Functional and Structural Alterations With 24-Hour Myocardial Hibernation and Recovery After Reperfusion
A Pig Model of Myocardial Hibernation
Background Short-term myocardial hibernation of 3 hours resulting from a moderate resting coronary flow reduction has been reproduced in pigs. This study was designed to determine whether any structural changes accompany short-term hibernation caused by a moderate flow reduction maintained for 24 hours and whether any such structural alterations are reversible after reperfusion.
Methods and Results A severe left anterior descending coronary artery (LAD) stenosis was created with a reduction of resting flow to ≈60% of baseline and maintained for 24 hours. Regional coronary flow was measured by a flowmeter; wall thickening was determined by echocardiography, and local metabolic changes were measured. Of 17 pigs, 11 completed the study protocol of 24 hours. The LAD flow was reduced from 0.91±0.11 to 0.52±0.13 mL·min−1·g−1, a 43% mean decrease, at 15 minutes after the LAD stenosis and was maintained at 0.56±0.11 mL·min−1·g−1 at 24 hours. The reduction of regional coronary flow initially produced acute myocardial ischemia, as evidenced by reduced regional wall thickening (from 37.2±6.9% at baseline to 11.5±6.8%), regional lactate production (−0.34±0.28 μmol·g−1·min−1), and a decrease in regional coronary venous pH (from 7.41±0.035 at baseline to 7.30±0.030). At 24 hours, the reductions in coronary flow and wall thickening were maintained relatively constant and the rate-pressure product was relatively unchanged, but lactate production ceased and regional H+ concentration normalized, with a tendency toward a further reduction in regional oxygen consumption, from 3.10±0.90 mL·min−1·100 g−1 at 15 minutes after stenosis to 2.52±0.95 mL·min−1·100 g−1 at 24 hours (P=.06), indicating metabolic adaptation of the hypoperfused regions. Of 11 pigs, 6 were free of myocardial infarction; 3 had patchy necrosis involving 4%, 5%, and 6% of the area at risk; and 2 other pigs had a few scattered myocytes with necrosis, detected only by light and electron microscopy. Ultrastructural changes consisted of a partial loss of myofibrils and an increase in mitochondria and glycogen deposition. Regional wall thickening recovered 1 week after reperfusion in most pigs, and the ultrastructural changes reverted to normal.
Conclusions In this pig model, moderately ischemic myocardium undergoes metabolic and structural adaptations but preserves the capacity to recover both functionally and ultrastructurally after reperfusion.
Recent investigations have demonstrated that regional myocardial dysfunction in coronary disease may be a consequence of infarcted, stunned, or hibernating myocardium.1 2 3 4 5 6 7 8 9 The term “myocardial hibernation” has been proposed to describe persistent regional myocardial contractile dysfunction caused by reduced regional coronary flow.1 2 Stunned myocardium, defined as postischemic myocardial dysfunction despite restoration of adequate perfusion, results from one or more discrete episodes of ischemia and gradually recovers over hours to days.3 4 In contrast, revascularization or an intervention that improves oxygen supply-and-demand balance is required to improve or restore mechanical function of hibernating myocardium.1 2 5
Although the difference between stunned and hibernating myocardium is well defined conceptually, differentiating between these two conditions can be difficult clinically. In clinical studies,6 7 8 functional recovery of akinetic and even dyskinetic LV regions has been documented after revascularization in chronic coronary disease, indicating that viable myocardium existed in these regions. This dysfunctional but viable myocardium could be either hibernating or stunned, depending on the status of myocardial perfusion. With use of PET, myocardial flow assessed by 13N-ammonia has been reported to be either normal or only mildly reduced8 9 in reversibly dysfunctional regions, which has been considered to be more consistent with stunning than with hibernation, leading to the debate about whether true myocardial hibernation can exist over a prolonged period with persistently reduced perfusion.5 10 11
The concept of long-term myocardial hibernation in animal models also has been controversial. Bolukoglu et al12 and Liedtke et al13 have described a model of long-term coronary stenosis with a reduction of phasic velocity of 50% without a reduction of mean coronary flow. Regional wall thickening was normal immediately after placement of the stenosis but was mildly reduced later, after the stenosis had been maintained for 4 to 7 days. However, if the definition of myocardial hibernation requires both a reduced resting coronary flow and reduced regional wall thickening or contractile dysfunction, whether their model represents myocardial hibernation would be debatable. Furthermore, Shen and Vatner14 have presented evidence from a long-term animal model to support their contention that the term “hibernation” should be reconsidered and replaced by “long-term stunning.”
Nevertheless, short-term myocardial hibernation has been well documented in several short-term animal models, with mildly to moderately reduced resting coronary flow and myocardial dysfunction with minimal necrosis in pigs and dogs.15 16 17 18 19 20 Metabolic adaptation, characterized by regeneration of high-energy compounds and recovery of lactate production, has been observed in pigs with short-term myocardial hibernation caused by a moderate flow reduction maintained for 60 minutes to 3 hours.15 16 However, whether such a moderate flow reduction can be sustained for longer periods without significant ultrastructural alterations or extended infarction has not been demonstrated in species with sparse preexisting collateral circulation, such as pigs or humans. The comparability of these ultrastructural alterations, if any, to changes in myocardial regions subtending long-term coronary stenosis in patients has not been investigated.7 8 Furthermore, although functional recovery of such hypoperfused myocardium with ultrastructural alterations in patients has been documented, whether such functional recovery is associated with morphological recovery has not been studied.7 8 20 Therefore, this study was designed to examine in pigs whether any necrosis or ultrastructural abnormalities occur in hibernating myocardium with a 24-hour moderate coronary flow reduction and whether the structural alterations, if any, are reversible after reperfusion.
