Chronic Myocardial Hibernation in Humans
1997; Lippincott Williams & Wilkins; Volume: 95; Issue: 7 Linguagem: Inglês
10.1161/01.cir.95.7.1961
ISSN1524-4539
AutoresJean‐Louis Vanoverschelde, William Wijns, Μ. Borgers, Guy R. Heyndrickx, Christophe Depré, Willem Flameng, Jacques Melin,
Tópico(s)Heart Rate Variability and Autonomic Control
ResumoHomeCirculationVol. 95, No. 7Chronic Myocardial Hibernation in Humans Free AccessResearch ArticleDownload EPUBAboutView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticleDownload EPUBChronic Myocardial Hibernation in HumansFrom Bedside to Bench Jean-Louis J. Vanoverschelde, William Wijns, Marcel Borgers, Guy Heyndrickx, Christophe Depré, Willem Flameng and Jacques A. Melin Jean-Louis J. VanoverscheldeJean-Louis J. Vanoverschelde the Division of Cardiology, University of Louvain, Brussels (J.-L.J.V., C.D., J.A.M.); the Cardiovascular Center, Aalst (W.W., G.H.); the Janssen Research Foundation, Beerse (M.B.); and the Department of Cardiovascular Surgery, Katholieke Universiteit Leuven (W.F.), Belgium. , William WijnsWilliam Wijns the Division of Cardiology, University of Louvain, Brussels (J.-L.J.V., C.D., J.A.M.); the Cardiovascular Center, Aalst (W.W., G.H.); the Janssen Research Foundation, Beerse (M.B.); and the Department of Cardiovascular Surgery, Katholieke Universiteit Leuven (W.F.), Belgium. , Marcel BorgersMarcel Borgers the Division of Cardiology, University of Louvain, Brussels (J.-L.J.V., C.D., J.A.M.); the Cardiovascular Center, Aalst (W.W., G.H.); the Janssen Research Foundation, Beerse (M.B.); and the Department of Cardiovascular Surgery, Katholieke Universiteit Leuven (W.F.), Belgium. , Guy HeyndrickxGuy Heyndrickx the Division of Cardiology, University of Louvain, Brussels (J.-L.J.V., C.D., J.A.M.); the Cardiovascular Center, Aalst (W.W., G.H.); the Janssen Research Foundation, Beerse (M.B.); and the Department of Cardiovascular Surgery, Katholieke Universiteit Leuven (W.F.), Belgium. , Christophe DepréChristophe Depré the Division of Cardiology, University of Louvain, Brussels (J.-L.J.V., C.D., J.A.M.); the Cardiovascular Center, Aalst (W.W., G.H.); the Janssen Research Foundation, Beerse (M.B.); and the Department of Cardiovascular Surgery, Katholieke Universiteit Leuven (W.F.), Belgium. , Willem FlamengWillem Flameng the Division of Cardiology, University of Louvain, Brussels (J.-L.J.V., C.D., J.A.M.); the Cardiovascular Center, Aalst (W.W., G.H.); the Janssen Research Foundation, Beerse (M.B.); and the Department of Cardiovascular Surgery, Katholieke Universiteit Leuven (W.F.), Belgium. and Jacques A. MelinJacques A. Melin the Division of Cardiology, University of Louvain, Brussels (J.-L.J.V., C.D., J.A.M.); the Cardiovascular Center, Aalst (W.W., G.H.); the Janssen Research Foundation, Beerse (M.B.); and the Department of Cardiovascular Surgery, Katholieke Universiteit Leuven (W.F.), Belgium. Originally published1 Apr 1997https://doi.org/10.1161/01.CIR.95.7.1961Circulation. 1997;95:1961–1971Since the pioneering work of Tennant and Wiggers,1 it has been known that total ischemia leads to a prompt cessation of contraction and eventually results in the appearance of cell damage and irreversible myocardial necrosis. Accordingly, in the minds of many cardiologists, the discovery of an abnormal regional contraction in a patient with coronary artery disease has long been equated with the presence of irreversible myocardial necrosis. With the advent of recanalization therapy, however, evidence progressively accumulated that prolonged regional "ischemic" dysfunction did not always arise from irreversible tissue damage and, to some extent, could be reversed by restoration of blood flow.2345 These observations have led to the speculation that chronically hypoperfused myocardium, which is often referred to as "hibernating,"23456 could maintain viability by simply reducing its metabolic demand to match the decreased supply for as long as myocardial perfusion was inadequate. The chronic impairment of contractile function in this setting has been regarded as a protective mechanism by which the heart spontaneously downgrades its myocardial function, minimizes its energy requirements, and prevents the appearance of irreversible tissue damage.245 The concept of chronic hibernation thus consists of two parts: a unique clinical observation that bears important implications for the management of patients with chronic coronary artery disease23456 and a pathophysiological hypothesis, yet to be demonstrated, implying that the chronic dysfunction is due to a chronic reduction of resting MBF.45 Other aspects that