Portal de Periódicos da CAPES

Mechanisms and Models in Heart Failure

2005; Lippincott Williams & Wilkins; Volume: 111; Issue: 21 Linguagem: Inglês

10.1161/circulationaha.104.500546

1524-4539

Douglas L. Mann, Michael R. Bristow,

Cardiovascular Function and Risk Factors

HomeCirculationVol. 111, No. 21Mechanisms and Models in Heart Failure Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticlePDF/EPUBMechanisms and Models in Heart FailureThe Biomechanical Model and Beyond Douglas L. Mann and Michael R. Bristow Douglas L. MannDouglas L. Mann From the Winters Center for Heart Failure Research, Department of Medicine, Baylor College of Medicine, Michael E. DeBakey Veterans Affairs Medical Center, Houston, Tex (D.L.M.), and Division of Cardiology and Cardiovascular Institute, University of Colorado Health Science Center, Denver (M.R.B.). and Michael R. BristowMichael R. Bristow From the Winters Center for Heart Failure Research, Department of Medicine, Baylor College of Medicine, Michael E. DeBakey Veterans Affairs Medical Center, Houston, Tex (D.L.M.), and Division of Cardiology and Cardiovascular Institute, University of Colorado Health Science Center, Denver (M.R.B.). Originally published31 May 2005https://doi.org/10.1161/CIRCULATIONAHA.104.500546Circulation. 2005;111:2837–2849is corrected byCorrectionDespite repeated attempts to develop a unifying hypothesis that explains the clinical syndrome of heart failure, no single conceptual paradigm for heart failure has withstood the test of time. Whereas clinicians initially viewed heart failure as a problem of excessive salt and water retention that was caused by abnormalities of renal blood flow (the "cardiorenal model"1), as physicians began to perform careful hemodynamic measurements, it also became apparent that heart failure was associated with a reduced cardiac output and excessive peripheral vasoconstriction. This latter realization led to the development of the "cardiocirculatory" or "hemodynamic" model for heart failure,1 wherein heart failure was thought to arise largely as a result of abnormalities of the pumping capacity of the heart and excessive peripheral vasoconstriction. However, although both the cardiorenal and cardiocirculatory models for heart failure explained the excessive salt and water retention that heart failure patients experience, neither of these models explained the relentless "disease progression" that occurs in this syndrome. Thus, although the cardiorenal models provided the rational basis for the use of diuretics to control the volume status of patients with heart failure, and the cardiocirculatory model provided the rational basis for the use of inotropes and intravenous vasodilators to augment cardiac output, these therapeutic strategies have not prevented heart failure from progressing, nor have they led to prolonged life for patients with moderate to severe heart failure.1,2In the present review we will summarize recent advances in the field of heart failure, with a focus on the new therapeutic strategies that have been developed for treating systolic heart failure. For a complete discussion on recent advances in the diagnosis and treatment of diastolic heart failure, the interested reader is referred to several recent reviews on this topic.3–5 To provide the proper framework for this discussion, we will review current and emerging therapies within the context of the extant conceptual biological models that clinician scientists have used for envisioning the syndrome of systolic heart failure. However, as discussed at the conclusion of this review, our current working models for heart failure are insufficient for explaining several of the new and emerging therapies for treating systolic heart failure. To this end, we suggest a simplified conceptual model for heart failure that both unites and extends several of the existing working models for heart failure.Heart Failure as a Progressive ModelFigure 1 provides a general conceptual framework for discussing the development and progression of heart failure. As shown, heart failure may be viewed as a progressive disorder that is initiated after an "index event" either damages the heart muscle, with a resultant loss of functioning cardiac myocytes, or alternatively disrupts the ability of the myocardium to generate force, thereby preventing the heart from contracting normally. This index event may have an abrupt onset, as in the case of a myocardial infarction, it may have a gradual or insidious onset, as in the case hemodynamic pressure or volume overloading, or it may be hereditary, as in the case of genetic cardiomyopathies. Regardless of the nature of the inciting event, the common feature in each of these index events is that they all, in some manner, produce a decline in pump function of the heart. In most instances patients will remain asymptomatic or minimally symptomatic after the initial decline in pumping capacity of the heart or will develop symptoms only after