature to become fully perfused after periods of regional or global ischemia has been characterized as the no-reflow phenomenon.8 9 The mechanistic basis for the no-reflow phenomenon is complex and has not yet been completely defined, although both intravascular and extravascular mechanisms are suspected. In the heart, microvascular "plugging" by leukocytes and/or platelets11 12 13 14 and mechanical compression of microvessels due to intracellular and/or extracellular edema4 15 16 17 or contracture16 17 18 19 are thought to contribute to no-reflow.
In a study investigating the differential effects of blood components on injury to the coronary microcirculation after normothermic ischemia and reperfusion, Reynolds and McDonagh13 found, in isolated rat hearts, that in the absence of leukocytes, 30% to 35% of surface capillaries failed to perfuse after 30 minutes of global, normothermic ischemia. Perfusion with DWB reduced capillary perfusion a further 30%. The authors concluded that although leukocytes contribute to coronary reperfusion injury, leukocytes alone do not fully account for capillary perfusion deficits after myocardial ischemia.
We recently reported that after 30 minutes of global, normothermic ischemia, increased mean cardiac muscle fiber cross-sectional area consequent to intracellular edema and myocyte contracture was highly related to decreased coronary capillary diameter (measured across the shortest axis of the vessel).17 Additionally, using india ink as a marker of flow, we determined that upon reperfusion with an asanguineous solution, decreases in perfused capillary number and length were correlated with decreases in capillary diameter. While intracellular edema had the greatest overall effect on increased fiber area, fiber shortening accounted for a significant increase in the size of muscle fibers in the subendocardium, where perfusion deficits were most pronounced. We concluded that during myocardial ischemia, capillary compression consequent to increased fiber size was associated with incomplete perfusion with an acellular solution. Moreover, upon whole blood reperfusion, capillary compression will likely exacerbate no-reflow due to mechanical plugging by leukocytes and other blood components.
The purpose of the present study was to evaluate the combined effects of hypothermic preservation and blood reperfusion on microvascular perfusion and left ventricular pump function. After 6 hours of static, hypothermic preservation, isolated rat hearts were reperfused with either DWB or a modified Krebs' solution containing K2RBC. Left ventricular pump function was measured with an intraventricular balloon. After 60 minutes of reperfusion, the hearts were perfused with india ink, then fixed with glutaraldehyde and evaluated functionally and morphometrically to determine (1) the effects of extended hypothermic preservation on left ventricular function and tissue morphology; (2) differences in capillary reperfusion consequent to differences in the composition of the reperfusate (K2RBC versus DWB); and (3) the relationship between microvascular perfusion and left ventricular pump function of the heart after long-term, hypothermic preservation.
|
Methods |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
Experimental Model
Ten male retired breeder Sprague-Dawley rats weighing
500 g were used. Each experimental group contained five rats, as
described below. Animals were fed food and water ad libitum and
received humane care in compliance with the "Principles of
Laboratory Animal Care" formulated by the National Society for
Medical Research and the "Guide for the Care and Use of Laboratory
Animals" prepared by the National Academy of Sciences and published
by National Institutes of Health (NIH publication 80-23, revised
1987).
Isolation of Rat Heart
The hearts were isolated as previously
described by McDonagh et
al.14 20 Briefly, the rats were anesthetized with
sodium pentobarbital (60 mg · kg-1 IP), and a
tracheotomy was performed. The animals were placed on a rodent
respirator (Harvard Apparatus) and ventilated. A median
sternotomy was performed, exposing the heart and great vessels. Loose
ligatures were placed around the ascending aorta and the right
subclavian, right common carotid, and innominate arteries. Heparin (300
U) was injected into the right atrium, and the right subclavian artery
was ligated. A Jelco (No. 20) catheter was inserted into the right
common carotid artery, advanced until the tip reached the aortic arch,
and secured.
The heart was then arrested in a manner consistent with the clinical protocol currently used at the University of Arizona Heart Center for the arrest and storage of human donor hearts before transplantation. Specifically, an apparatus consisting of two syringes, one containing 20 mL high-K+ (30 mmol/L; high-K+ cardioplegia; see "Solutions"), the other 20 mL low-K+ (5 mmol/L; low-K+ cardioplegia; see "Solutions") modified Krebs-bicarbonate solution at 4°C was immediately attached to the catheter, and the high-K+ solution was injected at a pressure of <80 mm Hg. The injection of high-K+ solution was immediately followed by the infusion of the low-K+ solution. After arrest and cooling, the heart was carefully excised, and an intraventricular balloon was inserted into the left ventricle through the left atrium. The balloon catheter was secured with a ligature. The heart was immersed in the low-K+ solution and stored at 4°C for a period of 6 hours.
