Growth Hormone, Acromegaly, and Heart Failure: An Intricate Triangulation

Luigi Saccà, Raffaele Napoli, Antonio Cittadini


Clin Endocrinol. 2003;59(6) 

In This Article

Role of the Neurohormonal System

A cardinal feature of heart failure is the persistent activation of the neurohormonal system. This includes a variety of hormones, neurotransmitters, and cytokines, in particular, TNF-α and IL-6 (Francis et al., 1990; Swedberg et al., 1990). The neurohormonal system is activated in heart failure to prevent blood pressure and cardiac output from falling, and this is achieved by the combined effects of vasoconstriction and fluid retention (haemodynamic model of heart failure). On a long-term basis, this adaptive response becomes detrimental because neurohormones exert cardiotoxic effects and cause extensive remodelling of the myocardial interstitial compartment. Indeed, the degree of neurohormonal activation is maximal in decompensated heart failure and bears a negative prognostic value with regard to disease progression (neurohormonal model of heart failure; Swedberg et al., 1990; Packer, 1992). Convincing evidence in support of this view derives from large-scale clinical trials demonstrating improved survival of patients receiving drugs that block such specific components of the neurohormonal system as angiotensin II, catecholamines and, according to recent data, aldosterone (SOLVD Investigators, 1991; Heidenreich et al., 1997; Pitt et al., 1999). These studies firmly establish the concept that heart failure is indeed an endocrine disease ( Table 4 ).

It has been long believed that chronic GH excess elicits adrenergic overactivity and this would be why the myocardial structural alterations in acromegaly resemble those observed in phaeochromocytoma (Kline, 1961; Van Loon, 1979). More insights into the influence of GH on the sympathetic nervous system have been provided by two recent studies. The first was conducted in patients with GH deficiency and was based on direct recording of sympathetic nerve firing by microneurography (Sverrisdóttir et al., 1998). The data showed that GH deficiency is associated with enhanced sympathetic nerve activity, suggesting that GH normally exerts a restraining effect. In this context, also particularly relevant is the finding that GH administration to rats with postinfarction heart failure reduced the myocardial content of norepinephrine - an event that was associated with improved cardiac energy status (Omerovic et al., 2000). A similar finding was reported in a preliminary, uncontrolled study in human dilated cardiomyopathy. After 3 months of GH therapy, the myocardial release of norepinephrine during physical exercise was attenuated (Capaldo et al., 1998). Incidentally, a marked reduction of the circulating levels of aldosterone was also reported in that study (Capaldo et al., 1998). This observation is relevant in view of the recently documented role of aldosterone in the progression of heart failure (Pitt et al., 1999).

The second study was carried out in acromegalic patients and assessed muscle sympathetic nerve activity using tritiated norepinephrine combined with the forearm balance technique (Capaldo et al., 2000). In this study, muscle norepinephrine turnover was similar in acromegalics and healthy subjects and was raised to comparable levels by insulin infusion. Complementing these observations, it was recently reported that the plasma levels of catecholamines and their response to the upright posture are normal in acromegalics (Bondanelli et al., 1999). It is clear that GH exerts no sympatho-excitatory influence. This observation is particularly interesting in view of the well-documented detrimental effect of catecholamine excess on the myocardium (Communal et al., 1998) and its role in the progression of heart failure (SOLVD Investigators, 1991).

Most of the recent data support the concept that GH activates the renin-angiotensin-aldosterone system(RAAS). Acute administration of GH to normal subjects raises plasma renin activity and aldosterone concentration (Ho & Weissberger, 1990). This is one of the mechanisms mediating the fluid retention effect of GH. Indeed, when GH is administered with captopril or spironolactone, no increment of the extracellular volume is observed (Møller et al., 1997). However, it is doubtful whether chronic GH elevation causes sustained activation of RAAS. For example, in children with GH deficiency, no increment in plasma renin activity and aldosterone can be detected after 1 year of therapy (Ludens et al., 1969; McCaa et al., 1978). This suggests that the early response of RAAS to GH is followed by homeostatic adaptation of the system to the expanded body fluid and sodium compartment. For the same reason, the chronic GH elevation of acromegalic patients has been reported to be associated with normal or even depressed RAAS activity (Kalberg & Ottosson, 1982). In line with this view is the recent observation that transgenic mice overexpressing IGF-I show downregulation of angiotensinogen in cardiomyocytes and reduced levels of angiotensin II (Leri et al., 1999).

Collectively, the data do not support a role for the neurohormonal system in the development and progression of acromegalic heart disease. Recent data indicate that patients with chronic heart failure may present with low IGF-I levels, because of decreased activity of the GH/IGF-I axis or because of peripheral GH resistance (Giustina et al., 1996; Anker et al., 1997). Interestingly, these patients have greater neurohormonal and cytokine activation as compared with patients with normal/high GH/IGF-I activity (Niebauer et al., 1998).