An Incomplete Adaptation to Ischemia
Background We tested the hypothesis that hibernating myocardium represents an incomplete adaptation to a reduced myocardial oxygen supply.
Methods and Results In 38 patients, areas of hibernating myocardium were identified by angiography, multigated radionuclide ventriculography, thallium scintigraphy with reinjection, and low-dose dobutamine echocardiography. Biopsies removed at cardiac surgery showed structural degeneration characterized by a reduced protein and mRNA expression and disorganization of the contractile and cytoskeletal proteins myosin, actin, desmin, titin, α-actinin, and vinculin by electron microscopy, immunohistochemistry, and in situ hybridization. Additionally, an increased amount of extracellular matrix proteins resulting in a significant degree of reparative fibrosis was present. Dedifferentiation, ie, expression of fetal proteins, was absent. Apoptosis indicating suicidal cell death was found by the terminal deoxynucleotidyl transferase end-labeling method and electron microscopy. Radionuclide ventriculography showed improvement of regional function at 3 months postoperatively compared with preoperative values (mean values, 23.5% and 48%, respectively), and the echocardiographic wall-motion score index decreased from 3.4 to 1.8. The degree of severity of the morphological changes (three stages) correlated well with the extent of postoperative functional recovery: more advanced clinical improvement was observed in patients with slight and moderate morphological degeneration (stages 1 and 2), but recovery was only partial in severe degeneration (stage 3).
Conclusions Cellular degeneration rather than adaptation is present in hibernating myocardium. The consequence is progressive diminution of the chance for complete structural and functional recovery after restoration of blood flow. The practical consequence from this study should be early revascularization in patients showing areas of hibernating myocardium.
The term “hibernating myocardium” refers to the presence of persistently impaired LV function at rest, due to a reduced coronary blood flow that can be partially or completely restored to normal after revascularization.1 2 LV dysfunction would thus result from a reduced oxygen supply that is met by a reduced oxygen demand, ie, a new steady state, in myocardial areas supplied by a stenosed or totally occluded artery. This new adaptation, as an endogeneous mechanism, would act as protection against ischemia, preserving the structural integrity of the tissue, and myocardial necrosis would be absent under these conditions.3 4
Numerous studies exist on the identification of viable myocardium by different methods used in clinical investigations such as dobutamine echocardiography,5 6 7 thallium scintigraphy,8 9 10 angiography, and a combination of all of these.5 6 7 11 12 Several reports from Borgers and his group13 14 15 16 are concerned with the structural alterations in the areas afflicted and their relation to clinical measurements, and it was concluded that dedifferentiation but not degeneration of myocytes occurs in hibernating myocardium.15
Our previous studies in human myocardium17 18 19 showed that the ultrastructural appearance and the pattern of protein localization in different types of heart diseases are almost similar to those observed in hibernating myocardium. It seems that the myocardium has only a limited repertoire of reacting to noxious stimuli and that degeneration of myocytes is a common feature.
The aim of this study, therefore, was to define the major structural intracellular as well as extracellular changes of human hibernating myocardium and the possibility of cell death. These data were correlated with the clinical findings to achieve a concept about the cellular adaptive mechanisms occurring in myocardial hibernation.
Thirty-eight selected patients with coronary heart disease (37 men and 1 woman, mean age, 61±7 years) were included in the present study. Clinical data for all patients are listed in Table 1⇓.
When by angiography LV function was found to be reduced, multigated radionuclide ventriculography was carried out, and global and regional EFs were determined. 201Tl scintigraphy with reinjection was used to determine viable and ischemic myocardial tissue, and echocardiography with low-dose dobutamine was used to determine the functional capacity of the ventricular region afflicted before the indication for CABG surgery was established.
The time interval between the preoperative examinations and the revascularization was no more than 8 days.
During CABG surgery, transmural Tru-cut needle biopsies were removed from the center of the area previously diagnosed as hibernating. The patients were again studied by echocardiography at 10 to 14 days and by all methods used preoperatively 3 months after the operation, and the degree of restitution of perfusion and of functional recovery was determined (Fig 1⇓).
Patients with previous infarction (<3 months ago), with left bundle-branch block, pacemakers, incomplete revascularization, or perioperative or postoperative myocardial infarction were excluded from the study. Informed written consent from each patient for every investigation and approval of the local hospital review board had been obtained.
