Growth hormone: a newcomer in cardiovascular medicine

Time for primary review 31 days. Growth hormone (GH) controls linear growth and a variety of other functions, including nutrient metabolism, skeletal muscle performance, regulation of body composition, and even psychological well-being. The recent finding that GH plays a role in cardiovascular physiology has enlarged GH's spectrum of action even further [1, 2]. GH is essential for cardiac development and for preserving cardiac morphology and performance in adult life. Patients with GH deficiency, particularly those with the congenital form, present with cardiac atrophy and significant impairment of cardiac performance and exercise capacity [3]. Epidemiologic studies have shown that GH deficiency is associated with a high incidence of cardiovascular mortality, mostly due to heart failure [4]. In single case reports, GH deficiency was associated with severe dilated cardiomyopathy, which was unresponsive to conventional therapy but successfully treated with replacement therapy [5–7]. Clinical studies of GH's role in cardiac physiology are intersecting with experimental attempts to treat heart failure with growth factors. Unlike traditional therapeutic approaches that aim at slowing the development of excessive, pathologic cardiac hypertrophy, the new strategy is to favor the induction of ‘physiologic’ forms of hypertrophy. Indeed, both GH and IGF-I have already proven beneficial in several models of experimental heart failure [8–14]. Endeavors to treat human cardiac failure with a growth factor approach are just beginning. In a preliminary study of a small group of selected patients with idiopathic dilated cardiomyopathy, a 3-month treatment with GH improved ventricular geometry, function, and energetics [15]. These data will hopefully encourage the implementation of more robust clinical trials. At this stage of events, it may be useful to examine this novel approach, the underlying rationale, and the studies that generated the idea to approach heart failure with growth factors. The importance of GH in cardiovascular physiology has emerged from cross-sectional assessment of cardiac morphology and function in clinical states of altered GH secretion. Patients with GH deficiency acquired in adult life may have a reduced left ventricular (LV) mass and performance. These changes are not impressive and are not found in all patients [16–19]. In contrast, in patients with congenital lack of GH there are unequivocal abnormalities of cardiac structure and function [3, 20, 21]. Conceivably, it is the age at which GH deficiency sets in that determines the severity of heart involvement. Cardiac involvement is more prominent if GH deficiency starts during the early development of the heart and consists in cardiac atrophy and ventricular dysfunction, both diastolic and systolic [3, 21]. The resulting clinical condition is characterized by a hypokinetic syndrome with low cardiac output and increased peripheral vascular resistance [3, 22]. Patients with GH deficiency have a decreased exercise capacity, which is not an unexpected finding in view of their reduced skeletal muscle mass and strength [3, 23]. However, we found a significantly lower response of cardiac output to physical exercise in patients with childhood-onset GH deficiency, which supports the idea that cardiac dysfunction contributes to their limited physical activity and poor quality of life [3]. When patients with GH deficiency are treated with replacement therapy, there is a significant improvement of their cardiac status, which is particularly impressive in those with the childhood-onset form of the disease [3, 21, 22]. LV mass, systolic performance, and cardiac output improve consistently after GH, as does exercise capacity. These data strongly support a causative role of GH in the cardiac abnormalities associated with GH deficiency. The cardiac consequences of chronic GH excess have been inferred from observations made in the acromegaly model and are less straightforward than those observed in subjects with GH deficiency. Acromegaly is often associated with such major complications as arterial hypertension, diabetes mellitus, and coronary artery disease, each of which may per se affect cardiac function. Consequently, it has not been easy to isolate the genuine effect of chronic GH excess on the heart and to reconstruct the natural course of the cardiac involvement in acromegaly. Nor have animal models been particularly useful for a variety of reasons, including the limited duration of cardiac exposure to GH excess. In an attempt to appraise the pure cardiac consequences of chronic GH excess, we recently examined a large group of patients affected by uncomplicated acromegaly [24]. In these patients, LV mass was significantly increased – a finding that agrees with autopsy data [25]. In most patients, the increase in LV mass fulfilled the criteria for cardiac hypertrophy. Except for the late stage of uncontrolled acromegaly when cardiac failure often ensues, there was no LV cavity dilation. We concluded that concentric cardiac hypertrophy is a distinct feature of chronic GH overactivity. This view has a solid experimental foundation. It is supported by data obtained in strictly selected acromegalic patients without evidence of other confounding variables that may affect myocardial growth. In addition, if acromegalic patients are treated with the somatostatin analog octreotide, the inhibited GH secretion is paralleled by rapid regression of ventricular hypertrophy and correlates with the improved diastolic function [26, 27]. The structural abnormalities of the acromegalic heart cause derangements in ventricular diastolic function, i.e. prolongation of the isovolumic relaxation time and a marked fall in the early to late filling velocity ratio [24]. Similar findings were