Myocardial Reperfusion
At the end of the preservation period,
the isolated heart was
suspended for perfusion on a modified Langendorff
apparatus21 via the aortic cannula. Five
hearts (K2RBC group) were continuously perfused at 37°C for 60
minutes with a modified Krebs-bicarbonate solution containing
washed red blood cells14 20 (K2RBC; see
"Solutions"). The second group of hearts (n=5; DWB group)
was
initially reperfused with diluted whole blood (hematocrit, 0.18 to
0.20) in a modified Krebs-bicarbonate solution for 7
minutes14 20 (DWB; see "Solutions").
Reperfusion
with the K2RBC solution was continued for 53 minutes. Total reperfusion
time for both groups was 60 minutes. A Radiometer ABL330 blood gas
analyzer was used to perform blood gas analysis on the
perfusate to control for pH, PO2,
PCO2, and bicarbonate levels.
The intraventricular
balloon was attached to a
pressure transducer and filled with PBS (250 µL). Maximum LVDP,
left ventricular end-diastolic pressure,
and peak ±dP/dt were processed and recorded with a Gould
WindowGraf recorder. The hearts were paced with a Grass stimulator
at a constant rate of 250 beats per minute.14
Throughout the 60-minute Langendorff perfusion period, the volume in the intraventricular balloon was maintained at a volume that, based on previous work with this model,14 would yield a diastolic pressure of 5 mm Hg in a nonischemic heart. At the end of the reperfusion period, hearts were removed from the modified Langendorff apparatus and immediately perfused with india ink solution, then glutaraldehyde17 for morphometric analysis.
Determination of Perfused Capillaries
The percentage of
perfused capillaries was determined by
subsequent perfusion with an india ink solution (see
"Solutions"), as previously described.17 Briefly,
immediately after the 60-minute reperfusion period, hearts were
perfused with the ink solution at a pressure of 80 mm Hg for 1.5
minutes. The perfusion pressure and duration were selected on the basis
of previously reported data that established that this pressure and
duration are adequate to ensure complete perfusion of the capillaries
that are capable of supporting the flow of india ink.22
Tissue Preparation
After perfusion with this india ink
solution, the hearts were
perfused with a glutaraldehyde solution, as previously
described by Poole et al.23 Briefly, the hearts were
perfused with a glutaraldehyde fixative (6.25%
glutaraldehyde solution in 0.1 mol/L sodium cacodylate,
adjusted to 430 mOsm with NaCl; total osmolarity of the fixative, 1100
mOsm; pH 7.4) for 2 to 4 minutes at a pressure of 60 mm Hg, as opposed
to 80 mm Hg, to minimize the washout of the india ink solution in
perfused vessels during perfused fixation of the tissue. After fixation
of the tissue, the heart was immersed in glutaraldehyde
and stored at 4°C.
At the time of processing, as previously
described,24 a
portion of the left ventricular free wall (0.5x0.5 cm) was
taken from the heart and sectioned into Epi and Endo segments (Fig
1
). These segments were cut into longitudinal sections
(1 cm x2 mm x1 mm) and washed three times with 0.1 mol/L
sodium
cacodylate buffer. After the third wash, the tissue sections were
soaked in an osmium solution (1% OsO4 in 0.125 mol/L
sodium cacodylate) for 2 hours. The tissue was then sequentially
dehydrated with 70%, 80%, 95%, and 100% ethanol, followed by 100%
propylene oxide. At this point, the tissue was soaked in propylene
oxide/epoxy resin (1:1) for 2 hours, then held overnight in epoxy
resin. The tissue was then embedded in epoxy resin, oriented such that
both transverse and longitudinal sections could be obtained, and cured
at 60°C for 2 days.
|
Tissue Sectioning
The tissue blocks were sectioned on an LKB
8800 Ultrotome III.
As illustrated in Fig 1
, eight blocks were sectioned from the
subepicardium and subendocardium of each heart such that four
longitudinal and four transverse sections, 1 µm thick, were obtained
and stained with 0.1% toluidine blue solution. The sectioning angles
needed to obtain transverse and longitudinal sections were determined
as previously described in detail by Mathieu-Costello.25
Briefly, consecutive sections of the same block were taken at different
angles with respect to the muscle fiber axis by changing the specimen
holder orientation by 1° and 5° for longitudinal and transverse
sections, respectively (Fig 1). Sections were considered
longitudinal
when a change of ±1° gave sections with longer sarcomeres. Sections
were defined as transverse when a change in the sectioning angle by
±5° in either direction yielded sections with smaller A-band
spacing. Whereas a perfect transverse section through all fibers would
remove all banding patterns, small differences in fiber and myofibril
alignment result in the presence of few bands across the cross section
of most fibers. Such deviations, about 5° from perfect transverse
sectioning, do not measurably affect estimates of fiber
cross-sectional area or capillary numerical
density.26
Morphometric Analysis
Each 1-µm section was projected
from a Zeiss
Photomicroscope III onto a Sony 21-in color monitor (magnification
x2300) with a Sony color video camera (SSC-S20). As many (8 to 15)
nonoverlapping quadrates as possible were systematically subsampled.
