very nice article:

Abstract

GH is an agent widely used in sport to improve physical performance and has been proposed as adjunctive therapy in several clinical conditions. However, its short-term effects on the normal human heart are poorly understood.

Sixty young normal volunteers (30 males and 30 females) were enrolled in a multicenter, double-blind, placebo-controlled study. All subjects were randomized to receive GH (0.03 or 0.06 mg/kg·d) or placebo. A complete Doppler-echocardiographic examination was performed at baseline and after 4 wk of treatment.

Low-dose GH did not significantly affect echocardiographic parameters. In contrast, high-dose GH increased left ventricular mass index by 12% (P < 0.05). The type of growth response was concentric, because left ventricular wall thickness but not diameter increased, leading to a 10% increase of relative wall thickness. These structural changes were associated with functional changes, including a significant increase in cardiac index and a decrease in peripheral vascular resistance; diastolic function was not altered. Fractional shortening and systemic blood pressure were unchanged in the two treatment groups.

In conclusion, administration of GH for 4 wk at doses that simulate GH abuse in sport caused a high cardiac output state associated with concentric left ventricular remodeling.

THE PREVALENCE OF drug abuse in sport has been claimed to be near 15–25%, and it represents a public health problem, considering the potential health hazards (1, 2, 3). The prohibited substance GH exerts anabolic and lipolytic effects (4, 5, 6). In addition, GH is deemed to affect skeletal muscle, cardiac, and respiratory performance (7, 8). The use of GH in sport is also attractive in view of the ease of escaping detection. Although the scale of the problem cannot be precisely assessed, GH abuse is thought to affect both elite athletes and the broader community, particularly adolescents (1). This view is supported by the increasing number of Internet sites describing alleged benefits of GH and IGF-I and by the seizure of GH vials concealed in athletes’ luggage.

The real efficacy of GH in improving exercise capacity and muscle strength in normal subjects and athletes is controversial (9, 10, 11). This can be partly accounted for by: 1) the relatively low doses of GH employed in previous studies; 2) the short treatment duration; 3) the lack of concomitant treatment with other prohibited agents; and 4) the end-point analysis. In this regard, although the physiological changes induced by GH abuse on the cardiovascular system probably have a profound impact on physical performance, they have never been systematically investigated.

Thus, we have evaluated the cardiovascular short-term (28 d) effects of high doses of recombinant human GH in normal fit subjects. This study is part of a larger, multicenter study, named GH-2000, aimed at detecting GH abuse in sport.

Subjects and Methods

Subject selection

Sixty healthy volunteers (30 males and 30 females) were enrolled at two institutions participating in the GH-2000 project, the Sahlgrenska University at Gothenberg and the University of Naples Federico II. Before entering the study, all subjects underwent a physical examination, and a detailed history was obtained. Females underwent a pregnancy test. All subjects were fit, i.e. they had a minimum of two training sessions per week, for at least 1 yr, and were between 18 and 35 yr of age. Exclusion criteria consisted of any disease state. Moreover, all women were using reliable contraception, and pregnancy and lactation were considered exclusion criteria.

Study design

The study was double-blind, randomized, and placebo-controlled. The subjects were assigned to treatment with placebo or 0.03 or 0.06 mg/kg·d of recombinant GH on the basis of a four randomized block design. The two treated groups will be referred to as low- and high-dose group, respectively, although even the low dose was remarkably higher than the dose routinely used in clinical practice (see below). GH (Genotropin, Pharmacia \|[amp ]\| Upjohn, Inc., Stockholm, Sweden, or Norditropin, Novo Nordisk, Gentofte, Denmark) was self-administered as daily sc injections at bedtime. GH dose was established on the basis of anecdotal reports collected by the GH-2000 group in the sport environment, which suggested that athletes take large doses of GH, i.e. as much as 0.08 mg/kg·d (9). It needs to be stressed that the doses employed were much higher than both the usual replacement dose and those administered in heart failure trials, but were justified by the different experimental setting.