The study protocol was approved by the Committee on Animal Care at Hartford Hospital, and the animal care guidelines of the American Heart Association were followed. Seventeen pigs weighing 25 to 43 kg were fasted for 12 hours and premedicated with ketamine (1 mg/kg IM), penicillin (2200 U/kg IV), and gentamicin (3 mg/kg IV). General anesthesia was induced with ketamine (1 mg/kg IM) and fentanyl and was maintained with isoflurane (0.5% to 1.5%) with an oxygen–nitrous oxide mixture (40% to 50%:50% to 60%). The pigs were intubated and connected to a respirator. The ventilation rate and volume were adjusted to maintain normal arterial blood gases. The isoflurane concentration was titrated to suppress the pain reflex without deeper anesthesia to minimize the dose-dependent cardiodepressive and ventilatory effects of isoflurane and retain dynamic coronary autoregulation. Rectal temperature was monitored and maintained at 97°F to 98°F by an electrically heated surgical table and drapes.
A midline thoracotomy was performed, and the heart was suspended in a pericardial cradle. A 7F Eppendorf catheter was inserted into the right femoral artery to monitor pressure and obtain blood samples. A 6F pigtail catheter was inserted through the atrial appendage into the LV to monitor LV pressure. Left atrial pressure was monitored with another 6F pigtail catheter. The jugular veins were used to administer fluids and medications during the study. Full anticoagulation was achieved and maintained with heparin 200 IU/kg IV followed by 30 IU/kg IV every hour. A 3F to 4F coronary perfusion catheter was advanced retrogradely through the coronary sinus to the proximal interventricular vein running parallel to the LAD to monitor oxygen content, lactate, and pH.
To prevent local coronary spasm from surgical manipulation, 3% lidocaine drops were intermittently applied locally at the proximal LAD site at which the artery was manipulated. The LAD was carefully dissected free over 1 to 2 cm to accept a probe (Transonic Inc) to measure coronary flow. The cuff of the flow probe was carefully aligned parallel to the vessel to ensure accurate measurement. A hydraulic cuff occluder was placed around the LAD immediately distal to the flowmeter. To ensure a normal coronary reserve at baseline, the hyperemic reaction of the LAD was determined by dividing the peak coronary flow after a brief (10-second) occlusion by the resting coronary flow. In 10 pigs, the LAD stenosis was created by gradually filling the hydraulic occluder with saline to reduce the resting LAD coronary flow to ≈60% of baseline (≈40% reduction). In the other 7 pigs, a silk tie was used with an 18- to 22-gauge arterial puncture needle with an outer diameter of 0.4 to 1.3 mm to create a graded stenosis. To do this, the size of the needle was first chosen according to the size of the proximal LAD to make a severe stenosis with a reduction in resting coronary flow to ≈60% of baseline. The selected needle was placed parallel to the proximal LAD. A 2-0 silk tie was placed around the proximal LAD and the needle and then tightened. The needle was withdrawn immediately (within 10 seconds), leaving a stenotic LAD lumen the size of the needle or smaller. This procedure was repeated with a series of ties or a smaller needle until the desired reduction of coronary flow was achieved. On average, two ligations were needed. The end of the tie was then fixed to prevent distal migration.
Baseline measurements of wall thickening by echocardiography, heart rate, LV pressure, aortic pressure, regional coronary flow, coronary venous lactate level, H+ concentration, and oxygen content were obtained under stable conditions. Stability was defined as two consecutive measurements at 5-minute intervals with a pH difference <0.02, a coronary flow difference <3 mL, and a mean blood pressure difference <5 mm Hg. LAD flow was reduced to ≈60% of baseline, and the reduction was maintained for 24 hours. The stability of the coronary stenosis was verified by serial measurements of coronary flow at 15 and 60 minutes and 24 hours under the same conditions of anesthesia.
In 8 of 11 pigs, after maintenance of a stable LAD stenosis for >60 minutes, the chest was closed in layers with the pericardium left open. In 4 pigs, the catheters and flow probe were left in place; in the other 4, the catheters and flow probe were removed before the chests were closed. The pigs were then allowed to recover in their cages. Aspirin and intravenous heparin were given postoperatively to prevent thrombotic coronary occlusion. After 24 hours of LAD stenosis, the pigs were restudied. The catheters and flow probe were checked or reinstalled in the same position as before, and all measurements were repeated under the same anesthesia. Because the LAD was isolated on the prior day, reinstallation of the flow probe in 4 pigs was done within a few seconds without occlusion of the LAD. Myocardial biopsies were taken from the myocardial region perfused by the LAD, around the midpapillary muscle level, with a Tru-Cut biopsy needle (20-mm, 14-gauge; Travenol Laboratories). The LAD stenosis was then released, and the chest was closed in layers with the pericardium left open. Because regional myocardial dysfunction did not immediately improve on release of the stenosis (see the “Results” section), the pigs were kept alive for 7 days to document recovery of regional myocardial dysfunction. One pig died 16 hours after release of the coronary stenosis as a result of massive hemorrhage. In all surviving pigs, transthoracic echocardiograms were performed daily to monitor regional LV wall thickening. Epicardial echocardiograms were repeated at 1 week in 6 pigs and at 4 weeks in 1 pig before death. The heart was harvested for pathological examinations (see below).
In the remaining 3 pigs, regional LAD flow was measured by a transonic flowmeter, and heart rate and blood pressure were monitored continuously for 24 hours under the same conditions of anesthesia to evaluate temporal variations of LAD flow after creation of the stenosis in this model. The pigs were then killed at 24 hours for morphological evaluations with gross inspection after TTC staining, light microscopy, and electron microscopy.
Epicardial two-dimensional echocardiography was performed to evaluate the severity of the reduction of regional myocardial wall thickening at the short-axis view of the midpapillary muscle level of the LV. Images were obtained from the epicardial surface of the right ventricle. Identical echocardiographic views were obtained before and after creation of the LAD stenosis, at 24 hours, after release of the stenosis, and before death. Transthoracic echocardiography was performed daily in the surviving pigs to monitor changes in wall thickening. Wall thickness was measured at the midpapillary muscle level as described previously.20 Anterior wall thickness was measured at the midanterior wall. The inferior (control segment) wall thickness was measured at the opposite wall. End-diastolic wall thickness was measured when the LV cavity was maximal. End-systolic wall thickness was measured when the LV cavity first became minimal, corresponding to the end of the T wave on the ECG. Regional LV wall thickening was calculated as end-systolic minus end-diastolic wall thickness divided by end-diastolic wall thickness, expressed as a percentage. Two observers blinded to each other's results performed all echocardiographic measurements. The mean values of the measurements are presented.