Point-counting techniques were performed by use of a square grid
test A 14427 superimposed onto themonitor. All
point counts were collected, stored, and processed with an Apple
IIe computer.
Capillary diameters were measured across the
longest and shortest
principal diameters of capillaries on transverse sections
[(c)l and
(c)s,
respectively]. The degree of capillary compression was calculated as
the percentage of the mean shortest principal capillary diameter in
relation to the mean greatest principal capillary diameter, calculated
as
100x{1-[(c)s÷(c)l]},
and is designated by the symbol (c). A
minimum of 100
capillaries were measured in each of the four subepicardial and
subendocardial transverse tissue sections of each heart. Capillaries
were defined as those vessels having a major diameter 
8 µm.
Sarcomere length was measured on longitudinal sections. Capillary
number per fiber cross-sectional area [QA(0)] and the
percentage of capillaries supporting flow [defined as those
capillaries containing india ink; QA(0)P] were
estimated directly by standard point-counting techniques in
transverse, ie, cross, sections.27 The length of
capillaries (both total and perfused) per unit volume of muscle fiber
[JV(c,f) and JV(c,f)P,
respectively], the capillary orientation concentration
parameter (K), and anisotropy coefficient [c(K,0)] were
determined as described by Mathieu et al.24 Calculation of
capillary surface area per volume of muscle fiber
[SV(c,f)],
SV(c,f)=JV(c,f)x
x[(c)l+(c)s÷2],
represents a modification of that described by Poole and
Mathieu-Costello.26
Solutions
Low-K+ cardioplegia. In mmol/L, this was Na+ 153, K+ 4.5, Mg2+ 0.75, Ca2+ 0.25, and Cl- 140. Additionally, the perfusate contains 2 g/100 mL dextrose, 100 mg/l00 mL lidocaine, and 25 g/100 mL human albumin. The solution has a final pH between 7.4 and 7.8, with an osmolality between 306 and 315.
High-K+ cardioplegia. In mmol/L, this was Na+ 153, K+ 30, Mg2+ 0.75, Ca2+ 0.25, and Cl- 140. Additionally, the perfusate contains 2 g/100 mL dextrose, 100 mg/l00 mL lidocaine, and 25 g/100 mL human albumin. The solution has a final pH between 7.4 and 7.8, with an osmolality between 306 and 315.
K2RBC. In mmol/L, this was NaCl 115, KCl 5, CaCl2 2.5, KH2PO4 1.2, MgSO4 · 7H2O 1.2, NaHCO3 25, dextrose 5, and CaNa2 EDTA 0.08. Additionally, the perfusate contains 2 g/100 mL BSA (Sigma Chemical Co, fraction V) and washed human red blood cells to a 0.20 hematocrit.
DWB. In mmol/L, this
was NaCl 115, KCl 5, CaCl2
2.5, KH2PO4 1.2, MgSO4
1.2 · 7H2O 1.2, NaHCO3 25, dextrose 5, and
CaNa2 EDTA 0.08. This solution was mixed at a ratio of 1:1
with whole rat blood that had previously been heparinized in situ (10
U/mL). Final hematocrit was 0.20.
India ink. Nontoxic india ink containing 100% carbon particles (Hunt's Speedball No. 3232) was dialyzed against the modified Krebs-bicarbonate solution containing low K+, minus the albumin, at 4°C for 24 hours. This approximately doubled the volume of the ink. The solution was then filtered through Whatman No. 1 filter paper and heparinized (final concentration, 100 U/mL). Immediately before use, the solution was warmed to 37°C and equilibrated with 95% O2/5% CO2.7
Statistical Analysis
For all variables, the SEM indicates the
variability between
quadrates at the sampling location analyzed in each heart.
Additionally, statistics between tissue blocks were used to
analyze the regional variability of the estimates of capillary
density, tortuosity, etc, in each preparation. Differences between
K2RBC and DWB hearts and Epi and Endo within hearts were assessed by
two-factor ANOVA with replication. Differences between K2RBC and
DWB hearts were examined by unpaired Student's t test.