the dysfunction has been present for some time. Thus, when viewed within this conceptual framework, left ventricular (LV) dysfunction is necessary but not sufficient for the development of the syndrome of heart failure. Download figureDownload PowerPointFigure 1. Pathogenesis of heart failure. Heart failure begins after an index event produces an initial decline in pumping capacity of the heart. After this initial decline in pumping capacity of the heart, a variety of compensatory mechanisms are activated, including the adrenergic nervous system, the renin-angiotensin system, and the cytokine system. In the short term these systems are able to restore cardiovascular function to a normal homeostatic range, with the result that the patient remains asymptomatic. However, with time the sustained activation of these systems can lead to secondary end-organ damage within the ventricle, with worsening LV remodeling and subsequent cardiac decompensation. As a result of resultant worsening LV remodeling and cardiac decompensation, patients undergo the transition from asymptomatic to symptomatic heart failure.As shown in Figure 1, the compensatory mechanisms that are activated after the initial decline in the pumping capacity of the heart are able to modulate LV function within a physiological/homeostatic range, such that the functional capacity of the patient is preserved or is depressed only minimally. The portfolio of compensatory mechanisms that have been described include early activation of the adrenergic nervous system and salt- and water-retaining systems in order to preserve cardiac output,6–8 as well as activation of a family of vasodilatory molecules, including natriuretic peptides, prostaglandins (PGE2 and PGEI2), and nitric oxide, to counteract the excessive vasoconstriction resulting from excessive activation of the adrenergic and renin-angiotensin systems.9,10 However, our understanding of the family of molecules that may be involved in this process is far from complete. Although patients with depressed systolic function may remain asymptomatic or minimally symptomatic for years, at some point patients will become overtly symptomatic, with a resultant striking increase in morbidity and mortality. The transition to symptomatic heart failure is accompanied by further activation of neurohormonal and cytokine systems, as well as a series of adaptive changes within the myocardium, collectively referred to as "LV remodeling." Although there are further modest declines in the overall pumping capacity of the heart during the transition to symptomatic heart failure, the weight of experimental and clinical evidence suggests that heart failure progression occurs independently of the hemodynamic status of the patient.Neurohormonal Mechanisms for the Progression of Heart FailureIn the latter part of the 1980s and early 1990s, evidence began to appear that certain other types of medical therapy might have a beneficial effect on the natural history of LV dysfunction or myocardial failure, despite initial hemodynamic effects that were either unimpressive11–13 or even adverse.11,12,14,15 These 2 types of therapies, namely, ACE inhibitors and β-adrenergic blocking agents, have dramatically changed the way in which we conceptualize heart failure. As will be discussed below, data generated from both experimental model systems and clinical trials suggest that both types of therapy may prevent the progression of pump dysfunction that characterizes the natural history of heart failure and may halt or even reverse the progressive cardiac dilatation that occurs as heart failure progresses. It is important to emphasize that the beneficial effects of these treatments are not pharmacological but rather are due to favorable effects on the biology of the failing heart. The aforementioned observations led to a point of view that heart failure should be viewed as a "neurohormonal model," in which heart failure progresses as a result of the overexpression of biologically active molecules that are capable of exerting deleterious effects on the heart and circulation.16 "Neurohormone" is largely a historical term, reflecting the original observation that many of the molecules that were elaborated in heart failure were produced by the neuroendocrine system and thus acted on the heart in an endocrine manner. However, it has since become apparent that a great many of the so-called classic neurohormones such as norepinephrine and angiotensin II are synthesized directly within the myocardium and thus act in an autocrine and paracrine manner. Furthermore, molecules such as angiotensin II, endothelin, natriuretic peptides, and tumor necrosis factor (TNF) are peptide growth factors and/or cytokines that are produced by a variety of cell types within the heart, including cardiac myocytes, and thus do not necessarily have a neuroendocrine origin. Nonetheless, the important unifying concept that arises from the neurohormonal model is that the