Regional Myocardial Blood Flow
Regional coronary blood flow was measured with a cuff flow probe connected to a transonic flowmeter. At the conclusion of each experiment, the flowmeter was calibrated against a known flow rate to ensure accuracy. Methylene blue was injected into the LAD or circumflex artery to stain the tissue supplied by the vessel. The stained tissue was dissected and weighed to determine the regional myocardial mass perfused by the stenotic coronary artery. Coronary blood flow is expressed as milliliter per minute per gram of wet tissue.21
Myocardial Metabolic Measurements
Arterial and coronary venous blood samples were obtained anaerobically in cold, dry syringes containing heparin fluoride to inhibit glycolysis. Samples were divided for blood gases and glucose and lactate content, stored in ice, and processed immediately after the experiment. Blood gases were measured in duplicate, and the values were averaged. Plasma for lactate content was deproteinated with perchloric acid, neutralized with potassium hydroxide and imidazole buffer, and analyzed with the enzymatic method. Regional myocardial oxygen consumption was calculated by multiplying the arterial–coronary venous oxygen content difference by the regional transmural blood flow supplied by the LAD. Lactate consumption or production was calculated by multiplying the arterial–coronary venous lactate difference by regional transmural myocardial blood flow. A positive value indicates lactate consumption; a negative value indicates production.
Pathological and Histochemical Morphology
At 24 hours of the LAD stenosis, transmural myocardial biopsies were taken from the LAD region around the midpapillary muscle and a control region supplied by the circumflex artery. Adequately sized myocardial samples of 3 mm by 7 to 10 mm were obtained with a Tru-Cut needle. The biopsy site was sealed by a suture, which also served to identify the site for later samples. Immediately after biopsy, the specimens were immersed in 3% glutaraldehyde fixative and buffered to pH 7.4 with 0.1 mol/L sodium dihydrogenophosphate. The tissue fragments were postfixed for 1 hour at 4°C with 1% osmium tetroxide, buffered to pH 7.4 with 0.5 mol/L veronal acetate containing 93 mmol/L sucrose, subsequently dehydrated in ethanol, and embedded in Epon for electron microscopy. For each heart, six tissue blocks were obtained randomly and included the middle and either side of the transmural needle biopsy sample to ensure that the samples studied included subendocardial, middle, and subepicardial layers. All tissue blocks were evaluated.
At the conclusion of the study (24 hours for 3 pigs, 40 hours for 1 pig, 7 days for 6 pigs, and 4 weeks for 1 pig), methylene blue was injected distally into the stenotic LAD to delineate the area at risk. The LV was cut into cross sections at 0.5-cm intervals from apex to base. The area at risk (area stained with blue) was dissected and weighed. The LV sections (both the normal part and the area at risk) were then immersed in a 0.09-mol/L sodium phosphate buffer (pH 7.4) containing 1.0% TTC for 30 minutes at 37°C to identify myocardial necrosis. Myocardium with deep red staining by TTC was considered viable; myocardium not stained by TTC was deemed necrotic. In 3 pigs with patchy necrosis (not stained by TTC), total surface area, necrotic area, and normal area of each LV section in regions supplied by the LAD (stained by blue dye) were traced on transparent paper. The infarct size for each pig was calculated by integrating necrotic areas from all LV sections and expressing them as a percent of the total area at risk. All LV sections, including areas with patchy necrosis by TTC, were then fixed with 5% formalin, embedded in paraffin, sliced into 5-μm sections, and stained with hematoxylin and eosin and trichrome stain for light microscopic examination. The myocardial samples in the region from which myocardial biopsies were taken at 24 hours with the LAD stenosis were fixed with glutaraldehyde and prepared as described above for electron microscopy to evaluate recovery of the ultrastructure after reperfusion. Histological surveys of the LAD-perfused regions and normal regions were performed by a cardiac pathologist (Dr Fallon) who was unaware of the origin of the samples. The same pathologist also performed qualitative electron microscopic evaluation independently.
To evaluate quantitative changes of cellular organelles in myocytes, a morphometric technique for the quantitative electron micrograph was used. This previously described technique was modified for this study.22 23 Myocytes for morphometry were selected according to the arbitrary sampling method advocated by Weibel.22 All myocytes with nuclei were selected from each sequential square along the grid of the electron micrographic slide, photographed, and printed with ×3000 magnification. For each heart, at least 20 myocytes were randomly selected and evaluated in each normal and LAD-perfused region from all available blocks of the transmural biopsy samples. The principle of a box-counting planimetry system was applied as described previously.22 Briefly, a grid system consisting of vertical and horizontal lines providing 360 minisquares (1×1 mm) on transparent paper was used. The grid system was then superimposed on a selected myocyte on an eletron micrographic print. The number of squares enclosed in a certain structure was counted. The squares intersected by the profile border were counted as fractions in proportion to the fraction covered by the structure. Counting of the number of points overlying a certain structure (organelle) in a cell resulted in quantitative determination of volume of the structure under investigation in relation to total volume of a cell, which was estimated by the total squares a cell was overlying. Percentages were used to express the quantitative relation between entire cell volumes and volumes of intracellular structures. The volumes of myofibrils, mitochondria, and cytoplasm were calculated according to established principles.22 Because myocytes in the hibernating region were not uniformly affected, the percentages of affected cells also were counted. Cells in which myofibrils comprised <55% of the total volume of the cell were classified as affected cells; the selection of 55% as the cutoff point was based on data from the normal control region (see the Table⇓) in which myofibrils occupied 63.8±4.9% of the total cell volume. One hundred sequentially selected myocytes from the LAD-perfused regions and normal regions were evaluated in each pig.
