Effects of Growth Hormone Administration on Cardiovascular Risk in Cured Acromegalics With Growth Hormone Deficiency

The aim of the study is to evaluate the effects of physiologic growth hormone (GH)
replacement on cardiovascular risk markers, cardiac autonomic function, arterial
distensibility, body composition, and quality of life in men and women with GH deficiency
following treatment of acromegaly. We hypothesize that this population will represent a
newly identified group of patients for whom GH replacement will be of benefit.

Treatment modalities in acromegaly include transsphenoidal surgery and radiation therapy,
which can both result in hypopituitarism. A significant subset of cured acromegalics
therefore develop pituitary hormone deficiencies. Although replacement of adrenal, thyroid
and gonadal hormones is routine practice, clinicians do not replace GH in this subgroup,
even in profoundly GH deficient subjects, as there are no randomized studies proving benefit
in this population. With the accumulation of evidence on the beneficial effects of GH
replacement, this therapy is becoming standard of care in all subjects with GH deficiency
(GHD), except in this acromegaly subgroup where GH has been traditionally withheld. The GHD
syndrome is manifested by an increase in cardiovascular risk, which is potentially
reversible with GH therapy. Cardiovascular disease is the leading cause of death in
acromegalics. Although cure of acromegaly is associated with a reduction in mortality
attributable to GH excess, GHD may be a contributing factor to cardiovascular morbidity and
mortality in this group of patients, as it is in patients with other pituitary tumors. It is
therefore crucial to determine how cured acromegalics with hypopituitarism are affected by
the GHD syndrome, and it is essential to study how this particular population responds to GH
therapy. Because these patients typically have large macroadenomas and are treated with
surgery and radiation therapy, long-term management of hypopituitarism is critical. As with
all endocrine disorders, the goal of therapy is normal hormone replacement, not taking
patients from a state of hormone excess to one of permanent hormone deficiency.

Cardiovascular status in acromegaly

Acromegaly is associated with a 2-3 fold increase in mortality compared to the general
population [2] . GH excess has been recognized to have multiple effects on the heart and
cardiovascular system. GH excess affects cardiovascular health indirectly by increasing the
prevalence of cardiovascular risk factors including hypertension, insulin resistance/type 2
diabetes, and dyslipidemia [3]. In addition, endothelial dysfunction is more prevalent in
acromegaly than in normal controls [4]. Impaired endothelium-dependent vasodilatation with
exaggerated sympathetic-mediated vasoconstrictor response has been recently described in
acromegalic patients [5]. Although flow-mediated dilatation has been shown to improve in
cured acromegalics, it has not been shown to return to normal. Reports on the prevalence
of increased carotid intima-media thickness (IMT) are conflicting. Some studies have
documented an increase in IMT in active acromegaly [6] and some have not [7].

A specific acromegaly-related cardiomyopathy -- independent of hypertension, diabetes and
dyslipidemia -- has been extensively described. Impairment in ejection fraction after
physical activity is observed in up to 73% of patients, which may lead to exercise
intolerance in some of them [9].

Morphological and functional cardiac changes are reversed with normalizing GH/IGF-I levels
[10]. Although ventricular hypertrophy has been shown to regress, it is unclear what
proportion of patients recover a normal ventricular mass. Several echocardiographic studies
have shown that with control of disease activity diastolic filling is improved, but the
effect on ejection fraction and exercise tolerance is variable. Data on reversibility of
cardiovascular disease in acromegaly are heterogeneous due to evolving definitions of cure
for acromegaly, often short duration of studies, varying duration of disease activity,
differences in gender and gonadal status, as well as possible distinct effects of
somatostatin analogs on the heart and vessels. Dysrhythmias are also more common in
acromegaly than in controls [11, 12]. Some studies have shown that permanent myocardial
scarring may occur [13, 14].

In our proposed study population sequelae of previous GH excess may coexist with
manifestations of GH deficiency.

Cardiovascular status in GHD

Cardiovascular morbidity and mortality in adults with GHD has been shown to be increased in
a number of retrospective studies [15-18]. Increased arterial IMT, increased prevalence of
atherosclerotic plaques and endothelial dysfunction have been reported in GH deficient
adults both in childhood and adulthood onset forms [19-22].

The GHD syndrome is characterized by a cluster of factors that are associated with increased
cardiovascular risk, such as central adiposity [23-25], increased visceral fat [26], insulin
resistance [27], dyslipoproteinemia [28] and decreased plasma fibrinolytic activity [29].
GH administration has beneficial effects on a number of these factors, but it is unknown
which mechanisms are implicated in GH action on the process of atherosclerosis.

In addition to alterations in atherosclerotic markers, abnormalities in cardiac function and
structure have been reported among patients with GHD, possibly contributing to the increased
cardiovascular mortality [30-31]. GHD is also associated with cardiac autonomic dysfunction
that may contribute to cardiovascular mortality [33] and improves with GH replacement
therapy. Of particular importance regarding patients with acromegaly, it has been shown
that twelve months of GH replacement improves left ventricular mass and cardiac performance
in young adults with GHD [34]. Therefore, hypopituitary patients with a history of
acromegaly who are now GH deficient may be particularly good candidates to benefit from
physiologic GH replacement.