overexpression of portfolios of biologically active molecules can contribute to disease progression independently of the hemodynamic status of the patient, by virtue of the deleterious effects that these molecules exert on the heart and circulation.The evidence in support of the foregoing point of view is derived from 2 lines of investigation. First, a number of experimental models have shown that pathophysiologically relevant concentrations of neurohormones17–19 or overexpression of single components of their signal transduction cascade20–22 is sufficient to mimic some aspects of the heart failure phenotype. Second, clinical studies have shown that antagonizing neurohormones leads to clinical improvement in patients with heart failure.23–30 Thus, a logical explanation for the progression of heart failure is that long-term activation of a variety of neurohormonal mechanisms produces direct end-organ damage within the heart and circulation. Accordingly, progressive activation of neurohormonal mechanisms may explain why heart failure may develop insidiously many years after an acute myocardial infarction, despite the absence of ongoing ischemia. The neurohormonal model also explains why the so-called heart failure phenotype appears remarkably consistent in patients with different etiologies for their heart failure, insofar as disease progression is ultimately driven by very similar portfolios of biologically active molecules, regardless of the inciting cause.Thus far, a variety of proteins, including norepinephrine, angiotensin II, endothelin, aldosterone, and TNF, have been implicated as some of the potentially biologically active molecules whose biochemical properties are sufficient to contribute to disease progression in the failing heart. Disease progression may also be engendered by the loss of the beneficial effects of endogenous vasodilators such as nitric oxide, natriuretic peptides, prostaglandins, and kinins, which are insufficient to counteract the peripheral vasoconstriction that results for endothelial cell dysfunction and the vasoconstrictor properties of angiotensin II and norepinephrine. The most powerful compensatory mechanism activated to support the failing heart is perhaps an increase in cardiac adrenergic drive.31 Unlike other compensatory mechanisms, adrenergic activation accesses all the known means by which myocardial performance can be stabilized or increased.7 These include an increase in contractile function, increase in heart rate, cardiac myocyte hypertrophy, and volume expansion/increased end-diastolic volume (via β-adrenergic signaling of nonosmotic vasopressin release).7 However, in addition to the positive effects on stabilizing myocardial performance, increased myocardial adrenergic signaling, particularly through β1-adrenergic receptor pathways,32 is also highly cardiomyopathic.18,22,33 A summary of some of these helpful and harmful adrenergic receptor pathways is given in Table 1, although this table is somewhat oversimplified. TABLE 1. Biological/Physiological Responses Mediated by Postjunctional Adrenergic Receptors in the Human HeartBiological ResponseAdrenergic ReceptorBeneficial effects Positive inotropic responseβ1, β2 >>α1C Positive chronotropic responseβ1, β2 Vasodilationβ1 (epicardial), β2 (small vessel)Harmful effects Cardiac myocyte growthβ1> β2 >>α1C Fibroblast hyperplasiaβ2 Myocyte damage/myopathyβ1> β2, α1C Fetal gene inductionβ1 Myocyte apoptosisβ1 Proarrhythmiaβ1, β2, α1C Vasoconstrictionα1CAs implied by a greater number of harmful than helpful effects of activation of the adrenergic receptor pathways listed in Table 1, the net effect of a sustained increase in cardiac adrenergic activity in the failing heart is to promote myocardial disease progression and to accelerate the natural history of heart failure. Indeed, repeated observations of the salutary effects of β-blocking agents in clinical trials have shown that chronically elevated β-adrenergic signaling has adverse effects on contractile function, remodeling, and heart failure morbidity and mortality. As shown in Table 2,134–136 these effects appear to be primarily delivered through β1-receptor signaling, inasmuch as both β1-receptor selective (metoprolol CR/XL and bisoprolol) or nonselective agents (carvedilol, bucindolol) have similar salutary effects in terms of molecular responses and clinical outcomes. The reasons for this are 2-fold: the increased myopathic potential of β1- versus β2- or α1-receptor signaling that is summarized in Table 132,34 and the binding affinity selectivity of norepinephrine for β1 versus β2 or α1 receptors.32 Thus, the beneficial effects of β-blocking agents appear to be due to the class effects of β1-receptor blockade, at least in terms of molecular responses35 and clinical outcomes.28,36TABLE 2. Class Effects of β-Adrenergic Blockade in Chronic Heart FailureEffectStudiesβ-BlockersCV indicates cardiovascular; HF, heart failure; MERIT-HF, Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure; CIBIS-II, Cardiac