All parametric data are expressed as mean±SD. ANOVA was used to compare parametric data among different stages by use of a commercially available statistical software package (RS1, BBN Software Co). If there was a statistical difference by the ANOVA, a paired t test was used to examine parametric data between two stages. A paired t test also was used to examine parametric data between normal control and hibernating regions. Corrections for multiple comparisons were applied by use of the Tukey honestly significant difference test when applicable. Nonparametric data in normal and hibernating regions were tested by χ2 or Fisher's exact test. A value of P≤.05 was considered significant.
Data were collected and analyzed for only the 11 of the 17 pigs that completed the experimental protocol to 24 hours. In these pigs, heart rate was 115±10 bpm; mean blood pressure was 78±6 mm Hg at baseline and did not change significantly after placement of the stenosis. Mean left atrial pressure increased from 9±3 to 13±4 mm Hg (P<.05) after the stenosis was created and did not change significantly thereafter. During placement of the stenosis, the other 6 pigs developed ventricular fibrillation and died. The ventricular fibrillation was proceeded by an abrupt reduction of the already-reduced coronary flow to zero, most probably as a result of complete spastic occlusion of the LAD, because autopsy did not show thrombus at the LAD stenosis. The survival rate was similar for stenoses created with a hydraulic occluder (6 of 10) and with silk ties (5 of 7, P=NS).
Coronary Flow and Contractile Function
All pigs had coronary flow reserve of 3.0 or more (4.2±0.48) at baseline. Resting coronary flow in the LAD region, 0.91±0.11 mL·min−1·g−1 at baseline, was reduced to 0.52±0.13 mL·min−1·g−1 at 15 minutes after creation of the stenosis (Fig 1A⇓) and was maintained at 0.54±0.11 mL·min−1·g−1 at 60 minutes (P=NS versus that at 15 minutes) and 0.56±0.11 mL·min−1·g−1 at 24 hours (P=NS versus that at 60 minutes). There was no difference in the flow reduction between stenoses created with silk ties or a hydraulic occluder at either 15 minutes or 24 hours. Regional anterior LV wall thickening, 37.2±6.9% at baseline, was reduced to 11.5±6.8% at 15 minutes (Fig 1B⇓) and remained unchanged at 24 hours (11.2±6.4%, P=NS). No difference in anterior wall thickening was noted between stenoses created with silk ties or a hydraulic occluder at 15 minutes and 24 hours. The inferior wall thickening was 38.3±7.6% at baseline, did not change significantly after placement of the stenosis, and increased slightly (P=NS) at 24 hours (Fig 1B⇓).
In 3 pigs, coronary flow, heart rate, and blood pressure were monitored continuously over the 24 hours (Fig 2⇓). LAD flow (Fig 2A⇓) was reduced from 0.90 to 0.47 mL·min−1·g−1 (48% reduction), from 0.97 to 0.52 mL·min−1·g−1 (46% reduction), and from 1.01 to 0.56 mL·min−1·g−1 (45% reduction) in the 3 pigs at 15 minutes. The mean LAD flow over 24 hours was 0.54±0.10, 0.59±0.06, and 0.60±0.11 mL·min−1·g−1, corresponding to 40%, 41%, and 40% flow reductions, respectively. During the 24 hours, the flow reduction varied from 59% to 25% and was associated with variations in blood pressure and heart rate. The rate-pressure product varied from 8096 to 18 900 mm Hg·bpm (Fig 2B⇓). Higher coronary flows were associated with higher rate-pressure products (Fig 2⇓), but these values remained much below baseline levels in each pig throughout the 24-hour observation period.
Regional Oxygen Consumption and Metabolism
At baseline (Fig 1C⇑), lactate consumption (0.45±0.37 μmol·g−1·min−1) was observed in all pigs in the LAD region. Lactate production (−0.34±0.28 mL·min−1·g−1) developed by 15 minutes after creation of the stenosis (P<.01). This initial lactate production at 15 minutes ceased, and lactate consumption returned at 24 hours despite maintenance of the stenosis and the flow reduction. The change in regional lactate balance was associated with similar changes in regional coronary venous pH. The regional coronary venous pH decreased from 7.41±0.046 to 7.33±0.034 at 15 minutes but returned to normal (7.40±0.05) by 24 hours.
At 15 minutes, regional oxygen consumption had decreased by 36% from 5.16±0.94 to 3.10±0.90 mL·min−1·100 g−1 of myocardium (P<.01; Fig 1D⇑). A tendency toward a further reduction of oxygen consumption, to 2.52±0.95 mL·min−1·100 g−1 (P=.06), was observed at 24 hours even though regional wall thickening and rate-pressure product were unchanged (P=.1). The initial decrease in oxygen consumption after creation of the stenosis was associated with an increase in arterial coronary venous oxygen extraction (from 56.7±6.5% at baseline to 69.4±7.2%, P<.01) and a decrease in regional coronary venous oxygen saturation (from 43.2±7.5% at baseline to 30.5±6.3%, P<.01). By 24 hours, regional coronary venous oxygen saturation (45.6±7.0%) and arterial coronary venous oxygen extraction (54.2±8.8%) normalized.
Myocardial glucose consumption increased from 3.7±1.6 at baseline to 4.6±4.8 mg·min−1·100 g−1 at 15 minutes after the LAD stenosis (P<.05). The increased glucose consumption persisted at 60 minutes (4.0±2.1 mg·min−1·100 g−1) and 24 hours (4.5±2.4 mg·min−1·100 g−1).