Dobutamine Two-dimensional Echocardiography
A Hewlett-Packard wide-angle phased-array imaging system (Sonos 1500/2.5-MHz transducer) was used for transthoracic echocardiography. The echocardiographic studies were performed at rest and during the intravenous infusion of dobutamine using a standard protocol.
Dobutamine infusions at a rate of 5 and 10 μg/kg BW−1 · min−1 were given over 10 minutes each into a peripheral vein. Before the infusion rate was increased, images of standard views (parasternal long axis, parasternal short axis at basal and midventricular level, and apical four and two chamber) were acquired and recorded on half-inch videotape (VHS).
For evaluation of regional function, the LV was divided into 16 segments according to the recommendations of the American Society of Echocardiography. The following scoring system was used to evaluate regional function of the LV: 1: normal, 2: slightly hypokinetic, 3: severely hypokinetic, 4: akinetic, and 5: dyskinetic.
All segments were scored at baseline and after infusion as responders/nonresponders to dobutamine. A wall-motion score index of the global LV and of the hibernating area was calculated for each patient by the sum of the score of segments divided by the number of segments evaluated.
An improvement of regional function in at least two adjacent abnormal segments by a factor ≥1 during dobutamine infusion at 10 to 14 days and/or 3 months after revascularization was considered as indicating hibernating myocardium.
A stress–rest–reinjection protocol was used. After an overnight fast without any medication for 12 hours, the patients underwent supine symptom-limited exercise testing on a bicycle ergometer starting at a workload of 25W and with increments of 25W every 2 minutes. 201Tl at a dose of 111 MBq was injected intravenously at peak exercise.
Myocardial SPECT acquisition was performed using a conventional circular-field single-head rotating digital gamma camera (Sopha, DS7). Imaging started immediately after the exercise. Thirty-two projections were acquired over 30 seconds, each over a 180° arc using body contour in step-and-shoot mode. The data were then reconstructed by back-projection and reoriented into vertical long-axis, short-axis, and horizontal long-axis slices using the Wiener filter for the entire LV.
Redistribution images were acquired 4 hours after exercise. Delayed images at rest always were taken under optimal individual medication. Immediately after redistribution imaging, a 37-MBq additional thallium dose was administered at rest, and reinjection images were acquired 30 minutes thereafter.
Qualitative and Quantitative Analysis
The SPECT images were analyzed qualitatively in 13 sectors as well as quantitatively for four representative tomograms.
The qualitative analysis is based on the visual interpretation of regional tracer activity. Stress defects were classified on the basis of their behavior in reinjection images in totally reversible (complete redistribution), partially reversible (partial redistribution), and fixed (absent redistribution) defects in the baseline study.
For the quantitative analysis, the count activities within the myocardial sectors were expressed as percentage of the peak activity in each slice (circumferential profile analysis). From the center of the LV, 18 radii were constructed, with the segment with the highest activity taken as 100%. The segmental activities were expressed as percent of the segment with the highest activity. The activity of 201Tl in these three short-axis tomograms (one basal, two midventricular) and one vertical long-axis view was compared with the database of a reference population of the same sex (probability <5% for significant coronary artery disease; 46 women and 38 men investigated at our center).
Uptake on stress images below 2 SDs of the mean in this reference population was classified abnormal. Quantified segmental uptake was considered reversible when an increase of the thallium uptake of 15% or more was measured on reinjection in comparison with the stress images in the baseline study.
A region was classified as hibernating myocardium when it showed reversibility of a defect and/or an uptake value in the range of the mean normal uptake ±2 SDs in the baseline study, and hypokinetic or akinetic wall motion at rest in the baseline radionuclide ventriculography.
For identification of the perfusion area of the three main coronary vessels, three short-axis tomograms and one midventricular sagittal long-axis tomogram were chosen and subdivided into three descriptor territories with respect to the location of segments.
For every patient, the region classified as hibernating myocardium was related to one of these three descriptor territories and accordingly to one coronary vessel. The reduced perfusion was estimated by the lowest value of thallium uptake in the hibernating area after reinjection in the baseline study compared with the database of the reference population.
Three months after revascularization, an increase in the thallium uptake to a value of ≥55% of peak myocardial activity indicated normal perfusion because this was the lowest value of the mean ±2 SDs determined in the reference population.
Resting and exercise LV wall motion was assessed by multigated radionuclide ventriculography in patients in a supine position.
Using a dose of 740 MBq of 99mTc-labelled in vivo red blood cells, we positioned a small-field-of-view gamma camera (LEM, Siemens), equipped with a high-sensitivity collimator interlinked to a Sopha workstation, in a modified 45° left anterior oblique angulation with a slight caudal tilt.