Insufficiency Bisoprolol Study II; COPERNICUS, Carvedilol Prospective Randomized Cumulative Survival Study; and MDC, Metoprolol in Dilated Cardiomyopathy Trial.Reduction in total mortalityMERIT-HF,28 CIBIS-II,36 COPERNICUS134Metoprolol CR/XL, bisoprolol, carvedilolReduction in CV mortalityMERIT-HF, CIBIS-II, COPERNICUS, BEST42Metoprolol CR/XL, bisoprolol, carvedilol, bucindololReduction in CV or HF hospitalizationsMDC,135 CIBIS-II, MERIT-HF, COPERNICUS, US Carvedilol,27 BEST136Metoprolol tartrate, metoprolol CR/XL, bisoprolol, carvedilol, bucindololImproved HF symptomsMDC, CIBIS-II, MERIT-HF, US CarvedilolMetoprolol tartrate, metoprolol CR/XL, bisoprolol, carvedilolReduced need for cardiac transplantationMDC, BESTMetoprolol tartrate, bucindololReduction in myocardial infarctionBEST136BucindololThe adverse effects of β-adrenergic signaling on heart failure natural history would seem to dictate that any type of antiadrenergic therapy would be equally effective, as long as it inhibited the β1-adrenergic signaling. However, recent clinical trial data indicate that the type of antiadrenergic therapy, particularly receptor blockade versus reducing norepinephrine release, is critically important.37–39 The likely explanation for the polar difference in the response of these 2 general classes of antiadrenergic agents is that, during the crucial early period of adrenergic inhibition, sympatholytic agents produce an irreversible removal of adrenergic support, with inability to recruit adrenergic drive when needed to support cardiac function. In contrast, β-blockers are mass-action agents whose inhibition can be easily reversed by norepinephrine competition, which allows for retention and recruitment of the powerful adrenergic support mechanism on an as-needed basis. Extensions of these observations include the potentially favorable effects of therapeutic approaches that allow the beneficial aspects of adrenergic inotropic support to be maintained in the presence of β-blockade40 or the addition to β-blockade to positive inotropic device therapy.41 On the basis of experience with the β-sympatholytic agent bucindolol in the Beta-Blocker Evaluation of Survival Trial (BEST),42 it is apparent that β-blocking agents can interact with certain characteristics of heart failure subpopulations to produce differences in clinical response. This is in contrast to the rather unvarying pharmacological properties and clinical responses to ACE inhibitors. These observations highlight the complexities encountered in therapeutic development in heart failure, wherein surprises predominate, and the only way to directly test hypotheses is in phase III clinical trials.Of major relevance to antiadrenergic strategies in heart failure is the impact of adrenergic receptor polymorphisms on myocardial disease progression and on therapeutic response. For example, a double adrenergic receptor polymorphism, an α2C deletion/loss of function genotype (α2C Del322-325), combined with a high-functioning β1-receptor genotype (β1 Arg389), confers a 10-fold risk for the development of heart failure.43 The α2C polymorphism likely leads to a reduction in the natural brake on norepinephrine release provided by α2 receptors, and the increased adrenergic drive in these individuals then presumably damages the heart to a greater extent in individuals with the high-functioning β1 receptor polymorphism. Transgenic mice with genetic ablation of the α2C receptor have elevated norepinephrine levels and develop evidence of cardiomyopathy.44 Importantly, the α2C polymorphism is enriched in black persons,43 and it provides a potential explanation for certain characteristics of heart failure in this population, including worse cardiac function and prognosis per a given degree of functional incapacity. When transgenically overexpressed in mouse hearts,45 the high-functioning β1Arg389 receptor variant, which by prevalence is the wild-type form of the β1-adrenergic receptor, is much more cardiomyopathic than the lower-functioning β1Gly389 polymorphic counterpart. There is evidence from the BEST study (S. Liggett, MD, P. Lavori, MD, M.R. Bristow, MD, unpublished data, 2005) that the clinical response to bucindolol was affected in a predictable manner by these genetic variants. On the basis of these and other observations, we may be close to the time when genotyping will be a necessary prerequisite to selecting the proper treatment for chronic heart failure patients, at least in terms of antiadrenergic therapy.Is the Neurohormonal Model Adequate to Explain the Progression of Heart Failure?Despite the many strengths of the neurohormonal model in terms of explaining disease progression and the many insights that neurohormonal models have provided in terms of drug development for heart failure, there is increasing clinical evidence to suggest that our current neurohormonal models fail to completely explain disease progression in heart