Morphological Changes With LAD Stenosis at 24 Hours
Of the 11 pigs, 6 had no evidence of myocardial necrosis. Three pigs had patchy myocardial necrosis by TTC staining and gross pathological inspection that involved 4%, 5%, and 6% of the area perfused by the LAD (area at risk) in the subendocardium around the papillary muscle. Histological survey confirmed the TTC-detected necrosis. In 2 other pigs, a few foci of necrosis of a few myocytes confined to the subendocardium were detected only by light microscopy. In 2 of the 3 pigs killed at 24 hours without reperfusion, the microscopic picture was consistent with typical coagulation necrosis, with enhanced sarcoplasmic eosinophilia, nucleolysis, and polymorphonuclear leukocyte margination and infiltration. In the 3 pigs with patchy necrosis after reperfusion for 7 days, chronic inflammatory infiltrates and granulation tissue were noted in the patchy infarct regions. Electron microscopy revealed patchy myocyte necrosis with amorphus densities in the mitochondria, clumping of nuclear material, and ruptured sarcolemma in 4 of the 5 pigs with TTC and/or microscopic necrosis. Myocyte necrosis typically was confined to one or a few cells with adjacent normal intact myocytes.
Decrease in Myofilaments and Increase in Mitochondria and Glycogen
No significant changes were noted in the control regions (Fig 3A⇓). In the LAD region, electron microscopic examination of the biopsies from all 11 pigs revealed loss of myofilaments or sarcomeres to various degrees (Fig 3B⇓). Although the depletion of myofilaments was most obvious in the perinuclear area (Fig 3B⇓), myofilament loss extended from there to the cell periphery and occurred randomly in scattered myocytes surrounded by normal cells. The empty myofilament areas were replaced by glycogen deposits and mitochondria. Mitochondrial size in these spaces varied widely but usually appeared smaller than mitochondria in normal regions. Glycogen deposits appeared variable on the electron micrographs (Figs 3B and 4⇓⇓) in different pigs and even in different myocytes for the same pigs. Myocytes were not uniformly affected; instead, damage usually was localized to one or a few adjacent myocytes, whereas other regions of the same myocyte appeared to have a normal density of myofibrils and mitochondria (Fig 4⇓). Semiquantitative counting of myocytes revealed that 39±18% were affected (ie, myofibrils ≤55% of total cell volume) in hibernating regions compared with 6±2% in normal regions (P=.01). Other ultrastructural changes included dilated sarcoplasmic reticulum and intrasarcomere and intercellular edema. The capillary vascular structure was intact, with no hemorrhage or thrombus observed in the vascular system. Quantitative evaluation confirmed that myofibrils or sarcomeres were decreased (P<.01) and cytoplasm (P<.01) and mitochondria (P<.05) were increased compared with normal regions (the Table⇑).
Morphological Changes With Reperfusion
Fig 3C⇑ shows an example of ultrastructural recovery of hibernating myocardium after 7 days of reperfusion. After release of the LAD stenosis (7 days for 6 pigs and 4 weeks for 1 pig), the appearance of the myofilaments and mitochondria returned to normal in all but 1 pig in which some depletion of myofilaments persisted after 1 week. In 3 pigs, inflammatory cells still were noted in the patchy area of myocardial necrosis. The other pig died of hemorrhage 16 hours after release of the stenosis; no myocardial necrosis was found, but myofibrillar loss and intercellular edema were present, as in the biopsies at 24 hours.
Coronary Flow and Function With Reperfusion
Regional coronary flow was 2.08±0.15 mL·min−1·g−1 at 3 minutes after release of the LAD stenosis, reflecting hyperemia. Regional flow decreased to 1.13±0.16 mL·min−1·g−1 at 20 minutes, which was not different from baseline.
Regional wall thickening recovered to 25.6±7.3% at 1 week but was still less than at baseline (33.0±5.9%, P=.06). In 5 pigs, transthoracic echocardiography was performed daily for 1 week after release of the stenosis. Progressive functional recovery could be tracked: wall thickening was 32±5.8% at baseline; 10.2±6.3% at 15 minutes after creation of the stenosis; 12.3±7.9% after release on day 1; and 13.8±6.6%, 15.3±6.9%, 18.9±6.5%, 21±6.9%, 23.8±6.3%, and 24.2±7.2% on days 2 through 7, respectively. In 1 pig, regional wall thickening improved only from 0% to 8.5% by transthoracic echocardiography during the first week of reperfusion. This pig was allowed to survive for another 3 weeks, by which time regional wall thickening was 23%, or 70% of the baseline wall thickening of 32%. No infarction was detected in this pig.
This study demonstrates that myocardial hibernation with a sustained moderate (43% on average) reduction in resting regional coronary flow and metabolic adaptation can be maintained in pigs for 24 hours without extensive infarction. In this model, an acute, moderate reduction in coronary flow initially produced acute myocardial ischemia, as evidenced by reduced regional wall thickening, regional lactate production, and increased regional coronary venous H+ concentration. Despite persistence of the reduction in coronary flow and regional wall thickening at 24 hours and no change in the rate-pressure product, which is an index of external myocardial oxygen demand, ischemic lactate production ceased and regional H+ concentration normalized, with a tendency toward further reduction in regional oxygen consumption. This pattern indicates sustained metabolic adaptation of the hypoperfused segment at 24 hours. This metabolic adaptation was associated with ultrastructural myocardial changes, characterized by a partial replacement of myofibrils or sarcomeres with numerous mitochondria and glycogen deposits. These functional and morphological changes recovered almost completely, usually within 1 week after reperfusion of the hibernating region. This animal model demonstrates that acutely ischemic myocardium with a moderate flow reduction undergoes sequential metabolic and morphological alterations and preserves the ability to recover both functionally and morphologically after revascularization.