Two hundred beats were collected at rest and during graded bicycle exercise starting at 25W, with 25W increments when sufficient beats were registered. Quantitative sectorial EF calculation was done by drawing nine radii to the LV border dividing the ventricle into an equal number of sectors.20
A postoperative functional improvement of at least 5% of the LVEF was the criterion for recovery of hibernating myocardium.
All patients underwent coronary angiography with multiple projections and left ventriculography in biplane view (DCI, Philips). Coronary narrowing was assessed as percent diameter stenosis.
For comparison and matching of the different methods, all LV segments were grouped into three vascular territories corresponding to the three main coronary vessels.
The anterior and anterolateral walls were considered as perfusion area of the left anterior descending coronary artery. The lateral LV wall was seen as perfusion area of the left circumflex artery and the inferior and posterior walls as perfusion area of the right coronary artery.
Data are presented as mean values ±SD. The Friedman and the Dunn tests (echocardiography), the paired t test (radionuclide ventriculography), and the Wilcoxon test (201Tl uptake) were used for the evaluation of differences between the different time points of the study. The Kruskal-Wallis and the Dunn tests were used for the evaluation of differences between stages 1, 2, and 3 in echocardiography as well as in radionuclide ventriculography. Fibrosis was evaluated using ANOVA and Bonferroni. A value of P<.05 was considered significant.
Two transmural biopsies were removed from each patient and either immediately frozen in liquid nitrogen for immunohistochemistry and in situ hybridization or immersed in 3% glutaraldehyde buffered with 0.1 mol/L Na cacodylate (pH 7.4, 440 mOsmol) for electron microscopy.
Small tissue samples were embedded in Epon following routine procedures. Semi-thin sections were stained with periodic acid–Schiff’s reagent (PAS) for glycogen and evaluated in the light microscope. From these, the degree of fibrosis was determined from 10 fields of vision per tissue sample by the point-counting method following stereological principles. Ultra-thin sections were stained with uranyl acetate and lead citrate and viewed and photographed in a Philips CM 10 electron microscope.
Histology and Immunohistochemistry
The following antibodies were used: anti-α-actinin, desmin, vinculin, collagen IV (Sigma); collagens I and III (Bioscience); collagen VI (Telios/Biomol); fibronectin (ICN Biomedicals); laminin and vimentin (Dianova); titin T12 (Boehringer/Mannheim); and phalloidin (Sigma) for the staining of actin. The myosin antibody was a generous gift from Dr Decker, Chicago, IL.
The tissue samples were mounted with Tissue Tek (Sakura Fine-tec) and cryosections were air-dried and fixed with acetone at −20°C for 10 minutes. Incubation with the first antibody for 60 minutes was followed by treatment with the biotinylated second antibody for 60 minutes. The last incubation was carried out with fluorescein isothiocyanate (FITC)–linked streptavidin (Amersham). Nuclei were stained with actinomycin D (Molecular Probes) diluted 1:100. Frequent rinsing with PBS was done between all steps. The sections were covered with Mowiol (Hoechst), coverslipped, and viewed in a Leica Aristoplan microscope with fluorescence equipment or in a confocal laser microscope (Leica). Documentation was carried out on professional Kodak Ektachrome 100 HC film for color slides. All reproductions were made from slides.
The probe for fibronectin was a 3.5-kb HindIII–Xba I or a 360-bp Sac I–Bgl II fragment of clone pRCabFN1 (generous gift of Dr K. Boheler, London, England). The probes for human collagen-α1(I) (Hf677) and for human actinin-1 (α-actinin; HFBCY25) were bought from American Type Tissue Collection. For in situ hybridization analysis, a 1.5-kb EcoRI subclone of collagen-α1(I) and a 1.7-kb EcoRI fragment of human actinin-1 were used. The probe for myosin was a 156-bp DNA fragment designated pBS αβ and donated by Dr K. Schwartz, Paris, France.
The probes were tested in conventional Northern blot analysis for specificity. Labelled sense and antisense RNA probes were generated by in vitro transcription using a kit (Promega) and 100 μCi of [α-35S]UTP (1000 Ci/mmol). Usually, 80% to 90% of incorporation and about 250 ng of labelled probe were obtained. An aliquot was then routinely checked on a denaturing 1% agarose gel containing 0.66 mol/L formaldehyde followed by digestion with RNase-free DNase and partial hydrolysis to generate fragments of 200-bp average length. Labelled RNAs were stored in 0.1 mol/L dithiothreitol (DTT) at −80°C until further use.