Adipose tissue has receptors for GH [35], which has lipolytic activity [36]. A decrease in
central fat as assessed by waist-to-hip ratio have been reported in some studies [26,
37-39], but not in others [40-42]. Consequences of increased abdominal adiposity include
increased risk of cardiovascular disease, type 2 diabetes and cerebrovascular disease [43].
Long-term GH treatment decreases total body fat [26, 37-40, 44] including visceral fat [38].
Lean body mass and muscle function are improved with GH therapy in adults with GHD [39, 40,
45]. GH increases lean body mass and decreases adipose tissue mass when given to adults with
GHD [38] or the elderly [46]. Administration of GH causes insulin resistance acutely [26]
but long-term therapy may restore glucose sensitivity [47] through its effects on body
composition.

GH treatment increases lipoprotein (a) (Lp (a)) levels but its effects on other lipoproteins
are still controversial. Some studies have reported decreases in LDL cholesterol with or
without increases in HDL cholesterol [34, 37, 48, 49] with GH administration, while others
have not [20, 50, 51]. Key factors likely involved in the discrepant findings include
heterogeneity of patients studied in terms of age of onset of the GHD (childhood versus
adulthood), gender, severity of GHD and methodological issues such as dose and duration of
GH administration. In addition, many of the studies have no control period. There is a
decrease in the hepatic expression of LDL receptors in GHD, which is reversed by GH therapy.
This phenomenon could be linked to the exaggerated postprandial increase in
triglycerides-rich particles observed in GHD, which is also normalized by the administration
of GH [52].

Inflammation plays a central role in the pathophysiology of atherosclerosis [53]. Each
atherosclerotic lesion represents a different stage of a chronic inflammatory process in the
arterial wall, and different markers along the inflammatory cascade have been reported to
predict cardiovascular risk [54]. Among those, high-sensitivity testing for C-reactive
protein (CRP) is one of the best validated [55]. Several prospective studies support a
strong link between levels of CRP and future risk of coronary events [56-60]. CRP adds
considerable value to the total and HDL cholesterol measurement in the prediction of
cardiovascular risk [61].

These distal markers reflect the consequences of elevated proinflammatory cytokines such as
interleukin-6 (IL-6). GH is known to have important immunomodulatory effects [62-64]. We
therefore hypothesized that the effects of GH on the process of atherosclerosis might be
mediated through the cytokine-inflammatory pathway. We have recently investigated the
effects of physiologic GH replacement in cardiovascular risk markers in men with GHD [65].
In this study we found that CRP and IL-6 levels decreased in GH treated men compared to
controls despite no significant change in serum lipid levels. Other emerging inflammatory
markers include intercellular adhesion molecule-1 (ICAM-1), P-selectin and CD 40 ligand
(CD40L), which is thought to reflect platelet activation and may promote atheromatous plaque
destabilization. Myeloperoxidase was recently shown to predict the early risk of myocardial
infarction and the risk of major adverse cardiac events in the following six months [66].
And lately placental growth factor (PlGF) has been found to be an independent marker of
adverse outcome in patients with acute coronary syndromes [67]. The effect of the GH-IGF-I
axis on these markers is unknown.

We also recently have investigated levels of inflammatory markers in women with
hypopituitarism compared with healthy controls. We found that women with hypopituitarism
have increased levels of IL-6 and CRP, suggesting that chronic inflammation may be involved
in the pathogenesis of atherosclerosis in this population [68]. In addition to inflammatory
markers, thrombogenic cardiovascular risk markers such as fibrinogen [69], tissue-type
plasminogen activator (tPA) and plasminogen activator-inhibitor 1 (PAI-1) are thought to be
surrogate markers of vascular health. It will be critical to determine whether physiologic
GH replacement has beneficial effects in patients with a history of acromegaly, and to
define the influence of GH and gonadal status on these risk factors.

Quality of life has been shown to be poorer in GH deficient females treated for acromegaly
than in females with other causes of GHD [70]. Short-term GH replacement caused a
non-significant improvement in quality of life scores in subjects with GHD following cure of
acromegaly, but the effects of longer GH treatment duration have not been published in this
specific subgroup. Our study will provide more data on the quality of life of subjects
following cure of acromegaly (GH deficient versus GH sufficient) and on the effects of GH
therapy in the GH deficient group.

Data on body composition and cardiovascular risk markers in patients with cured acromegaly
are rare. No studies have yet been published comparing these endpoints in GH sufficient and
GH deficient subjects with a history of acromegaly. Our hypothesis is that GH sufficient
subjects have a more favorable profile than GH deficient subjects. Several studies have
shown a normalization of mortality rates in subjects with cured acromegaly compared to
subjects with active acromegaly. However it has not been demonstrated that this improvement
was mediated by a normalization of the cardiovascular risk factors. Collecting
cross-sectional data in this patient population may contribute to answer this question.