Correlation analyses were performed with standard
least-squares regression techniques. Significance was accepted at
P<.05.
|
Results |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
The morphometric data for the Epi and Endo sections of hearts subjected to 6 hours of hypothermic, static preservation followed by reperfusion with the K2RBC solution (group 1, n=5) or DWB (group 2, n=5) and Langendorff perfusion for 60 minutes are presented in Tables 1
and 2. Measures of left
ventricular function are included in Table 2.
|
|
Structural Analysis
Total Capillarity
Total
capillary number per fiber cross-sectional area
[QA(0)] was essentially the same for both groups (Table
1). Analysis by unpaired Student's t test
revealed
no differences between these groups. However, two-factor ANOVA
revealed that in both groups, Epi was significantly greater than Endo
(P<.05).
The capillary orientation coefficient, c(K,0),
ranged from 1.02 to
1.15, indicating that capillary tortuosity and branching contributed
between 2% and 15% to total capillary length per fiber volume,
JV(c,f), compared with straight, unbranched capillaries
oriented strictly parallel to the muscle fiber axis.24
There was no significant difference in the capillary orientation
coefficient between groups 1 and 2 or between Epi and Endo within each
group. This observation is supported by the absence of statistical
differences in capillary length per fiber volume between groups 1 and
2, whereas differences between Epi and Endo in the K2RBC-reperfused
hearts and DWB-reperfused hearts persisted (Table 1,
P<.05).
As would be expected in the absence of
statistically significant
differences in capillary length per fiber volume and capillary
diameters between groups 1 and 2, no differences in capillary surface
area per volume of muscle, SV(c,f), were found between and
within hearts in group 1 (Table 1). However, the differences
noted in
QA(0) and JV(c,f) between Epi and Endo in both
groups did not persist for capillary surface area per fiber volume
(Table 1). The tendency for a slightly higher Endo capillary
diameter,
although not statistically significant, most likely produced this
effect (Table 2).
Perfused Capillarity
There were no differences in perfused capillary number per fiber
area, QA(0)P, between Epi and Endo
within groups 1 and 2 (Table 1). However, the perfused
capillary number
per fiber area was significantly lower in the hearts reperfused with
DWB compared with those reperfused with the solution containing K2RBC
(P<.05, Table 1). Thus, in group 1, 75±6% of
the
capillary number per fiber area was perfused, compared with 56±3% in
group 2 (Fig 2, left; P<.01). No differences
in the % QA(0)P were measured between Epi and
Endo in either experimental group.
|
Capillarity tortuosity and branching
did not appreciably contribute to
perfused capillary length per fiber volume,
JV(c,f)P, as evidenced by essentially
equal values for c(K,0), the capillary orientation coefficient, for
total and perfused capillaries (Table 1). The same instance
observed
for perfused capillary number, therefore, persisted for perfused
capillary length per fiber volume. Although no differences were
measured between Epi and Endo within groups 1 and 2,
JV(c,f)P in hearts reperfused with the solution
containing DWB was significantly less than that calculated for the
hearts reperfused with K2RBC (Table 1, P<.05).
Consistent with alterations in capillary number per fiber area
and capillary length per fiber volume were the changes observed with
respect to perfused capillary surface area per fiber volume,
SV(c,f)P. As illustrated in Fig 2, right,
SV(c,f)P was significantly greater in hearts in
group 1 than in hearts in group 2 (76±6% and 58±3%,
respectively,
P<.05). Values for group 1 Epi were not different from
group 1 Endo, nor were there differences between Epi and Endo in hearts
in group 2 (Table 2).
Capillary Diameter
Capillary diameter measured across the short axis of the vessel,
(c)s, was not different between and
within
groups 1 and 2 (Table 2). Likewise, there were no differences
in
capillary diameter measured across the greatest axis of the vessel,
(c)l, between and within hearts in
group 1
and group 2 (Table 2). The degree of capillary compression was
calculated as the percentage of
(c)s to
(c)l and is designated by the
symbol %
(c). Comparison of
(c) revealed that the degree of
capillary compression was the same between and within the two groups
(Table 2).
The strength of the correlation between both
the percentage of perfused
capillary number per fiber area [% QA(0)P]
and perfused capillary surface area per fiber volume [%
SV(c,f)P] and the degree of capillary
compression [% (c)] was the
same (r=.78 and
r=.79, respectively, P<.01; Fig 3).