Placebo preparations were reconstituted in a volume of solvent identical with that of GH preparations and were also self-injected sc each day. To minimize side effects, the GH dose given during the first week of treatment was half the target dose. The dose was reduced by 50% in case of persisting side effects, and the treatment was stopped if side effects did not vanish after dose reduction. If a subject dropped out during the treatment phase, the washout phase started the following day.

The treatment period was 28 d, followed by 48 d of washout. Blood samples, weight, height, and body mass index were measured in each subject at days 5, 28, and 84. Doppler echocardiographic examination (see below) was performed at days 5 and 28 in all subjects. All 30 subjects studied at the University of Naples Federico II underwent an additional echocardiographic examination at day 84. All subjects gave their written informed consent to their participation, and the study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the guidelines of good clinical practice. The protocol was approved by the Ethics Committees of the participating centers and the National Health Authorities of Sweden and Italy.

Echocardiography

An ultrasound system equipped with a 3.5 MHz transducer (Apogee CX, Interspec Inc., Ambler, PA; or Sequoia C256, Acuson, Mountain View, CA) was used for complete M-mode and two-dimensional echocardiographic analysis. M-mode and two-dimensional recordings were made with the patients in the lateral recumbent position, according to American Society of Echocardiography recommendations (12). In each center, a single operator invariably performed all examinations. The investigator reading the echoes was blinded as to whether the recordings were from placebo- or GH-treated patients. The methods are described in detail elsewhere (13). Measures of left ventricular (LV) architecture and function were assessed according to standard formulae. Peripheral vascular resistance (dynes·sec·cm-5) was calculated as: mean aortic pressure/cardiac output x 80.

Analytical procedures

Serum samples were stored at -80 C until analysis. Samples from each subject were analyzed in the same run to reduce analytic variations. Laboratory staff was blinded to the treatment code, which was broken after the assay results were entered into the database.

Serum IGF-I was measured with a RIA using a monoclonal antibody after acid-ethanol extraction (14), with intra-assay coefficient of variation (CV) of 6.6, 4.4, and 2.3%, and between assay CVs of 9.7, 7.0 and 4.6% at 13.6, 36.7, and 173 nmol·liter-1, respectively. IGF binding protein (IGFBP)-3 was assayed with in-house RIA and polyclonal antibodies (15). Serum IGFBP-3 intra-assay CVs were 6.2, 5.5, and 4.5% at 2.5, 5.7 and 12.6 mg·liter-1, and between assay CVs were 11.9, 14.5, and 13.1% at 2.5, 5.7 and 12.6 mg·liter-1, respectively.

Statistical analysis

All values are mean ± SEM. Statistical analysis was performed with the STATVIEW software package. Between-group comparisons of echocardiographic indexes were performed with the two-way ANOVA with repeated measures in one factor (time). One-way ANOVA was used for the other comparisons. Where appropriate, comparisons to determine the significant changes within the same group over time and between groups at each time point were performed with the Newman-Keuls test, after the samples were tested for normal distribution. Linear regression analysis was used as appropriate. If the distribution was not normal, corresponding nonparametric tests were used. A value of P < 0.05 was considered significant.

Results

(table1: http://jcem.endojournals.org/cgi/con...l/87/4/1654/T1)
(table2: http://jcem.endojournals.org/cgi/con...l/87/4/1654/T2)
The clinical and hormonal characteristics of the study population at baseline and after 4 wk of treatment are shown in Table 1. There were 25 subjects in the placebo group, 17 in the low-, and 18 in the high-GH group. In Table 2, all side effects are reported. Overall, 46% of the subjects in the active treatment group had side effects, compared with 12% in the placebo group. Specifically, 7 of 17 subjects enrolled in the low-dose group had side effects during the study, and the dose was reduced in one, while in the high-dose group 9 of 18 subjects had side effects and two reduced the dose. In the placebo group, three subjects had side effects. Among these, one subject reduced the dose, and one subject stopped treatment at day 14 because of headache and tachycardia. The most frequently reported side effect was transient fluid retention. Increased sweating and arthralgia were also reported.