In Situ Hybridization Analysis
Cryostat sections (4 μm) were placed on glass slides coated with 3-aminopropyltriethoxysilane (Sigma) and fixed with 4% paraformaldehyde for 10 minutes. After the sections were washed, they were dehydrated through graded ethanol, dried, and used immediately according to Simmons et al.21
Each section was hybridized overnight in a humidified chamber at 50°C with 60 μL of hybridization buffer including about 2.5×106 cpm of denatured antisense or sense cRNA probe (5 ng). After several washes with increasing stringency,22 dehydrated slides were exposed to Kodak NTB-2 emulsion for up to 2 weeks. Analysis was done after slides were counterstained with 0.1% toluidine blue.
Control hybridizations include the use of serial sections of the same tissue under identical conditions for hybridization with a labelled sense RNA probe to check probe specificity and the pretreatment of sections with RNase A (Sigma) prior to hybridization to test for the nonspecificity of the probe.
In Situ Detection of Apoptotic Cells
In situ detection of apoptosis was performed using the ApoTag In Situ Apoptosis Detection Kit (Oncor Inc) with some modifications. In brief, cryosections were fixed in 4% paraformaldehyde for 20 minutes. Endogenous peroxidase was blocked with 0.75% H2O2. The sections were postfixed in ethanol/acetic acid (3:1) at −20°C for 10 minutes. Nuclear proteins were dissolved by incubation with 2 μg/mL proteinase K (Sigma) for 15 minutes. The labeling procedure using a mixture of terminal deoxynucleotidyl transferase (TdT) and reaction buffer containing digoxigenin-labeled dUTP was carried out according to the kit instructions.
DAB staining was used to visualize labeled apoptotic nuclei (brown). Sections were counterstained with methyl green. Control sections were incubated in the absence of the TdT enzyme. For a positive control, sections were digested with 1 μg DNase-I/mL (Sigma) in DNase buffer (30 mmol/L Trizma base, pH 7.2; 140 mmol/L sodium cacodylate; 4 mmol/L magnesium chloride; 1 mmol/L DTT) for 10 minutes. All other steps of the procedure were the same as described above.
The mean value of the wall-motion score index for hibernating myocardium regions decreased from 3.4±0.4 at baseline to 1.8±0.6 during dobutamine infusion. Immediately after revascularization, regional function had recovered only partially (2.9±0.6 at 10 to 14 days). Segmental function was significantly improved at 3 months (1.8±0.6).
A good correlation was observed between the degree of functional recovery and the severity of morphological degeneration. Stage 1: 4.0±0.5 at baseline, 1.6±0.8 during dobutamine infusion, 2.9±0.6 at 10 to 14 days after revascularization, and 1.4±0.3 at 3 months postoperatively. Stage 2: 3.4±0.4 at baseline, 1.7±0.6 during dobutamine infusion, 2.9±0.7 at 10 to 14 days, and 1.9±0.6 at 3 months after revascularization. Stage 3: 3.9±0.1 at baseline, 2.6±0.5 during dobutamine stimulation, 2.9±0.4 at 10 to 14 days, and 2.8±0.1 at 3 months postoperatively (Fig 2⇓ and Table 2⇓).
Qualitatively, visual analysis showed 346 of 494 analyzed segments (70%) to be abnormal after stress. On 4-hour images and after reinjection, 80 segments (23%) showed complete redistribution, 128 (37%) showed incomplete redistribution, and 138 (40%) showed a fixed defect in the baseline study.
The effect of revascularization was assessed by comparing prerevascularization and postrevascularization scans. Of 138 segments with a fixed defect before operation, 12 showed an increase of the 201Tl uptake. Of 128 segments with a partially reversible defect, 105 had an enhanced 201Tl uptake postoperatively. All 80 segments with complete redistribution showed a better 201Tl uptake 3 months after revascularization. Ischemia was absent in the postoperative SPECT images.
The quantitative evaluation of the 201Tl uptake, as an indicator of perfusion, showed in the hibernating region a significant increase from 40.53±9.9% preoperatively to 66.58±11.4% of the peak myocardial activity 3 months after revascularization (P<.001, Fig 3⇓).