were not included in the original description, ie, the rapidity of mechanical recovery after successful revascularization7 and the response of dysfunctional myocardium to inotropic stimulation, are now also considered to be integral parts of this condition. The purpose of this discussion is to review some of the more recent advances to the understanding of the pathophysiology of chronic myocardial hibernation in humans. Emphasis will be placed on regional perfusion-contraction matching in both the experimental and the clinical settings, on the peculiar morphological changes that have been shown to occur in the hibernating myocardium, on the determinants of mechanical reversibility on restoration of adequate coronary patency, and on the presence of recruitable inotropic reserve.Perfusion-Contraction Matching in Myocardial HibernationThe term "hibernation" was first used by Diamond et al6 in 1978 to describe the chronic wall motion abnormalities of patients with coronary artery disease but no previous myocardial infarction and their reversibility after revascularization, and it was subsequently popularized by Rahimtoola45 and Braunwald and Rutherford.2 In his 1989 description of the syndrome, Rahimtoola4 postulated that this peculiar condition resulted from the "relatively uncommon response to reduced MBF at rest whereby the heart downgrades its myocardial function to the extent that blood flow and function are once again in equilibrium, and as a result, neither myocardial necrosis nor ischemic symptoms are present." The definition of myocardial hibernation, as formulated by Rahimtoola, thus implies that (1) the heart can spontaneously adapt to chronic underperfusion (the "smart heart" hypothesis), (2) a new steady state between perfusion and contraction can be reached, and (3) this new equilibrium can be maintained for a prolonged period of time. This definition raises two important, albeit conceptually different, questions: the first is whether the heart can adapt to prolonged periods of underperfusion while avoiding necrosis; the second, and more clinically relevant, is whether chronic left ventricular ischemic dysfunction in humans represents such an adaptive response to a chronic reduction of resting MBF. To answer the first question necessitates development of animal models of sustained perfusion-contraction matching and demonstration that it can be perpetuated over the long term.8 To answer the second question requires simultaneous assessment of perfusion and contraction directly in patients with hibernating myocardium.Flow-Function Relations During Partial Coronary OcclusionThe tight coupling between coronary flow, myocardial oxygen consumption, and contractile performance of the heart is a fundamental principle of cardiac physiology. Because of the small extraction reserve of oxygen, decreases in coronary blood flow rapidly translate into decreases in contractile performance.1 Several studies have examined the relation between regional MBF (radioactive microspheres) and function (sonomicrometry) in both open-chest9 and conscious1011 dogs undergoing graded reductions in coronary flow. These studies have demonstrated the existence of a close coupling between the supply of myocardial substrates, including O2, of which the measurement of regional perfusion provides a rough estimate,12 and myocardial energy demand, as reflected by the steady-state level of regional contraction. The proportional decrease in regional myocardial flow and function in this setting has been called "acute perfusion-contraction matching" and is typical of acute myocardial ischemia. Reperfusion after very short periods of low coronary flow ( 1 week) but reversible regional left ventricular ischemic dysfunction in the experimental laboratory have ended up with models of perfusion-contraction mismatch. Canty and Klocke21 examined the temporal response of regional function after ameroid implantation in conscious dogs. In their model, regional contraction was found to decrease progressively during the course of ameroid occlusion. Yet, at the time of ameroid occlusion (2 to 4 weeks), the measurements of regional endocardial blood flow showed a dissociation between flow and function. Bolukoglu et al22 and Liedtke et al23 achieved sustained reduction in segmental shortening without necrosis in swine undergoing a 50% reduction of the left anterior descending coronary artery flow velocity for 7 days. In these experiments too, the decrease in segmental function was progressive over time and was not associated with reduced subendocardial blood flow by day 4. More recently, Shen and Vatner24 