obtained when diastolic function was assessed by equilibrium radionuclide ventriculography [28]. Systolic function is not apparently altered when examined at rest. Rather, there is evidence for a hyperkinetic condition, characterized by increased cardiac output and reduced peripheral vascular resistance. However, during physical exercise there is no increase of LV ejection fraction, as assessed by radionuclide angiography, indicating that the cardiac functional reserve is impaired in acromegalic patients [28]. Most data on the effects of GH overactivity on cardiac function are derived from patients with longstanding acromegaly and, consequently, with marked cardiac hypertrophy. Histologic examination of the acromegalic myocardium has revealed extensive interstitial fibrosis, sometimes associated with areas of myocyte necrosis and lymphomononuclear infiltration [25]. Therefore, it is not surprising that cardiac function is impaired after many years of GH overactivity. Yet, several observations suggest that the initial and genuine impact of GH excess on the heart is to increase its performance: (1) short-term administration of GH to normal subjects enhances systolic function [29]; (2) myocardial contractility in increased in animal models of relatively short-term GH excess [30, 31]; and (3) even in the intermediate stage of acromegalic heart disease there is a high output state [1]. The possibility that GH overactivity per se may in the short term enhance cardiac performance is attracting much attention in view of potential exploitation of this property in patients with heart failure. Although this issue can be only resolved by appropriate clinical trials, some insights may be gained from the analysis of cardiac function in patients with short-term acromegaly. We recently studied a group of such patients whose mean disease duration was less than 5 years (3.2±1 years). In these patients, LV mass was significantly increased and associated with enhanced systolic function and cardiac output [32]. More important, none of the indices of diastolic function was impaired, which suggests that interstitial remodeling had not yet occurred to an extent that compromised ventricular function. These data are reassuring with regard to the prospect of a therapeutic use of GH in cardiology and in other fields of internal medicine (chronic renal failure, malnutrition states, etc.). The study also resulted in a more accurate reconstruction of the natural history of the cardiac abnormalities in acromegaly. There seems little doubt that enhanced cardiac performance is a prerogative of the initial stages of acromegaly, although LV ventricular mass is already unequivocally augmented. It is believed that IGF-I mediates most of the effects of GH on cardiac growth and function. IGF-I is synthesized in the myocardial tissue where it acts through autocrine/paracrine mechanisms. In addition, IGF-I produced elsewhere may reach the myocardium borne by the bloodstream, and thus act as an endocrine factor. The GH receptor gene is expressed in the myocardium at a relatively higher extent than in many other tissues [33]. The mechanisms underlying GH receptor expression are largely obscure. It seems that GH regulates its own receptor [34, 35]. For instance, during the regeneration process after ischemic injury, expression of the GH receptor is delayed in hypophysectomized animals [36]. Interestingly, the GH receptor is upregulated in the myocardium of rats with experimental renal hypertension [37]or acute volume overload [38], suggesting a mechanistic link between GH activity and cardiac growth subsequent to a hemodynamic challenge. The intracellular signals generated by GH binding to its receptor are just beginning to be investigated. There is evidence that they include activation of the tyrosine kinase JAK2, MAP kinase, and STAT transcription factors [39]. GH administration increases cardiac IGF-I content in the rat [40]and induces IGF-I mRNA expression [41]. In addition, cardiac myocytes of rats express IGF-I receptors [42], and IGF-I has been implicated in cardiomyocyte growth and proliferation. In cultured cardiomyocytes IGF-I stimulates protein synthesis and accumulation of myofibrils, and induces the transcripts of muscle-specific gene, such as myosin light chain-2, skeletal muscle α-actin, and troponin I [43]. On the other hand, blockage of the expression of IGF-I receptors in cultured neonatal cardiomyocytes prevents protein synthesis and cardiomyocyte growth [44]. Myocardial IGF-I synthesis may occur following hemodynamic challenge. For example, IGF-I mRNA expression is increased in the myocardium after pressure overload [45, 46]. Interestingly, IGF-I mRNA increases at an early stage of ventricular hypertrophy and to a greater extent in myocardial segments exposed to the highest stress [47]. These data place the relation of IGF-I to cardiac hypertrophy in a new mechanistic perspective and suggest that IGF-I plays a role in the scenario of the intricate mechanisms that control cardiac growth, possibly independent of GH itself. Total understanding of the mechanisms underlying the changes in cardiac performance in patients with abnormal GH activity is hampered by the paucity of invasive data and the complexity of GH interaction with the cardiovascular system (Table 1). The effect of GH on cardiac function has been mostly investigated by echo-Doppler and radionuclide angiographic studies that provide useful information on systolic performance and on the ventricular loading conditions, but do not allow an accurate estimation of myocardial contractility. Effects of GH on heart function Effects of GH on heart function A simple way of assessing afterload in clinical studies is to measure end-systolic wall stress (ESS). This is the force acting per unit cross-sectional area of the ventricular wall and, according