Regional LVEF at rest increased from a mean value of 23.6±12.7% to 48.0±16.8% 3 months after revascularization. In correlation with the morphological alterations: in stage 1 the increase was from 27.0±11.6% to 58.8±9.9% after bypass surgery, in stage 2 from 24.2±12.8% to 45.6±16.7%. In stage 3 an improvement from 8.5±1.7% to 25.3±3.0% after 3 months postoperatively was observed (Fig 4⇓ and Table 3⇓).
Success of Revascularization
Aortocoronary bypass surgery with 3.4±0.98 venous and internal mammary artery bypass grafts per patient was performed. Three months after revascularization, all hibernating areas were revascularized.
Three different stages of severity of tissue deterioration were defined using a semiquantitative scoring system. The first stage represents slight degeneration characterized by the beginning of a loss of contractile material and slight fibrosis. In situ hybridization showed an almost normal density of the mRNA distribution, and apoptosis was never observed. Stage 2 (moderate) shows loss of myofilaments and cytoskeletal proteins, moderate fibrosis, and a slight reduction of the mRNA content. Stage 3 (severe) is characterized by a significant loss of contractile and cytoskeletal proteins, accompanied by significant fibrosis. mRNA content was significantly reduced for myosin and α-actinin, but it was increased for fibronectin and collagen I. Apoptosis of myocytes and interstitial cells was observed. Since different degrees of morphological alterations, ie, slight, moderate, and severe, were present in hibernating myocardium, it was concluded that continuous degeneration takes place.
Most of the biopsies were structurally altered, showing myocytes of different size, many of them either atrophied or hypertrophied, and an increased but varying amount of fibrosis. PAS staining revealed large intracytoplasmic areas filled with glycogen (Fig 5⇓). The degree of fibrosis as determined quantitatively correlated well with the different stages of degeneration (16.8±4.8%, 34.8±6.1%, and 57.5±7.5% for the three stages, respectively). All stages were significantly different, with a value of P<.001.
The main ultrastructural changes observed in hibernating myocardium included different size and shape of the nuclei often accompanied by chromatin clumping; mitochondrial abnormalities in size and shape; lack of contractile material; and presence of large areas containing nonspecified cytoplasm, vacuoles, lipid droplets, and large glycogen-filled regions (Fig 6⇓).
Nuclear changes indicating apoptosis accompanied by sequestration of cellular particles into the extracellular space were found in several cases (Fig 7⇓).
The extracellular space was widened and showed an augmentation of collagen fibrils and ground substance, as well as an increased number of fibroblasts and macrophages.
Actin and myosin. In contrast to normal cardiac tissue, the contractile proteins myosin and actin were reduced to different degrees in hibernating myocardium corresponding to the loss of contractile material in the electron microscope. In samples with slight alterations, the labeling was only slightly disturbed, but in severe alterations, large areas of individual cells showed significant defects.
Desmin. Desmin either was reduced or disorganized and accumulated in individual myocytes (Fig 8⇓). These changes were more pronounced with progression of the structural deterioration from stage 1 to stages 2 and 3.
α-Actinin. Disorganization and defects of different degrees increasing from stages 1 through 3 were present in hibernating myocardium (Fig 9⇓).
Titin. In hibernating myocardium, defects for titin labeling were obvious and corresponded to the lack of contractile proteins varying from slight to moderate to severe. In myocytes from stage 3 myocardium, labeling was observed only in the periphery of the cell (Fig 10⇓).
Vinculin. In hibernating myocardium, vinculin was expressed in increased amounts compared with normal tissue (Fig 11⇓).
α-Smooth muscle actin, vimentin, and natriuretic factor (ANF). In normal and hibernating myocardium, α-smooth muscle actin was found only in endothelium and smooth muscle cells of blood vessels but not in myocytes. Vimentin was absent in myocytes in hibernating myocardium. ANF, usually found only in normal atrial tissue, was absent in hibernating LV myocardium.
Fibronectin. The widened interstitial space contained large amounts of fibronectin separating the remaining myocytes from each other. The accumulation of fibronectin was more pronounced in stage 3 myocardium than in stage 1.
Collagens I, III, and VI. The amount of all types of collagen fibrils was significantly increased, ranging from slight to severe accumulation in the different stages, and corresponded to the increase in fibronectin (Fig 12⇓).
Laminin and collagen IV. Myocytes, endothelial cells, and smooth muscle cells are surrounded by these two proteins. In hibernating myocardium, thick layers of laminin and collagen IV were seen to surround the myocytes. This augmentation was more evident with more-pronounced severity of degeneration.