Critique of the Model of Myocardial Hibernation
Hydraulic occluders are widely used to create coronary stenoses, with the advantage that the severity of the stenosis is adjustable. However, a high mortality caused by abrupt occlusion was observed in this study and has been reported by others using pigs.12 Hydraulic occluders also may leak or rupture during long-term studies. Therefore, we used silk ties and a calibrated external needle to create stenoses (see the “Methods” section) in half of the pigs. In experienced hands, this technique yields a success rate similar to hydraulic occluders but avoids the problems of leakage or rupture. However, the silk tie technique could potentially induce myocardial stunning by temporarily occluding the artery when the external needle is tied against the vessel. Kloner and Przyklenk24 demonstrated that persistent abnormalities in systolic function occur only after ischemic episodes of 5 minutes or longer. In this study, the temporary occlusion was very brief, <10 seconds, or equal to the occlusion time used to test the hyperemic response. This technique should therefore not cause myocardial stunning. This conclusion is supported by the fact that anterior wall thickening in pigs with stenoses created by silk ties or hydraulic occluders was not different for similar degrees of flow reduction. The brevity of the occlusion also makes it unlikely to induce preconditioning.20
Hibernation Versus Stunning
Most authors define hibernation as myocardial dysfunction resulting from prolonged hypoperfusion1 2 5 19 and myocardial stunning as myocardial dysfunction persisting after restoration of coronary flow (postischemic dysfunction).3 5 19 24 Therefore, determining the level of myocardial perfusion associated with myocardial dysfunction is a key to differentiating hibernation from stunning. That coronary flow and segmental function were reduced to similar degrees at 15 and 60 minutes and 24 hours suggests that hypoperfusion in the model is stable over this time period. However, a major limitation of this study is that coronary flow was not monitored between 60 minutes and 24 hours in 8 of the 11 pigs. The reduced regional coronary flow is not likely to have returned to normal during this interval because the stenoses were severe and fixed. However, it is possible that myocardial oxygen demand may have increased during the recovery or waking period, causing repeated episodes of demand ischemia superimposed on the existing reduction of coronary flow. But regional wall thickening at 24 hours was unchanged compared with that at 15 minutes after placement of the stenosis, suggesting that if any myocardial stunning had occurred during this interval, it was not severe enough to produce prolonged deterioration of regional function. Furthermore, coronary flow was monitored continuously throughout the 24 hours in 3 pigs, and no instances of total coronary occlusion were observed, eliminating the possibility that myocardial stunning is an important feature of this model. Although regional coronary flow varied with changes in blood pressure and heart rate (Fig 2⇑), flow always remained below the normal baseline level over the 24 hours, indicating sustained hypoperfusion in the dysfunctional LAD regions.
Regional LV dysfunction in this model could not result from surgical manipulation of the coronary artery because we performed identical surgical procedures, including isolating the artery and testing the hyperemic response but without creating a stenosis, in a group of 5 pigs in a previous study without reducing regional wall thickening.20 Also, after the LAD was isolated but before the stenosis was created in this study, wall thickening in the anterior and inferior regions was not different, precluding an effect from surgical manipulation on regional myocardial function.
Thus, myocardial dysfunction in this model is caused by persistent hypoperfusion and represents myocardial hibernation. However, some ischemia caused by increased oxygen demand may have occurred during recovery in some of the pigs, superimposing a degree of ischemic and postischemic dysfunction (stunning) on the hibernating myocardium. Such a situation may reflect what happens clinically; myocardial hibernation may not always be a pure phenomenon.
Minimal Myocardial Necrosis
Small areas of patchy myocardial necrosis were detected in 5 of the 11 pigs in this study. In our previous study,20 similar small areas of patchy infarction were observed after 90 minutes in the same model of myocardial hibernation. The amount of myocardial necrosis does not appear to increase from 90 minutes to 24 hours, suggesting that the model is stable for at least this interval and that any ischemia that may have occurred during recovery caused no irreversible damage.
Flow Reserve Versus Change in Perfusion Pressure
As Fig 2⇑ shows, variations in regional coronary flow were seen despite the fixed coronary stenosis: higher flows were associated with higher rate-pressure products and higher perfusion pressures. Therefore, it is unlikely that regional flow variations were caused by changes in the stenosis itself, either loosening of the occluder or silk ties or intermittent spasm or thrombosis of the stenosis. The variations in flow probably reflect changes in perfusion pressure, oxygen demand, or both. This study was not designed to determine whether a limited flow reserve is present that can be recruited by increased oxygen demand in hypoperfused hibernating myocardium if perfusion pressure and the stenosis are kept constant. Previous studies have demonstrated that a limited flow reserve is present in coronary stenoses with moderately reduced resting flow25 26 ; however, this reserve cannot be recruited by pacing-induced ischemia in a short-term setting.26
Differences between the subendocardium and subepicardium could not be evaluated in this study. The regional flow reductions represent a transmural average, and the small areas of patchy necrosis were confined to the subendocardium, as expected. Because the morphological assessment was performed at only two time points, 24 hours and 7 days, the precise timing of the development of morphological abnormalities and their recovery cannot be determined. At least 5 of the 7 pigs recovered within 1 week; 1 died before recovering within the first day of reperfusion; and 1 recovered only after 4 weeks.
The concept of myocardial hibernation originated from clinical observations,2 and evidence for its existence has come primarily from clinical studies.5 8 19 However, the interpretation of clinical data has been controversial because few studies have provided simultaneous myocardial perfusion and functional measurements.7 10 11 27 In one report, resting coronary flow, including collaterals, was normal or only minimally reduced in dysfunctional myocardium with preserved metabolic activity as determined by PET.7 In a more recent study of patients with severe coronary disease, regional LV dysfunction, which recovered after revascularization, was associated with an average reduction of 34% in regional flow as measured by PET.8 However, serial measurements of LV function and perfusion were not performed.
Although short-term animal studies support the hy-pothesis of hibernating myocardium with downregula-tion,15 16 17 18 20 28 data from long-term animal studies are limited.5 12 13 14 Acute animal models with severe coronary stenoses have been maintained for 60 minutes to 6 hours with reduced regional coronary flow and regional wall thickening, with minimal infarction in pigs and dogs.15 16 17 18 20 25 These experiments have been considered to be consistent with short-term myocardial hibernation with functional and metabolic downregulation. Fedele et al15 and Pantely et al16 have shown that metabolic recovery of the ischemic myocardial region with regeneration of creatinine phosphate and cessation of lactate production and regional acidosis occurs within 60 minutes to 3 hours. Active downregulation with metabolic preservation has been demonstrated in myocardium hibernating for 60 minutes in pigs.28
Inotropic stimulation of short-term hibernating myocardium induces ischemic metabolism with lactate production,18 20 leading to infarction if the stimulation is prolonged for 90 minutes.29 Because sympathetic stimulation commonly occurs during normal activities in animals and patients, whether hibernating myocardium can be maintained over days or weeks without significant myocardial necrosis is unclear.