and Fallavollita et al25 in pigs as well as Gerber et al26 in dogs succeeded in producing regional contractile dysfunction over periods of 1, 3, and 6 months, respectively. In each of these studies, the severity of regional dysfunction was found to be out of proportion to the reduction in MBF, thus demonstrating perfusion-contraction mismatch. Although the above studies do not dismiss the possibility that chronic perfusion-contraction matching could exist in intact animals over prolonged periods of time, they nevertheless suggest that chronic underperfusion is not a necessary prerequisite to the development of chronic dysfunction in the presence of chronic coronary artery stenoses.Perfusion-Contraction Matching in Patients With Hibernating MyocardiumAssessment of perfusion-contraction matching in patients with hibernating myocardium requires the ability to measure blood flow and function simultaneously.8 Direct assessment of resting MBF in patients is complicated by two factors: the difficulty of measuring MBF in absolute terms in the clinical setting and the known tissue heterogeneity of ischemically injured myocardium. The contention that flow is decreased in human hibernating myocardium is based on the results of clinical studies of the relative distribution of radiolabeled flow tracers such as 201Tl,27282982Rb,30 or 99mTc-MIBI.31 The interpretation of these scintigrams usually assumes that the segments with maximum tracer uptake have normal flow and that any region with an apparent reduction of tracer uptake is underperfused. Because perfusion scintigraphy provides only estimates of relative differences in tracer distribution, a seemingly decreased perfusion to a dysfunctional segment may result in part from an absolute increase in flow to the remote hyperfunctioning tissue.32 The accuracy of relative perfusion scintigraphy is further affected by the limited spatial resolution of the current SPECT devices. This results in significant underestimation of true regional activity concentrations, a phenomenon known as "partial volume effect," which describes how counts measured from a myocardial region with reduced wall thickness will always be lower than those measured from a region with a normal wall thickness.33 The partial volume effect is particularly relevant to the situation of the hibernating myocardium, because the sole loss of systolic wall thickening is expected to result in a 20% to 25% underestimation of regional counts.34 The degree of underestimation can be even larger in the presence of significant wall thinning. Taken together, these limitations make it difficult to determine whether a dysfunctional myocardial segment that exhibits reduced radiolabeled flow tracer uptake at rest with SPECT is truly underperfused or not. Recent refinements in myocardial perfusion imaging, and particularly the advent of PET, have greatly enhanced our ability to measure flow directly in patients with coronary artery disease.353637 PET is a truly quantitative method. It has a much better spatial resolution than SPECT; it allows for accurate correction of photon attenuation and, to some extent, of partial volume effects; and finally, when mathematically and physiologically appropriate models are used to describe the biological behavior of the radiotracers in blood and myocardium, it allows for computation of quantitative estimates of regional myocardial perfusion. Several investigators have attempted to assess the level of resting MBF in patients with hibernating myocardium by PET. Initial studies in patients with previous myocardial infarction indicated that reversibly dysfunctional segments corresponded to areas with qualitatively reduced perfusion but preserved metabolism.383940 However, quantitative studies using [13N]ammonia found that reversibly dysfunctional segments after revascularization had normal or only mildly reduced baseline flow compared with remote, normally contracting areas in the same patients, and myocardial segments with persistent dysfunction had even lower values.4142On the basis of these studies, one could inadvertently conclude that the hibernating myocardium is indeed characterized by a mildly reduced perfusion. Two important aspects must nonetheless be considered. First, the level of flow reduction in most studies is not sufficient to justify ischemic dysfunction.8 Second, many if not all of the above studies have included a variable proportion of patients with previous myocardial infarction, which greatly complicates the interpretation of the flow data. Flow estimates with PET are indeed critically dependent on the mass of tissue that