Vimentin. Vimentin is an indicator of fibroblasts, and their number was focally increased in hibernating myocardium corresponding to the stages of degeneration (Fig 13⇓).
In Situ Hybridization
The labeling for contractile and cytoskeletal mRNA in normal myocardium was regular and homogeneously distributed. In hibernating myocardium, the density of the label for myosin and α-actinin mRNA was greatly reduced, indicating the possibility of reduced transcriptional activity (Fig 14⇓).
The label for fibronectin and for collagen I mRNA was discretely distributed over cells of the connective tissue. In hibernating myocardium, it was significantly augmented in most areas of the interstitial space.
Detection of Apoptosis
The ultrastructural nuclear changes interpreted as apoptosis were confirmed using the terminal deoxynucleotidyl end-labeling (TUNEL) method. Apoptotic cells were found in myocytes as well as in interstitial cells in numerous biopsies (Fig 15⇓).
In the present study, hibernating myocardium was defined as viable but underperfused myocardium with an LV dysfunction at rest, a functional improvement during dobutamine infusion, and recovery of regional function 3 months after CABG surgery. The preoperative blood flow was estimated by 201Tl uptake and showed restitution to normal values 3 months after operation. For inclusion in this study, all patients were required to meet the criteria of hibernation in all examinations, ie, echocardiography, 201Tl scintigraphy, and radionuclide ventriculography.
Cellular Recovery After Revascularization
Functional recovery of the hibernating myocardium was observed in all patients in various degrees depending on the severity of the morphological degeneration as documented by echocardiography at 10 to 14 days and by all methods used preoperatively repeated at 3 months after revascularization. This can be explained by the fact that, on the cellular level, a certain time is necessary for the transcription to start and for translation to occur (in myocytes in culture, it takes at least 10 days, depending on the culture conditions, for myofibrillogenesis to occur23 ). At 3 months postoperatively, a significant functional improvement was evident. However, the functional restitution was incomplete in patients showing severe morphological alterations, indicating that reversibility is limited because of cellular degeneration.
Cellular Mechanism of Hibernation
All structural proteins of the hibernating myocardium were altered. The mRNAs corresponding to cellular proteins were reduced and those of the extracellular space increased, indicating changes at the transcriptional level, because the expression of the respective genes seems to be predominantly regulated at the transcriptional level.24 The most obvious changes in the myocytes were the loss of myofilaments, disorganization of the cytoskeleton, and the occurrence of large areas filled with glycogen. The lack of titin and α-actinin, components of the “sarcomeric skeleton,” adds to the structural disorganization. As a consequence, the loss of myofilaments causes a reduction of the contractile capacity of myocytes, the disarrangement of the cytoskeleton results in loss of cellular stability, and the defects of the “sarcomeric skeleton” will lead to sarcomere instability. In addition, the reduction in titin filaments will produce a change of compliance since titin is the “third” elastic filament of the sarcomere.25
The pathophysiological situation of the cardiomyocytes will become more aggravated by the development of fibrosis. All constituents of the basement membrane, ie, laminin, collagen IV and VI, and fibronectin, were present in large amounts, which finally will lead to an encapsulation of the myocyte. Furthermore, the matrix protein fibronectin and the fibrillar collagens I and III fill the enlarged interstitial space, and macrophages and fibroblasts are present in large numbers. Macrophages will be stimulated to phagocytose the cellular debris, and the fibroblasts produce the different extracellular matrix proteins. Fibrosis is most probably due to loss of myocytes and has to be regarded as “replacement or reparative” fibrosis, whereas “reactive” fibrosis is assumed to exist by others.14 26 We believe, however, that the combination of cellular degeneration with the development of fibrosis in hibernating myocardium will significantly determine the degree and speed of recovery after bypass operation, which is a view slightly different from that postulated by the Belgian group, who claimed that hibernating myocytes are essentially healthy cells.14 27
The Problem of Cell Death
It is interesting to note that acute ischemic cell death28 29 is absent in hibernating myocardium. Apparently, the single ischemic events occurring in these patients are seldom severe enough to cause acute irreversible mitochondrial and nuclear damage. However, typical ultrastructural signs of apoptosis30 were observed, and the TUNEL reaction31 confirmed the presence of DNA fragments in myocyte nuclei. Since the biopsies are small, a quantitative analysis is difficult, but the fact that apoptosis occurs at all is a major finding in this study.
Apoptosis was found mainly in cells that showed stage 3 alterations, ie, the final stage of cellular degeneration is characterized by the occurrence of preprogrammed cell death.