However, the slopes of these relationships were
significantly steeper for the hearts reperfused with K2RBC compared
with DWB.
|
The strength of the correlation between perfused capillarity
and
capillary compression was not as significant in hearts reperfused with
DWB. The correlation between % QA(0)P and %
(c) was 0.92 (P<.01) for the
K2RBC group,
compared with 0.66 (P<.05) for the DWB group. Likewise, %
SV(c,f)P and % (c)
were very highly
correlated for group 1 (r=.92; P<.01), while
less so in group 2 (r=.67; P<.05). The slopes of
the lines describing the relationship between the percent perfused
capillary density and percent perfused capillary surface area and the
degree of capillary compression were different between K2RBC- and
DWB-perfused hearts. The slope of the line describing the relationship
between % QA(0)P and %
(c) was
-4.02 in group 1 compared with -2.12 in group 2,
P<.01.
The results for percent perfused capillary surface area were similar to
those for percent perfused capillary density. The slopes of the lines
describing the relationship between % SV(c,f)P
and % (c) were -4.06 and
-2.12 for groups 1 and 2,
respectively (P<.01), indicating that the rate of rise of
capillary perfusion is greater for hearts reperfused with K2RBC
compared with those reperfused with DWB.
It was also determined that the y intercepts of the lines describing the relationship between the percentage of perfused capillary number per fiber area and the percentage of perfused capillary surface area per fiber area were significantly different. For group 1, the y intercepts were 150.23 and 151.38 for % QA(0)P and % SV(c,f)P, respectively, compared with 98.79 and 100.31 for the hearts in group 2 (P<.05). These findings indicate that, at the same magnitude of capillary compression, hearts reperfused with DWB will not exhibit as high a level of capillary perfusion as those reperfused with K2RBC.
Functional Analysis
Left Ventricular Function
The mean systolic pressure of hearts reperfused with K2RBC for 1
hour after the 6-hour preservation period averaged 111±14 mm Hg. This
measure was not statistically different from that recorded from
those hearts that were reperfused with DWB immediately before perfusion
with K2RBC during the 1-hour assessment period (93±9 mm Hg). No
significant differences were measured in the diastolic
pressures of groups 1 and 2 (44±14 and 34±9 mm Hg,
respectively).
The LVDP tended to be lower in the hearts reperfused with DWB, 59±3
mm Hg, compared with those reperfused with K2RBC, 67±7 mm Hg;
however, the difference was not statistically significant.
As
illustrated in Fig 4, left ventricular
pump function of the preserved hearts was directly related to the level
of capillary perfusion upon reperfusion, regardless of the conditions
for reperfusion. For both QA(0)P and
SV(c,f)P, the LVDP of the isolated
hearts was directly related to the perfused capillary number per fiber
area (r=.86, P<.01; Fig 4, left)
and capillary
surface area per fiber volume (r=.87, P<.01; Fig
4, right).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
This investigation was designed to determine the effects of hypothermic preservation and blood reperfusion on coronary microvascular perfusion and left ventricular function. Morphometric techniques recently used in the heart17 26 enabled us to analyze the completeness of microvasculature reperfusion with solutions containing either washed red blood cells or diluted whole blood. A modified Langendorff technique21 was used with a perfusate containing washed red blood cells14 20 28 to acquire functional and morphological data from the same heart. To the best of our knowledge, this study is the first investigation to demonstrate a direct relationship between capillary compression and capillary perfusion with sanguineous solutions. Furthermore, this study is the first to demonstrate a direct relationship between microvascular perfusion and left ventricular function of the heart after preservation. Although clinical comparisons cannot always be made by this technique, the use of a human blood–based perfusate is physiological and has been shown to maintain cardiac ventricular function and coronary microvascular integrity in an isolated heart preparation.14 The results of this study are consistent with the hypothesis that factors associated with the compression of the capillaries constitute an important mechanism that contributes to no-reflow after hypothermic ischemia and reperfusion. Moreover, we demonstrated that the effect of microvascular compression is exacerbated by the introduction of whole blood components into the reperfusate. Finally, the contractility of the heart after extended hypothermic preservation is directly related to the severity of no-reflow, regardless of the mechanisms that underlie the perfusion deficit.
Capillary Compression
We previously found that the ability of
the coronary
capillary bed to support the flow of an acellular solution after 30
minutes of normothermic ischemia was inversely
related to the effective capillary diameter, which was defined as the
capillary diameter measured across the shortest axis of the
vessel.17 In the present study, using the same
morphometric techniques, we measured the number of india
ink–filled capillaries after 6 hours of static, hypothermic
preservation followed by reperfusion with a solution containing either
washed red blood cells or diluted whole blood.
In agreement with our earlier findings, for which we used a normothermic model of ischemia and reperfusion, we found that the proportion of ink-filled capillaries was inversely related to the degree of capillary compression after hypothermic ischemia and normothermic reperfusion. These results are consistent with our finding that microvascular compression is an important contributory mechanism to the no-reflow phenomenon after myocardial ischemia and reperfusion.