actively participates in tracer exchange within the region of interest. In the presence of marked spatial tissue heterogeneity, such as occurs in previously infarcted myocardium, flow estimates represent the transmural average between several values from very low in microinfarcted areas to almost normal in the noninfarcted epicardial zones and thus may not reflect the actual level of flow seen in the viable part of the wall. One approach to circumvent this problem is to study carefully selected patients with hibernating myocardium in whom any evidence of previous myocardial infarction is lacking. Vanoverschelde et al32 studied 26 patients whose clinical and angiographic characteristics were quite similar to those initially described by Rahimtoola.4 All had symptomatic coronary artery disease, no previous myocardial infarction, and complete chronic occlusion of a major coronary artery. In patients with normal resting wall motion, no difference in MBF measured with [13N]ammonia and PET was found between normal and collateral-dependent myocardium. In patients with abnormal resting wall motion, MBF was higher in remote than in collateral-dependent segments (95±27 versus 77±25 mL·min−1·100 g−1, P<.001). Yet no difference was found among collateral-dependent segments from patients with and without wall motion abnormalities (77±25 versus 85±14 mL·min−1·100 g−1, P=NS) (Fig 1). To corroborate their surprising results, the authors also investigated regional myocardial oxygen consumption using [11C]acetate and PET. In patients with normal resting wall motion, oxygen consumption was comparable between remote and collateralized segments. In patients with resting wall motion abnormalities, myocardial oxygen consumption was higher in remote than in collateral-dependent segments. It did not differ significantly, however, among collateral dependent segments of patients with and without regional wall motion abnormalities. Similar findings were also reported by Sambuceti et al.43 Another approach to circumvent the problems of tissue heterogeneity is to use [15O]H2O and PET to measure MBF.4445 Quantification of MBF by use of [15O]H2O indeed allows us to incorporate into the kinetic model an estimate of the fraction of tissue in the region of interest that is exchanging the freely diffusible water.4445 This approach provides values of flow per gram of perfusable tissue as opposed to per gram of region of interest. Because for any practical purposes, the exchange of water by scar tissue can be regarded as negligible, this technique thus measures flow predominantly in the nonnecrotic part of the wall. Using [15O]H2O, de Silva et al,46 Marinho et al,47 and Conversano et al48 found that 80% to 90% of hibernating segments exhibited baseline flow values that are within the range of resting flow values measured in normal regions.Repeated Stunning as a Plausible Mechanism for Chronic Myocardial HibernationIf resting flow is not reduced, what can then be the trigger for the chronic reduction in mechanical function? Even if resting perfusion is nearly normal, perfusion reserve is highly abnormal in hibernating segments. Vanoverschelde et al32 investigated the regional flow reserve of collateral-dependent myocardium using dipyridamole as the hyperemic agent. They found a wide range of hyperemic transmural flow values, from a fourfold increase to a 20% decrease in basal flow. Importantly, the collateral flow reserve of the dysfunctional segments was markedly blunted and correlated with the severity of chronic regional wall motion abnormalities. Accordingly, these authors suggested that repetitive intermittent episodes of ischemia (either exercise-induced or related to primary reductions in coronary blood flow [plaque events, vasoconstriction, platelet aggregation]) followed by stunning could be the mechanism leading to chronic regional ischemic dysfunction. It is worth mentioning that Braunwald and Kloner49 had already proposed such a mechanism back in 1982. Although because of the study design, no single episode of stunning could be demonstrated in the study by Vanoverschelde et al,32 subsequent observations by Shen and Vatner24 have provided evidence that chronic dysfunction in collateral-dependent myocardium can result from repeated episodes of ischemia followed by a perpetuated state of chronic stunning. These authors examined the time course of regional dysfunction after ameroid implantation in chronically instrumented pigs and found that the onset of dysfunction was not associated with permanently reduced subendocardial blood flow but was always preceded by repeated episodes of