Possible causes of apoptosis could be cytokines produced by macrophages and fibroblasts, withdrawal of growth factors, heat shock proteins induced by ischemia, or disturbances in the extracellular matrix causing loss of attachment of cells. Inflammation with cellular infiltration of mononuclear cells was absent and therefore cannot be considered as stimulus for cell death in hibernating myocardium.
Expression of Fetal Genes
Neither vimentin α-smooth muscle actin, nor ANF was found in normal or hibernating myocardium. This is in contrast to Ausma et al15 and Borgers et al,32 who found expression of α-smooth muscle actin and cardiotonin. In addition, these authors found titin in a punctate labeling pattern, which was interpreted as fetal expression. However, in isolated cultured adult cardiomyocytes, a punctate labeling pattern for titin is found not only during dedifferentiation but also in redifferentiation and during the final stage of degeneration before the cells die of apoptosis (our results). Therefore, the altered titin expression in hibernating myocardium can be explained by any of these pathomechanisms and is not exclusive for dedifferentiation.
Pathomechanism of Hibernation
As described above, hibernating myocardium is characterized by a reduction of the contractile apparatus and of the cytoskeleton, and by an increase in glycogen and degeneration of mitochondria that is associated with structural abnormalities reminiscent of cell dedifferentiation13 27 but that resembles even more closely cellular degeneration as described in the present study. In addition, it was suggested that whereas stunning implies an increased intracellular calcium content, hibernating might be a kind of low demand–low supply situation with a low intracellular calcium level.33 It can be imagined that in an initial stage, the contractility of the myocytes is suppressed because “stunning” occurs and is repeated, which may reduce the Ca2+ sensitivity of myofibrils33 34 and/or reduce the storage capacity for Ca2+ of the sarcoplasmic reticulum. This may lead to an “atrophy-of-inactivity” of the sarcomeres and would be one of several explanations for the loss of contractile material, a situation comparable to the “unloading” of myocytes in culture.
Since it must be assumed that patients with hibernating myocardium have experienced multiple episodes of ischemia, it can be inferred that molecular events, known to occur in singular ischemic events, accumulate. One of the known effects of severe reversible ischemia is the translocation of the glycolytic enzyme glyceraldehyde phosphate dehydrogenase (GAPDH) from the cytosol to the myofibrils where it decorates the myosin filaments.35 These become nonfunctional and will be degraded after binding to the ubiquitin complex. This could be another factor causing the loss of contractile filaments. It is interesting to speculate why the loss of contractile material is not counteracted by increased transcription. This may be due partially to lack of energy (ATP is needed to phosphorylate the bases of DNA) but also to lack of endocrine signals (tissue thyroxin T3, a signal necessary for myosin transcription, is reduced by the low blood flow). Furthermore, translation may be hampered by the markedly increased glycogen, which is known to form complexes with RNA.36
Glycogen storage may have been caused further by the continued loss of a key glycolytic enzyme (GAPDH), which forces glucose to enter the glycogen synthesis pathway.
Oversupply with free fatty acids, as it occurs with repeated ischemia via local or systemic sympathetic stimulation, may also compete with glucose as a substrate, thereby contributing to glycogen storage.
Since in the course of the degenerative process cellular particles are sequestered from the myocytes, these will atrophy. This reduction in myocyte size will reduce the demand for oxygen of individual cells and is interpreted as a protective mechanism aimed primarily at the survival of the cell but not at the maintenance of its function.
It is imaginable that a new balance between supply and demand will be established under these conditions. This delicate balance, however, can easily and immediately be disturbed by an increased-demand situation or by the progression of coronary artery stenosis. In this way, a new cycle is started that will produce more cellular degeneration until apoptosis occurs. Atrophy and death of myocytes are followed by replacement fibrosis that will further compromise myocyte function.
A summary of our hypothesis regarding the correlation between structure and function in hibernating myocardium and its pathophysiological consequences is illustrated in Fig 16⇓.
We conclude that hibernating myocardium is not completely adapted to chronic underperfusion: cellular degeneration and myocyte loss accompanied by reparative fibrosis occur, and the structural integrity of the myocardium deteriorates. The practical consequence of this finding is the recommendation that patients with hibernating myocardium should undergo CABG surgery without delay.
- Received March 11, 1997.
- Revision received May 29, 1997.
- Accepted June 5, 1997.