A porcine model of coronary stenosis sustained for 4 to 7 days was reported by Bolukoglu et al12 and Liedtke et al.13 Their coronary stenosis reduced the hyperemic response by 50% without a reduction in mean resting coronary flow.13 Wall thickening initially remained normal, implying that resting coronary flow was not reduced,30 but a reduction of regional wall thickening was observed after 7 days with the stenosis. Whether this model represents repeated myocardial stunning or hibernation is debatable.13 Shen and Vatner14 used an amaroid constrictor in minipigs to produce a severe coronary stenosis over several days, allowing collaterals to develop. Measurements obtained after 20 days revealed severe regional dysfunction but normal resting coronary flow, including collateral flow, a finding more consistent with stunning than with hibernation. Mills et al31 used a similar pig model to study the coronary vascular response to a chronic (4- to 32-week) reduction in perfusion pressure and flow. Direct comparisons with other studies is difficult because coronary flow was not measured early after placement of the stenosis and no serial functional evaluations or myocardial ultrastructural assessments were reported.
Our model of myocardial hibernation demonstrates that a persistent moderate flow reduction and regional LV dysfunction can be maintained for at least 24 hours without extensive infarction and with functional recovery after reperfusion. This is an important step toward establishing a long-term model of hibernation in which these conditions are sustained. In fact, we currently are extending the observation period of this model by maintaining the stenosis to 7 days; preliminary results indicate that hypoperfusion and regional LV dysfunction can be sustained without extensive infarction.32
Dynamic Metabolic Adaptations and Morphological Changes
This study showed that the metabolic adaptation of short-term myocardial hibernation occurs within 60 minutes and that this metabolic adaptation persists with further recovery of ischemic metabolism, such as lactate production and regional acidosis, over 24 hours. Minimal patchy myocardial necrosis was observed in 5 of 11 pigs and was not related to functional and metabolic changes. Significant ultrastructural changes, including loss of myofilaments and sarcomeres, were associated with the functional and metabolic adaptations of hibernating myocardium. The remaining intracellular spaces were partially filled by mitochondria. Whether the loss of myofilaments is quantitatively related to contractile dysfunction in hibernating myocardium cannot be determined from this study because contractile protein loss was not measured quantitatively.
Similar morphological changes have been observed in myocardium with regional dysfunction distal to chronic coronary stenoses in patients without histories of myocardial infarction7 8 ; however, significant fibrosis is noted in such patients but not in our pigs at 24 hours. The morphological changes of myocardial hibernation in the first few hours appear to be different from those at 24 hours. Specifically, no loss of contractile material was noted at 3 hours,33 whereas myofilament loss was obvious after 24 hours of myocardial hibernation. The timing and mechanism of clearance of myofilaments cannot be determined from this study. Because myocardial proteins have a variable half-time, ranging from a few hours to several days, the loss of myofilaments may not have peaked and may have been more pronounced if the stenosis had been maintained longer and the synthesis of new proteins had been slowed progressively by an inadequate oxygen supply. Although there was a tendency toward a further decrease in oxygen consumption during the 24 hours of myocardial hibernation, it is not clear whether basal oxygen consumption is reduced in hibernating myocardium. Dysfunctional but viable myocardium has been reported to show increased oxygen consumption compared with its basal level in stunned myocardium.34
A persistently high glucose consumption of regional hibernating myocardium over 24 hours was demonstrated in this study and is consistent with the PET study in patients with chronic hypoperfused myocardium and acute ischemic myocardium.35 The mechanism accounting for this high glucose consumption is not completely understood and cannot be ascertained from this study. The glycogen deposition appeared to be variable on the electron micrographs (Figs 3B and 4⇑⇑) both in different pigs and in different myocytes for the same pigs. A quantitative evaluation of glycogen was not attempted in this study.
Clinical and Future Research Implications
It is important to learn from an animal model how long hibernating myocardium can be maintained without the development of extensive necrosis or fibrosis because this information may be relevant to the clinical setting in which patients with unstable angina or non–Q-wave infarction may have severe regional wall motion abnormalities caused by a severe coronary stenosis with reduction of resting coronary flow. This study suggests that myocardial infarction is not inevitable when a persistent coronary flow reduction induces severe regional LV dysfunction. Further studies are needed to determine the stability and fate of hibernating myocardium, investigate interventions that could prevent hibernation or preserve myocardium during hibernation, and define the parameters that regulate myofibrillar and sarcomere loss and replacement during hibernation and recovery. The course of functional recovery may be dependent on the time required to regenerate contractile materials (myofilaments) or structure (sarcomeres) after revascularization.
This work was supported by a grant from the Beatrice Fox Auerbach Foundation and a grant from the Hartford Hospital Research Fund. We acknowledge the excellent technical assistance provided by Edward Hall, William Dyckman, and Luis Guerrero. We appreciate Drs Richard Cartun and Zhemin Li for their help in preparing histological slides, Dr Robert Levine for critical discussion, and John B. Newell and Jeffrey Mathew for their statistical expertise.
Selected Abbreviations and Acronyms
|LAD||=||left anterior descending coronary artery|
|LV||=||left ventricle/left ventricular|
|PET||=||positron emission tomography|
|TTC||=||triphenyl tetrazolium chloride|
Presented in part at the 67th Scientific Sessions of the American Heart Association, Dallas, Tex, November 14-17, 1994, and previously published in abstract form (Circulation. 1994;90[pt 2]:I-369).
- Received October 5, 1995.
- Revision received January 3, 1996.
- Accepted January 4, 1996.