Blood Components
Blood elements, particularly leukocytes, are
known to
aggravate the no-reflow phenomenon after myocardial
ischemia.5 6 7 8 9 10
The findings of the present
study are consistent with earlier studies of
normothermic ischemia and reperfusion. Engler et
al11 studied the accumulation of leukocytes during a
3-hour period of reduced coronary perfusion followed by
reestablishment of flow. They found that circulating leukocytes become
trapped in the coronary capillaries during the ischemic
insult. However, restoration of flow resulted in further entrapment of
the leukocytes, the magnitude of which was inversely related to
regional blood flow measurements. Also, the contribution of leukocytes
and platelets to the reduction in microvascular perfusion after 30
minutes of global normothermic ischemia was studied
by Reynolds and McDonagh.13 Upon direct visualization of
the coronary microcirculation, they found that when the hearts
were perfused with DWB before the ischemic insult, the number
of perfused capillaries was decreased by 62%. When hearts were
perfused with either leukocyte-free or
leukocyte/platelet–free solutions, the reduction in perfused
capillary density was only to 33% and 25%, respectively. This result
is similar to the 25% reduction observed in the present study with
reperfusion with K2RBC after hypothermic preservation. However, the
effect of DWB reperfusion on perfused capillarity after hypothermic
ischemia was not as pronounced as that observed after
normothermic ischemia.
In the present study, the use of the two
reperfusate solutions
enabled us to evaluate the effects of blood components on the
no-reflow phenomenon. We found that capillary perfusion was
significantly less in the DWB-perfused hearts than in the
K2RBC-perfused hearts in terms of both perfused capillary number per
tissue area and perfused capillary surface area per fiber volume
(P<.01). The difference in capillarity was found despite
essentially equal capillary compression in both groups (19% and 20%
for groups 1 and 2, respectively). Differences in capillary perfusion
between these groups, independent of capillary compression, are further
illustrated by regression analysis of the relationship between
capillary perfusion and capillary compression. A linear relationship
exists between both % QA(0)P and %
SV(c,f)P and the %
(c) in both
groups of hypothermically preserved hearts. There were, however,
significant differences in the y intercepts
(P<.05) and slopes (P<.01) measured between
hearts reperfused with solutions containing washed red blood cells
compared with diluted whole blood. This demonstrates that at an
equivalent degree of microvascular compression, capillaries in hearts
reperfused with DWB were not as well perfused as those reperfused with
washed red blood cells. Moreover, the rate of loss of capillary
patency, as described by the slopes of these lines, was greater in the
DWB- than in the K2RBC-reperfused hearts at any level of capillary
compression (Fig 3).
The contribution of the individual blood components, such as leukocytes and platelets, was not evaluated in the present study. However, the differences found between hearts reperfused with a solution containing only washed red blood cells in contrast to the full complement of cellular components contained in the diluted whole blood support the conclusion that leukocytes and/or platelets in the reperfusate contribute to no-reflow after hypothermic ischemia and reperfusion. Although the exact mechanisms were not investigated in this study, leukocyte–endothelial cell interactions most likely play a significant role. These mechanisms may involve the activation of selectin-mediated leukocyte rolling and the family of integrin/intercellular adhesion molecules responsible for leukocyte adhesion. Once adhered, leukocytes may then release oxygen radicals, which can damage both the microvascular endothelium and the myocytes.
Left Ventricular Pump Function
In previous myocardial
preservation studies, a direct relationship
was found between coronary perfusion during or after
hypothermic preservation of the heart and the recovery of pump function
upon reperfusion.7 29 After 24 hours of low-pressure,
hypothermic perfusion, Manciet and Copeland7 found the
percentage of filled microvessels in isolated rabbit hearts to be
inversely related to the amount of measured thiobarbituric
acid–positive material (principally malondialdehyde, a
by-product of lipid peroxidation) in the tissue. The severity
of lipid peroxidation was related to decreases in coronary flow
during preservation. The recovery of LVDP was, in turn, lowest in the
hearts that exhibited the greatest levels of lipid peroxidation.
Ledingham et al29 compared the efficacy of various
solutions for the preservation of left ventricular function
in rat hearts after 4 hours of static, hypothermic storage. The hearts
that had the greatest measured coronary flow rates upon
reperfusion also had the greatest recovery of pump function, as
evidenced by greater ±dP/dt and cardiac output.