failure. Our current medical therapies for heart failure will stabilize heart failure and in some cases reverse certain aspects of the disease process. However, in the overwhelming majority of patients, heart failure will progress, albeit at a slower rate. Moreover, as heart failure progresses, many patients will be refractory and/or intolerant to conventional medical therapy and often require withdrawal of conventional medical therapies.46 In addition, many types of neurohormonal inhibition have been shown to be ineffective or even harmful in heart failure patients (reviewed in Mann et al47). Although the precise mechanism(s) for this attenuation, loss, or lack of effectiveness of neurohormonal antagonism is not known, there are at least 5 potential explanations that warrant a brief discussion. One obvious explanation is that it may not be possible to achieve complete inhibition of the renin-angiotensin system or the adrenergic system in heart failure because of dose-limiting side effects of ACE inhibitors and β-blockers. A second explanation is that there may be alternative metabolic signaling for neurohormones that are not antagonized by conventional treatment strategies (eg, the conversion of angiotensin I to angiotensin II within the myocardium by tissue chymase).48,49 Indeed, the results of recent clinical trials in which angiotensin receptor antagonists and aldosterone antagonists have been shown to have benefit when added to conventional therapy with ACE inhibitors and β-blockers clearly support this point of view.30,50,51 Third, the currently available portfolio of neurohormonal antagonists, namely, ACE inhibitors and β-blockers, may not antagonize all of the alterations in biologically active systems that become activated in the setting of heart failure (Table 3). Indeed, given the inherent biological redundancy of all mammalian systems, it is perhaps predictable that there will be a number of biologically active molecules that are sufficient to contribute to disease progression by virtue of their toxic effects on the heart and the circulation. Thus, it is likely that with the current technologies for gene expression monitoring, as well as the innovative cloning strategies that are being used, it is only a matter of time before investigators identify new families/classes of biologically active molecules that are capable of contributing to disease progression. A fourth factor is that some heart failure–activated neurohormonal/cytokine signaling pathways capable of producing harmful effects in cardiac myocytes is isolated systems (eg, endothelin, TNF) may have also have favorable effects when functioning in the complex heart failure milieu. A fifth explanation for the loss of effectiveness of neurohormonal antagonism is that, at some point, heart failure may progress independently of the neurohormonal status of the patient. Thus, analogous to the limitations described for hemodynamic models for heart failure, neurohormonal models may be necessary but not sufficient to explain all aspects of disease progression in the failing heart. TABLE 3. Overview of LV RemodelingAlterations in myocyte biology Excitation contraction coupling Myosin heavy chain (fetal) gene expression β-Adrenergic desensitization Hypertrophy Myocytolysis Cytoskeletal proteinsMyocardial changes Myocyte loss Necrosis Apoptosis Alterations in extracellular matrix Matrix degradation Replacement fibrosisAlterations in LV chamber geometry LV dilation Increased LV sphericity LV wall thinning Mitral valve incompetenceLV Remodeling as a Cause of Disease Progression in Heart FailureNatural history studies have shown that progressive LV remodeling is directly related to future deterioration in LV performance and a less favorable clinical course in patients with heart failure.52–54 Although some investigators currently view LV remodeling simply as the end-organ response that occurs after years of exposure to the deleterious effects of long-term neurohormonal stimulation, others have suggested that LV remodeling may contribute independently to the progression of heart failure.52,55 Although a complete discussion of the complex changes that occur in the heart during LV remodeling is well beyond the intended scope of this brief review, it is worth emphasizing that the process of LV remodeling extends to and affects importantly the biology of the cardiac myocyte, the volume of myocyte and nonmyocyte components of the myocardium, and the geometry and architecture of the LV chamber (Table 3). Although each of these various components of the remodeling process may contribute importantly to the overall development and progression of heart failure, the reversibility of heart failure is determined by whether the changes that occur at the level of the myocyte, the myocardium, or the LV chamber are reversible. In this regard, the changes that occur at the level of the myocyte and the LV chamber appear to be at least partially reversible in some experimental