acute dysfunction induced by transient increases in regional demand. Altogether, these data suggest that repetitive stunning is a plausible mechanism that can account for the sustained prolonged contractile dysfunction of the hibernating myocardium. Alternatively, chronic dysfunction could result from a chronic decrease in coronary perfusion pressure in the poststenotic bed.5051 Coronary pressure has been shown to regulate contractile performance acutely, even in the absence of changes in coronary flow.52 Whether a decrease in coronary perfusion pressure can be involved in long-term regulation of contractile function in the hibernating myocardium is unknown and requires further investigation.Chronic Reduction in MBF as a Consequence Rather Than the Cause of Chronic HibernationThe fact that chronic contractile dysfunction in the presence of severe coronary artery disease most likely results from repeated episodes of ischemia followed by a state of chronic stunning does not exclude the possibility that MBF may eventually become reduced in the affected segments. Indeed, increasing evidence suggests that MBF progressively downgrades in response to reduced contractile function. Back in 1975, Heyndrickx et al13 already indicated that MBF measured 4 to 6 hours after reperfusion was often decreased by 20% to 25% in acutely stunned myocardium. Canty and Klocke21 made very similar observations in their dog model of chronic ameroid occlusion. In their experiments, subendocardial blood flow, which had remained normal up to the time of ameroid occlusion and peak contractile dysfunction, gradually decreased after ameroid occlusion. Interestingly, this occurred in the face of a progressive increase in coronary perfusion pressure and a slow normalization of regional contractile function. Recently, Berman et al53 studied the effects of sustained demand-induced ischemia in a pig model of short-term hibernation. They made the intriguing observation that, despite no changes in MBF during stress, transmural blood flow decreased after cessation of the stress and remained depressed for a prolonged period of time. Although it is possible that the above observations are related to some form of microvascular stunning, it is tempting to hypothesize that the progressive reduction in MBF seen under these conditions is somehow secondary to the reduction in resting contractile function and serves as a means to increase residual myocardial perfusion reserve.54Structural and Metabolic Alterations in the Hibernating MyocardiumAlthough there is little doubt that myocardial stunning contributes in one way or another to the chronic dysfunction of the hibernating myocardium, not all the features of chronic myocardial hibernation can be ascribed to stunning. Since the early 1980s, it has been known that chronically dysfunctional myocardial segments exhibit distinct morphological changes that can be demonstrated by both the light and the electron microscope. Flameng et al5556 were the first to report on the presence of such abnormalities. These authors conducted a series of studies on human myocardial biopsy samples harvested at the time of bypass surgery and provided evidence for specific alterations affecting both the cardiomyocytes and the extracellular matrix in chronically dysfunctional segments.325556575859 One striking feature of the changes seen in cardiomyocytes was the loss of contractile material (Fig 2).58 In some cells, this was limited to the vicinity of the nucleus, whereas in others it was very extended, leaving only few or no sarcomeres at the cell periphery. Myofibrillar loss was not accompanied by major cell volume changes, which is clearly different from atrophic degeneration. The space previously occupied by the myofilaments was filled with an amorphous, strongly PAS-positive material typical of glycogen. Under the electron microscope, the general organization pattern of the remaining peripherally located sarcomeres was well preserved. Mitochondria were small and scattered throughout the myolytic cytoplasm. Nuclei were tortuous and showed uniformly dispersed heterochromatin. Sarcoplasmic reticulum was virtually absent, as were T tubules. There were no signs of degeneration, such as cytoplasmic vacuolization, edema, mitochondrial swelling, membrane disruption, or lipid droplets, that would indicate acute ischemic damage or cellular atrophy. From a biochemical point of view, the tissue content of ATP, total adenine nucleotides, and PCr usually remained nearly normal.57 Mitochondrial function, as reflected by the ADP/ATP and PCr/ATP ratios, was also nearly intact,57 a finding