In addition to coronary flow, the differential effects of reperfusion with varied blood cellular components on the functional recovery of normothermically ischemic hearts have been studied. In a study examining the direct effects of whole blood reperfusion on recovery of contractile function in isolated hearts after 30 minutes of normothermic ischemia, McDonagh and Reynolds14 found that the recovery of LVDP was only 28% of that measured in nonischemic hearts. This recovery was significantly lower than that recorded from hearts that were reperfused with solutions containing either K2RBC or leukocyte-depleted DWB.
In agreement with these earlier studies, we have demonstrated that after 6 hours of hypothermic, static ischemia, left ventricular function was greatest in the hearts that exhibited the greatest proportion of perfused capillaries upon reperfusion. This finding was independent of the composition of the reperfusate. Although there was a trend toward reduced left ventricular contractility in the DWB group, we did not find a statistically significant decrease in the mean LVDP for hearts reperfused with DWB compared with mean LVDP for those reperfused with the K2RBC. As mentioned earlier, the effect of DWB reperfusion on no-reflow was also not as pronounced after hypothermic preservation compared with normothermic ischemia. Hearts that were reperfused with the K2RBC had significantly better capillary perfusion than hearts reperfused with DWB. Because the level of left ventricular pump function is directly related to the level of capillary reperfusion and because capillary perfusion in hearts reperfused with DWB was significantly worse than that observed in hearts reperfused with K2RBC, it follows that hearts reperfused with DWB immediately after ischemia would not function as well as those reperfused with the solution containing K2RBC.
In conclusion, using a novel experimental system that provides for functional and structural analysis to be accomplished on the same heart, we found that (1) after 6 hours of hypothermic, static preservation of the heart, capillary perfusion deficits are directly related to the degree of capillary compression; (2) these perfusion deficits are exacerbated by blood reperfusion; and (3) regardless of the composition of the reperfusate, left ventricular function after hypothermic ischemia is directly related to the ability of the coronary capillaries to support flow upon reperfusion. On the basis of these findings, we conclude that after an extended period of hypothermic preservation, compression of the capillaries is a significant contributory mechanism for impaired reperfusion of the myocardium. Further, whether as a function of mechanical plugging or activation of cellular adhesion molecules by platelets, monocytes, lymphocytes, or neutrophils, the inclusion of whole blood elements in the reperfusate intensifies this no-reflow effect. Finally, perhaps the most significant finding in this study with respect to the maintenance of myocardial function in the hypothermically preserved heart is that contractility of the heart after extended hypothermic preservation is directly related to the severity of no-reflow, regardless of the mechanisms that underlie the perfusion deficit.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
Funds for these studies were provided by the American Heart Association, Arizona Affiliate, and NIH grants HL-49230 and HL-07249. The authors thank David C. Poole, PhD, for his editorial assistance.
|
References |
|---|
|
TopAbstractIntroductionMethods Results Discussion References |
|---|
1. Emery RW, Cork RC, Levinson MM, Copeland J, McAleer MJ, Copeland JG. The cardiac donor: a six-year experience. Ann Thorac Surg. 1986;41:356-362. [Abstract]
2. Kushwaha SS, Crossman DC, Bustami M, Davies GJ, Mitchell AG, Maseri A, Yacoub MH. Substance P for evaluation of coronary endothelial function after cardiac transplantation. J Am Coll Cardiol. 1991;17:1537-1544. [Abstract]
3. Jeevanandam V, Auteri JS, Sanchez JA, Barr ML, Ott GY, Hsu D, Marobe C, Smith CR, Rose EA. Improved heart preservation with University of Wisconsin solution: experience and preliminary human experience. Circulation. 1991;84(suppl III):III-324-III-328.
4. Hearse DJ, Braimbridge MV, Jynge P. Protection of the Ischemic Myocardium: Cardioplegia. New York, NY: Raven Press; 1981:21-49.
5. Copeland JG, Jones M, Spragg R, Stinson EB. In vitro preservation of canine hearts for 24 to 28 hours followed by successful orthotopic transplantation. Ann Surg. 1973;178:687-692. [Medline] [Order article via Infotrieve]
6. Copeland JG, Stinson EB. Evaluation of hypothermic perfusion for 24-hour preservation of canine hearts. Transplant Proc. 1974;6:311-313.[Medline] [Order article via Infotrieve]
7. Manciet LH, Copeland JG. Lipid peroxidation results in compromised functional recovery of isolated rabbit hearts following low-pressure, hypothermic, perfused preservation for 24 hours. J Thorac Cardiovasc Surg. 1992;103:532-540. [Abstract]
8. Krug A, Du Mesnil de Rochemont W, Korb G. Blood supply of the myocardium after temporary coronary occlusion. Circ Res. 1966;9:57-62.