- Copyright © 1996 by American Heart Association
Rahimtoola SH. A perspective on the three large multicenter randomized clinical trials of coronary bypass surgery for chronic stable angina. Circulation. 1985;72(suppl V):V-123-V-135.
Braunwald E, Kloner RA. The stunned myocardium: prolonged, postischemic ventricular dysfunction. Circulation. 1982;66:1146-1149.
Bolli R. Mechanism of myocardial ‘stunning.’ Circulation. 1990;82:723-738.
Ross J Jr. Myocardial perfusion-contraction matching: implications for coronary heart disease and hibernation. Circulation. 1991;83:1076-1083.
Cigarroa CG, deFilippi CR, Brickner E, Alvarez LG, Wait MA, Grayburn PA. Dobutamine stress echocardiography identifies hibernating myocardium and predicts recovery of left ventricular function after coronary revascularization. Circulation. 1993;88:430-436.
Vanoverschelde JL, Wijns W, Depre C, Essamri B, Heyndrickx GR, Borgers M, Bol A, Melin JA. Mechanisms of chronic regional postischemic dysfunction in humans: new insights from the study of noninfarcted collateral-dependent myocardium. Circulation. 1993;87:1513-1523.
Maes A, Flameng W, Nuyts J, Borgers M, Shivalkar B, Ausma J, Bormans G, Schiepers C, De Roo M, Mortelmans L. Histological alterations in chronically hypoperfused myocardium: correlation with PET findings. Circulation. 1994;90:735-745.
Rahimtoola SH. Chronic myocardial hibernation. Circulation. 1994;89:1907. Letter.
Vanoverschelde JL. Reply to the letter to editor regarding chronic myocardial hibernation. Circulation. 1994;89:1907-1908.
Bolukoglu H, Liedtke AJ, Nellis SH, Eggleston AM, Subramanian R, Renstrom B. An animal model of chronic coronary stenosis resulting in hibernating myocardium. Am J Physiol. 1992;263:H20-H29.
Liedtke AJ, Renstrom B, Nellis SH, Subramanian R. Myocardial function and metabolism in pig hearts after relief from chronic partial coronary stenosis. Am J Physiol. 1994;267:H1312-H1319.
Shen YT, Vatner SF. Mechanism of impaired myocardial function during progressive coronary artery stenosis in conscious pigs: hibernating versus stunning? Circ Res. 1995;76:479-488.
Fedele FA, Gewirtz H, Capone RJ, Sharaf B, Most AS. Metabolic response to prolonged reduction of myocardial blood flow distal to a severe coronary artery stenosis. Circulation. 1988;78:729-735.
Pantely GA, Malone SA, Rhen WS, Anselone CG, Arai A, Bristow J, Bristow JD. Regeneration of myocardial phosphocreatine in pigs despite continued moderate ischemia. Circ Res. 1990;67:1481-1493.
Matsuzaki M, Gallagher KP, Kemper WS, White F, Ross J Jr. Sustained regional dysfunction produced by prolonged coronary stenosis: gradual recovery after reperfusion. Circulation. 1983;68:170-182.
Schulz R, Guth BD, Pieper K, Martin C, Heusch G. Recruitment of an inotropic reserve in moderately ischemic myocardium at the expense of metabolic recovery: a model of short-term hibernation. Circ Res. 1992;70:1282-1295.
Dilsizian V, Bonow RO. Current diagnostic techniques of assessing myocardial viability in patients with hibernating and stunned myocardium. Circulation. 1993;87:1-20.
Chen C, Li L, Chen LL, Vasquez de Prada J, Chen MH, Fallon JT, Weyman AE, Waters D, Gillam LD. Incremental doses of dobutamine induce a biphasic response in dysfunctional left ventricular regions subtending coronary stenosis. Circulation. 1995;92:756-766.
Schaper J, Meiser E, Stammler G. Ultrastructural morphometric analysis of myocardium from dogs, rats, hamsters, mice, and from human hearts. Circ Res. 1985;56:377-391.
Kloner RA, Przyklenk K. First evidence: postischemic dysfunction of viable myocardium. In: Kloner RA, Przyklenk K, eds. Stunned Myocardium. New York, NY: Marcel Dekker, Inc; 1993:17-40.
Pantely GA, Bristow JD, Swenson LJ, Ladley HD, Johnson WB, Anselone CG. Incomplete coronary vasodilatation during myocardial ischemia in swine. Am J Physiol. 1985;249:H638-H647.
Bristow JD, McFalls EO, Anselone CG, Pantely GA. Coronary vasodilator reserve persists despite tachycardia and myocardial ischemia. Am J Physiol. 1987;253:H422-H431.
Buxton DB. Dysfunction in collateral-dependent myocardium: hibernation or repetitive stunning? Circulation. 1993;87:1756-1758.
Arai AE, Pantely GA, Anselone CG, Bristow J, Bristow JD. Active downregulation of myocardial energy requirements during prolonged moderate ischemia in swine. Circ Res. 1991;69:1458-1469.
Schulz R, Rose J, Martin C, Brode OE, Heusch G. Development of short-term myocardial hibernation: its limitation by the severity of ischemia and inotropic stimulation. Circulation. 1993;88:684-695.
Vatner SF. Correlation between acute reduction in myocardial blood flow and function in conscious dogs. Circ Res. 1980;47:201-207.
Mills I, Fallon JT, Wrenn D, Sasken H, Gray W, Bier J, Levine D, Berman S, Gilson M, Gewirtz H. Adaptive responses of coronary circulation and myocardium to chronic reduction in perfusion pressure and flow. Am J Physiol. 1994;266:H447-H457.
Chen C, Gillam L, Chen L, Knibb D, Knight D, Waters D. Temporal hierarchy in functional and ultrastructural recoveries between short-term and chronic hibernating myocardium after reperfusion. Circulation. 1995;92(suppl I):I-552. Abstract.
Gayheart PA, Vinten-Johansen J, Johansen WE, Hester TO, Cordell RA. Oxygen requirements of the dyskinetic myocardial segment. Am J Physiol. 1989;257:H1184-H1191.