consistent with subsequent observations that oxygen consumption (measured with [11C]acetate and PET) is well preserved in the hibernating myocardium.3260 It is interesting to note that these critical observations were made and reported several years before the concept of chronic hibernation was put forward by Rahimtoola and that they have been ignored for more than a decade. It is only recently that the link between these structural alterations and myocardial hibernation has been established.3258 Their presence is now considered a hallmark of the hibernating myocardium, as are the changes in myocardial glucose utilization that have subsequently been described.Several investigators have indeed reported that under fasting conditions, the hibernating myocardium was taking up glucose more avidly than remote normal myocardium,406162 a feature that has been used to predict the reversibility of regional dysfunction after revascularization.40 The comparison of morphological data with the findings on metabolic imaging (which demonstrates increased rate of FDG transport and phosphorylation) has raised intriguing questions about the biochemical fate of exogenous glucose in "hibernating" cells.63 Although it was originally suggested that the increased glucose uptake in the hibernating myocardium resulted from stimulation of anaerobic metabolism by chronic ischemia,64 this explanation now appears unlikely in view of the normal or nearly normal levels of absolute MBF32414243464748 and oxygen consumption3260 measured in these segments. It would also hardly explain the accumulation of glycogen, a quite unusual finding in the setting of ongoing ischemia, which would rather be expected to result in the opposite.65 Even if ongoing ischemia is not implicated, it remains possible that a change in the pattern of myocardial substrate utilization from fatty acid to glucose contributes to the metabolic alterations of the hibernating myocardium. In this regard, it is worth mentioning that Liedtke et al23 recently presented strong evidence that such a metabolic switch occurred in pigs with chronic coronary stenosis. Chronic activation of glycogen synthase by ischemia has also been proposed to account for the metabolic alterations seen in the hibernating myocardium. McNulty and Luba66 recently showed that transient ischemia induced a sustained activation of the glucose-6-phosphate–independent form of glycogen synthase, allowing for a rapid replenishment of the glycogen stores during reperfusion. Similar observations were also reported by Bolukoglu et al.67 Together, these data thus suggest that the alterations of glucose metabolism seen in experimental myocardial ischemia and hibernation could result from a concerted deregulation of glycolysis and glycogen synthesis. Further studies are nonetheless required to verify this hypothesis in the clinical setting. Finally, because the uptake of glucose by dysfunctional but metabolically active myocardium was shown to be relatively independent of the hormonal milieu and dietary conditions,6162 some investigators have postulated that a change in the activity or in the expression of the two major cardiac glucose transporters, GLUT-1 and GLUT-4, could be involved in this phenomenon.68 So far, only two preliminary studies have attempted to measure the messenger RNA of these two glucose transporters by quantitative PCR in dysfunctional segments from patients with hibernating myocardium, and they produced conflicting results.6869Recent studies have suggested that the structural changes occurring in hibernating myocardium were the consequence of a dedifferentiation process. The hibernating cardiomyocytes indeed show many features of neonatal cardiomyocytes,70 including (1) depletion of contractile filaments, (2) presence of rough sarcoplasmic reticulum, (3) accumulation of glycogen, (4) occurrence of irregularly shaped nuclei with peculiar distribution of chromatin, (5) loss of organized sarcoplasmic reticulum, (6) lack of T tubules, and (7) vesiculization of the sarcolemma.58 Not all the characteristics of altered cardiomyocytes resemble those of embryonic cells, however. For instance, the remaining sarcomeres in the altered cells often retain their orderly arrangement at the cell periphery, whereas they are randomly distributed in embryonic cells. Also, the amount of glycogen seen in altered cardiomyocytes far exceeds that reported in the embryo. The hypothesis of dedifferentiation is further substantiated by immunohistological studies showing that hibernating cardiomyocytes reexpress contractile proteins that are specific to the feta
Referência(s)