9. Kloner RA, Ganote CE, Jennings RB. The "no-reflow" phenomenon after temporary coronary occlusion in the dog. J Clin Invest. 1974;4:1496-1508.
10. Opie LH. Myocardial ischaemia: metabolism and its modification. S Afr Med J. 1987;72:740-747. [Medline] [Order article via Infotrieve]
11. Engler RL, Dahlgren MD, Peterson MA, Dobbs A, Schmid-Schönbein G. Accumulation of polymorphonuclear leukocytes during 3-h experimental myocardial ischemia. Am J Physiol. 1986;251(Heart Circ Physiol 20):H93-H100.
12.
Golino P, Maroko PR, Carew TA. Efficacy of
platelet depletion in counteracting the detrimental effect of acute
hypercholesterolemia on infarct size and the no-reflow
phenomenon in rabbits undergoing coronary artery
occlusion-reperfusion. Circulation. 1987;76:173-180.
13. Reynolds JM, McDonagh PF. Early in reperfusion, leukocytes alter perfused coronary capillarity and vascular resistance. Am J Physiol. 1989;256(Heart Circ Physiol 25):H982-H989.
14. McDonagh PF, Reynolds JM. Brief perfusion with diluted whole blood after global ischemia increases reperfusion injury. Cardiovasc Res. 1994;28:1157-1165. [Medline] [Order article via Infotrieve]
15.
Leaf A. Cell swelling: a factor in
ischemic tissue injury.
Circulation. 1973;48:455-458.
16. Lee BY, Wilson GJ, Domenech RJ, MacGregor DC. Relative roles of edema versus contracture in the myocardial postischemic "no-reflow" phenomenon. J Surg Res. 1980;29:50-61. [Medline] [Order article via Infotrieve]
17. Manciet LH, Poole DC, Copeland JG, McDonagh PF, Mathieu-Costello O. Mechanistic basis for no-reflow phenomenon. Am J Physiol. 1994;266(Heart Circ Physiol 35):H1541-H1550.
18. Alanen K, Nevalainen TJ, Lipasti J. Ischaemic contracture and myocardial perfusion in isolated rat hearts. Virchows Arch A Pathol Anat Histol. 1980;385:143-149. [Medline] [Order article via Infotrieve]
19. Gavin JB, Nevalainen TJ, Seeyle RN, Webster VJ, Thomson RW. An association between the onset of rigor and the loss of vascular competence in early myocardial infarcts. Pathology. 1978;10:219-225. [Medline] [Order article via Infotrieve]
20. McDonagh PF, Niven AT, Roberts D. Direct visualization of the coronary microcirculation for pharmacologic and physiologic studies. Microvasc Res. 1984;28:180-196. [Medline] [Order article via Infotrieve]
21. Langendorff, O. Untersuchungen am überlebenden Saugethierherzen. Pflugers Arch. 1895;61:1-4.
22. Manciet LH, Larson DF, Copeland JG. Low-pressure perfusion results in effective microvascular perfusion of isolated rabbit hearts during hypothermic preservation for 24 hours. J Heart Lung Transplant. 1991;10:710-716. [Medline] [Order article via Infotrieve]
23.
Poole DC, Batray S, Mathieu-Costello O, Rakusen K.
Capillary geometrical changes with fiber shortening in rat
myocardium. Circ Res. 1992;70:697-706.
24. Mathieu O, Cruz-Orvie L, Hoppeler H, Weibel E. Estimating length density and quantifying anisotropy in skeletal muscle capillaries. J Microsc. 1983;131:131-146. [Medline] [Order article via Infotrieve]
25. Mathieu-Costello O. Capillary tortuosity and degree of contraction or extension of skeletal muscles. Microvasc Res. 1987;33:98-117. [Medline] [Order article via Infotrieve]
26. Poole DC, Mathieu-Costello O. Analysis of capillary geometry in rat subepicardium and subendocardium. Am J Physiol. 1990;259(Heart Circ Physiol 28):H204-H210.
27. Weibel ER. Point counting methods. In: Weibel ER, ed. Stereological Methods. New York, NY: Academic Press; 1979;1:101-161.
28. Gamble WJ, Conn PA, Kumar AE, Plenge R, Monroe RG. Myocardial oxygen consumption of blood-perfused, isolated, supported rat heart. Am J Physiol. 1970;219:604-612.
29. Ledingham SJM, Katayama O, Lachno DR, Yacoub M. Prolonged cardiac preservation: evaluation of the University of Wisconsin preservation solution by comparison with the St Thomas' Hospital cardioplegic solutions in the rat. Circulation. 1990;82(suppl IV):IV-351-IV-358.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||