and/or clinical models.14,56–58A number of changes that occur during the process of LV remodeling may contribute to worsening heart failure. Principal among these changes is the increase in LV wall stress that occurs during LV remodeling. Indeed, one of the first observations with respect to the abnormal geometry of remodeled ventricle was the consistent finding that the remodeled heart was not only larger but was also more spherical in shape.59 As depicted in Table 4, the increase in LV size and resultant change in LV geometry from the normal prolate ellipse to a more spherical shape creates a number of de novo mechanical burdens for the failing heart, most notably an increase in LV end-diastolic wall stress. Insofar as the load on the ventricle at end-diastole contributes importantly to the afterload that the ventricle faces at the onset of systole, it follows that LV dilation itself will increase the work of the ventricle and hence the oxygen utilization as well. In addition to the increase in LV end-diastolic volume, LV wall thinning also occurs as the ventricle begins to remodel. The increase in wall thinning along with the increase in afterload created by LV dilation leads to a functional "afterload mismatch" that may further contribute to a decrease in forward cardiac output.60–63 Moreover, the high end-diastolic wall stress might be expected to lead to episodic hypoperfusion of the subendocardium, with resultant worsening of LV function,64–66 as well as increased oxidative stress, with the resultant activation of families of genes that are sensitive to free radical generation (eg, TNF and interleukin-1β). TABLE 4. Mechanical Disadvantages Created by LV RemodelingIncreased wall stress (afterload)Afterload mismatchEpisodic subendocardial hypoperfusionIncreased oxygen utilizationSustained hemodynamic overloadingWorsening activation of compensatory mechanismsMyocardial desynchronizationGiven the potential central importance of LV remodeling in the progression of heart failure, the following section will focus on the basic cellular and molecular mechanisms that are responsible for this process. Although the complex changes that occur in the heart during LV remodeling have canonically been described in anatomic terms, the process of LV remodeling also has an important impact on the biology of the cardiac myocyte, changes in the volume of myocyte and nonmyocyte components of the myocardium, and the geometry and architecture of the LV chamber (Table 3). Although each of these various components of the remodeling process may contribute importantly to the overall development and progression of heart failure, it is extremely unlikely that any single aspect of the remodeling process itself will satisfactorily explain the progressive cardiac decompensation that occurs as heart failure advances. Accordingly, the remaining discussion will focus on the collective changes that occur in the cardiac myocyte, the myocardium, and the LV chamber, with an emphasis on those aspects of the remodeling process that might potentially contribute to disease progression.Alterations in the Biology of the Cardiac MyocyteIn both animal models and in the human heart, it is generally held that cardiac myocyte67 or global pump is the primary initiating event that leads to cardiac remodeling, although remodeling can occur in the absence of myocyte dysfunction in some experimental models.68,69 Numerous studies have suggested that failing human cardiac myocytes undergo a number of important changes that might be expected to lead to a progressive loss of contractile function, including decreased expression of α-myosin heavy chain gene with increased expression of β-myosin heavy chain,70,71 progressive loss of myofilaments in cardiac myocytes,72 alterations in cytoskeletal proteins,72 alterations in excitation contraction coupling,73 and desensitization of β-adrenergic signaling.74 Although many of the aforementioned changes may be thought of as beneficial in terms of protecting myocytes against the potential deleterious consequences of excessive neurohormonal activation, collectively these changes would be expected to lead to a defect in myocyte contractile function, as well as decreased loss of responsiveness to normal adrenergic control mechanisms, both of which are hallmarks of failing human myocardium. Indeed, when the contractile performance of isolated failing human myocytes has been examined under very simple experimental conditions, investigators have found that there is ≈50% decrease in cell shortening in failing human cardiac myocytes compared with nonfailing human myocytes.75 Moreover, as noted in the foregoing discussion, this defect in cell shortening has a number of important components that may act combinatorially to produce the observed phenotype of cellular contractile dysfunction. Thus, the contractile dysfunction that develops within myocytes during the process of LV remodeling is likely to involve ensembles of genes, i