to Laplace's law, it is directly related to ventricular pressure and radius, and inversely related to wall thickness. Systolic stress is a key determinant of ventricular performance: the higher the stress, the lower the shortening velocity and stroke volume. Therefore, the way GH affects wall stress may provide a clue as to how it interacts with overall cardiac performance. ESS is significantly reduced in acromegaly, even in the presence of arterial hypertension because the effect of the higher blood pressure to raise ESS is offset by the more pronounced wall thickening [22]. Conversely, ESS is markedly increased by GH deficiency and partially corrected by replacement therapy. Of the three components implicated in wall stress, i.e. LV pressure, radius and wall thickness, the major determinant of the changes induced by GH is LV wall thickness, whereas the role played by ventricular radius and pressure is marginal [22]. Exactly how GH affects LV geometry may be gauged by reasoning in terms of relative wall thickness, i.e. the average of interventricular septum and posterior wall thickness normalized by ventricular radius. As shown in Table 2, the effect of GH is to increase relative wall thickness or to restore it to normal when it is decreased as a consequence of lack of GH. This favorable effect of GH on LV geometry is possibly one of the key mechanisms whereby GH improves ventricular performance. Indeed, a remarkable improvement in relative wall thickness was observed in all patients with dilated cardiomyopathy, whether idiopathic or secondary to GH deficiency, when they were treated with GH therapy [5–7, 15]. Effects of GH on left ventricular relative wall thickness Based on data from Refs. [20, 22, 25, 30]. vs. controls. vs. before GH. Effects of GH on left ventricular relative wall thickness Based on data from Refs. [20, 22, 25, 30]. vs. controls. vs. before GH. The effects of GH on preload have been less satisfactorily explored in human models of altered GH secretion. When preload was indirectly assessed by the end-diastolic LV dimension, it was found to be normal in patients with longstanding acromegaly. Conversely, in subjects with GH deficiency LV end-diastolic dimension was reduced and fully corrected by replacement therapy [22]. In some studies of adulthood-onset GH deficiency [16, 18], replacement therapy led to an excessive increment in the LV end-diastolic dimension – a phenomenon also observed in patients with short-term acromegaly [22]. These observations have two implications. First, they support the idea that preload may be an additional mechanism contributing to the changes in cardiac performance associated with abnormal GH activity. Second, whereas it is well established that longstanding GH excess causes ventricular remodeling of the concentric kind, in the short term GH reshapes ventricular geometry according to a pattern reminiscent of eccentric hypertrophy. This view is supported by the observation that LV relative wall thickness is not increased in short-term acromegaly (Table 2). The idea is gaining ground that GH is able to improve cardiac function also independent of its growth effect on cardiac tissue. Supporting this view is the observation that short-term administration of GH to rats with postinfarction heart failure provides functional benefit even in the absence of changes in cardiac mass [12]. In agreement with these data, acute administration of IGF-I to normal human subjects improves LV systolic performance and cardiac output [48, 49]. This response may be mediated by an effect on myocardial contractility or, alternatively, it may be a consequence of acute changes in the ventricular loading conditions. The question whether GH affects myocardial contractility is difficult to address because of paucity of data and because of major difficulties inherent in the assessment of contractility in vivo. The LV ejection phase indices (shortening velocity and ejection fraction) are clearly impaired in subjects with GH deficiency and they improve significantly after GH therapy [3, 21]. However, when the systolic indices are normalized by the end-systolic wall stress, they no longer differ from control values [22]. This would seem to indicate that the altered systolic performance in the presence of GH deficiency is merely a consequence of the changes in afterload and that contractility is not affected. On the other hand, studies on contractility in animal models of GH deficiency or excess seem to support a role for GH in the regulation of contractility. For instance, the force of contraction is increased in the papillary muscle of rats with a GH-secreting tumor [50, 51]. Similarly, contractility is enhanced in the isolated perfused heart from rats treated for 4 weeks with exogenous GH or IGF-I [30, 31]. In line with these observations, contractility was depressed in GH-deficient dwarf rats. In this model, the developed pressure and wall stress in the perfused whole heart were impaired and fully corrected by GH treatment The mechanisms by which changes in GH activity affect contractility are not but there is evidence for an increased of the to 51]. On the other hand, although the to is reduced in dwarf there is no in receptor and or in suggesting that GH's effect is independent of the system GH a variety of effects that may indirectly affect cardiac performance (Table 1). may be in some subjects with and More often it is well possibly because GH's effect is offset by the improved renal and the increased causes the increase in as assessed by the LV end-diastolic dimension, observed in GH-deficient patients after replacement therapy and in short-term acromegaly. and volume GH administration does not raise arterial blood This phenomenon may be for by the that GH is a vascular resistance is decreased in acromegaly and markedly increased in GH deficiency [22]. This effect is due to a action of GH on the resistance Indeed, a response is also by administration of GH or IGF-I the artery – an experimental to raise growth factor only in the and to activation of mechanisms The effect of both GH and IGF-I is the of from the In patients with GH deficiency, is markedly decreased with human GH and peripheral vascular resistance A property of GH is to improve exercise capacity, an effect for by the increased skeletal muscle mass and strength [1]. In addition, GH a of that may to raise physical activity. In the enhanced exercise capacity may favorable effects on cardiac performance through a The study of the use of GH in experimental heart failure was to whether GH, through a effect on the the was able to the of postinfarction LV The study that a treatment with GH decreased consistently the incidence of LV after induction of myocardial The attempt to treat cardiac dysfunction with growth factors was made the rat model of In this model, administration of IGF-I induced a improvement of cardiac output without the extent and severity of the myocardial This favorable effect of IGF-I was in subsequent studies in rats with postinfarction heart failure. In the model, of the left coronary artery marked ventricular remodeling with impairment of systolic function and cardiac IGF-I treatment resulted in additional myocardial hypertrophy in areas without changes in and content IGF-I also improved cardiac output and reduced peripheral vascular resistance. These functional occurred even IGF-I increased the LV end-diastolic dimension, an that is not to the These data support the that growth factor therapy is to the heart in the experimental model of myocardial The question to are growth factors beneficial at a when and ventricular remodeling have already this GH with IGF-I to rats 4 weeks coronary The of the treatment was (1) IGF-I GH's effect on and (2) GH the induced by IGF-I and increases the of binding protein which further the of As observed in the early postinfarction model, also when to animals with established ventricular remodeling growth factors improved cardiac output and reduced consistently peripheral vascular resistance It is that both end-systolic and end-diastolic LV were significantly reduced by growth factor therapy. This effect is particularly important, the of ventricular in the of heart failure A more straightforward of the beneficial effect of GH in the early postinfarction LV remodeling was recently and it was shown that GH induced hypertrophy of the LV myocardium and, reduced LV Cardiac function was improved by GH treatment and assessment of LV function in in the isolated perfused whole heart that GH improved contractility a In a recent study the postinfarction model in hypophysectomized GH administration increased the left ventricular mass but only marginal effects on left ventricular dysfunction The for the lack of functional in this study is not although it may be related to the relatively small and the to myocardial GH has been with a measure of in patients with dilated both idiopathic and secondary to GH deficiency. A feature of dilated is LV by wall In other the relative wall thickness of LV is decreased the high wall The in contributes to myocardial energetics and, with the to the hemodynamic These provide the for a growth factor approach to dilated heart GH the altered LV geometry in dilated cardiomyopathy, with a in systolic wall stress and of LV systolic performance. Based on observations GH provide additional benefit by myocardial contractility and vascular resistance. from further to GH's effect in dilated from case of patients with GH deficiency developed severe heart failure that not to conventional therapy [5–7]. When these patients were treated with GH, their cardiac performance and clinical improved Although this effect was in all related to GH one a effect of the hormone at in two of the patients, the high of GH In a preliminary patients with heart failure secondary to idiopathic dilated were treated for with GH at a of 4 other This is as a replacement for patients with GH deficiency. the patients with dilated had a normal GH GH administration their IGF-I [15]. GH induced a growth response in the myocardium and LV ventricular geometry with a marked fall in systolic wall The ejection phase ventricular and cardiac performance improved whereas there was no clinical of increased – a potential consequence of GH's action on and on the secretion of the factor and the pressure decreased indicating of and ventricular filling pressure, in line with the improved exercise capacity and clinical finding of this preliminary study was that GH improved LV In other after GH therapy the heart more lower and This observation is in agreement with data obtained in rats with GH-secreting tumor [50, supports the idea that GH may be in its to affect the of the This mechanism of action is even more if one that in patients with dilated there is a of the myocardium reserve and a reduced of the heart to These data are but also a of although the improvement was impressive and the study from uncontrolled and on a small of selected More robust and are to the functional benefit of GH therapy in idiopathic dilated and, in other forms of heart failure ischemic dilated although most of the were improved several after of GH therapy, the was which on the effect of this approach on disease The effects of treatment with GH be in view of a potential cardiac tissue growth in excess of is with diastolic function and also in terms of effects [1, In addition, the from is that a short-term functional benefit no relation to the of patients with heart failure. Rather, in some the improvement in cardiac performance was by of disease As as such as the and are not the and the generated by GH and the growth factor approach to heart failure treatment yet