- Copyright © 1997 by American Heart Association
Rahimtoola SH. The hibernating myocardium in ischaemia and congestive heart failure. Eur Heart J.. 1993;14:212-220.
Ferrari R, La Canna G, Giubbini R, Milan E, Ceconi C, de Giuli F, Berra P, Alfieri O, Visioli O. Left ventricular dysfunction due to stunning and hibernation in patients. Cardiovasc Drugs Ther.. 1994;8:371-380.
Afridi I, Kleiman NS, Raizner AE, Zoghbi WA. Dobutamine echocardiography in myocardial hibernation: optimal dose and accuracy in predicting recovery of ventricular function after coronary angioplasty. Circulation.. 1995;91:663-670.
Cigarroa CG, deFilippi CR, Brickner ME, 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.
Dilsizian V, Perrone-Filardi P, Arrighi JA, Bacharach SL, Quyyumi AA, Freedman NM, Bonow RO. Concordance and discordance between stress-redistribution-reinjection and rest-redistribution thallium imaging for assessing viable myocardium: comparison with metabolic activity by positron emission tomography. Circulation.. 1993;88:941-952.
Grandin C, Wijns W, Melin JA, Bol A, Robert AR, Heyndrickx GR, Michel C, Vanoverschelde JL. Delineation of myocardial viability with PET. J Nucl Med.. 1995;36:1543-1552.
Dilsizian V, Bonow RO. Current diagnostic techniques of assessing myocardial viability. Circulation.. 1993;87:1-20.
Dilsizian V, Arrighi JA, Diodati JG, Quyyumi AA, Alavi K, Bacharach SL, Marin-Neto JA, Katsiyiannis PT, Bonow RO. Myocardial viability in patients with chronic coronary artery disease: comparison of 99mTc-sestamibi with thallium reinjection and [18F]fluorodeoxyglucose. Circulation.. 1994;89:578-587.
Vanoverschelde J-LJ, Wijns W, Depré 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.
Ausma J, Schaart G, Thone F, Shivalkar B, Flameng W, Depre C, Vanoverschelde JL, Ramaekers F, Borgers M. Chronic ischemic viable myocardium in man: aspects of dedifferentiation. Cardiovasc Pathol.. 1995;4:29-37.
Schaper J, Froede R, Hein S, Buck A, Hashizume H, Speiser B, Friedl A, Bleese N. Impairment of the myocardial ultrastructure and changes of the cytoskeleton in dilated cardiomyopathy. Circulation.. 1991;83:504-514.
Schaper J, Hein S, Speiser B, Froede R, Bleese N. Ultrastructural changes, the cytoskeleton and the extracellular matrix in chronic ischemia. J Mol Cell Cardiol. 1992;24(suppl I):S-07-1. Abstract.
Simmons DM, Arriza JL, Swanson JW. A complete protocol for in situ hybridization of messenger RNAs in brain and other tissues with radiolabeled single-stranded RNA probes. J Histotechnol.. 1989;12:169-181.
Hein S, Münkel B, Schäfer G, Schaper J. A confocal laser microscopy study of dedifferentiation and redifferentiation of cultured adult rat myocytes. Circulation. 1995;92(suppl I):I887. Abstract.
Boheler KR, Chassagne C, Martin X, Wisnewsky C, Schwartz K. Cardiac expressions of alpha- and beta-myosin heavy chains and sarcomeric alpha-actins are regulated through transcriptional mechanisms. J Biol Chem.. 1992;267:12979-12985.
Wang K, McCarter R, Wright J, Beverly J, Ramirez-Mitchell R. Regulation of skeletal muscle stiffness and elasticity by titin isoforms: a test of the segmental extension model of resting tension. Proc Natl Acad Sci U S A.. 1991;88:7101-7105.
Vanoverschelde J-L, Depré C, Wijns M, Bol A, Dion R, Gerber B, Borgers M, Melin J. Physiopathologie de l’hibernation chronique: apports de la tomographie par emissions positrons. Médicine/Sciences.. 1995;11:1315-1322.
Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol.. 1992;119:493-501.
Bakker A, Thone F, Borgers M, Jacob W. Contact sites between the inner and outer mitochondrial membranes in stunned versus hibernating myocardium. Cardiovasc Pathol.. 1995;4:195-202.
Barbato R, Menabò R, Dainese P, Carafoli E, Schiaffino S, Lisa FD. Binding of cytosolic proteins to myofibrils in ischemic rat hearts. Circ Res.. 1996;78:821-828.