In the placebo group, IGF-I levels did not change during treatment. IGF-I levels rose by 67% in the low-dose group, whereas the increase was more marked (+135%) in the high-dose group (Table 1). IGFBP-3 circulating levels also increased significantly from baseline values in the active treatment groups. There were no changes in placebo subjects (Table 1).
Low-dose GH did not affect heart rate, blood pressure, or body mass index. Similarly, Doppler echocardiographic parameters did not change significantly during treatment. However, there was a tendency for the LV mass index and for relative wall thickness to increase (+7% and +6% vs. baseline, respectively; Table 3 and Fig. 1). High-dose GH induced significant cardiac growth, as documented by a 12% increase of LV mass index over baseline values (Fig. 1 and Table 2). This increase was accounted for by an 8% increase of interventricular septum thickness and by an 8% increase of LV posterior wall thickness. The high-dose group also exhibited LV concentric remodeling, as shown by the 10% higher relative wall thickness vs. pretreatment values (Table 3). Diastolic blood pressure and heart rate did not change significantly.


The ejection phase indexes of ventricular function (fractional shortening and ejection fraction) did not vary during the treatment protocol, whereas there was a slight but significant increase of the cardiac index (+11% compared with pretreatment values), which better reflects overall cardiac performance and is closely related to indexes of exercise performance, such as maximal oxygen consumption. Such an increase was mainly accounted for by a slight increase of heart rate (Table 3). The increased cardiac index associated with unchanged blood pressure resulted in a slight but significant decrease of systemic vascular resistance (-11%).
Despite the significant increase in LV mass, LV filling indexes did not indicate any impairment of diastolic function. In the subgroup of subjects undergoing Doppler-echocardiographic examination also at day 84, the differences detected at day 28 were no longer evident (Fig. 1).

Discussion

This study demonstrates that short-term GH administration to normal subjects induces rapid changes in LV geometry and function.

Effect of GH on the normal myocardium

The ability of GH to stimulate cardiac growth has been consistently documented in human and animal studies, in both normal and diseased myocardium (5, 7, 16, 17, 18, 19). The presence of GH and IGF-I receptors in the myocardium (20, 21) and the demonstration of increased protein synthesis in the isolated cardiomyocyte and the whole heart subsequent to IGF-I administration provide solid molecular and cellular underpinnings for this effect (4, 5, 6). Until recently, IGF-I appeared to mediate the trophic effect of GH, because the addition of IGF-I, but not GH, to neonatal cultured cardiomyocytes induces cell hypertrophy after 48 h and the expression of muscle-specific genes (5). This view has been challenged by recent data that demonstrated direct effects of GH on neonatal cardiomyocytes, including cell hypertrophy and alterations of cellular metabolic profile, in the absence of changes in IGF-I mRNA (22). Therefore, the somatotropin receptor is expressed in a broad variety of tissues, and it not only stimulates the local (paracrine/autocrine) production of IGF-I but also mediates direct effects of GH that are still under active investigation.

The 11% increase in the cardiac index observed in the high-dose GH group is mainly accounted for by an increased heart rate. The presence of a high cardiac output state without changes in the ejection phase indexes (fractional shortening and ejection fraction) suggests that enhanced contractility was not the causative mechanism. Rather, the concomitant decrease of peripheral vascular resistance and wall stress points to changes in ventricular loading conditions. In this regard, GH causes peripheral vasodilation in the cardiovascular system (17, 19, 23), an effect that appears to be mediated by IGF-I binding to cognate receptors located in endothelial cells, followed by nitric oxide release (24). The small increment in heart rate observed in the high-dose group may be due to increased adrenergic sensitivity or sympathetic tone, although conflicting data have been reported so far. Human GH deficiency is indeed associated with evidence of either decreased (25) or increased sympathetic tone (26), while acr*****ly fails to exhibit a definite relationship with enhanced adrenergic system activity (27).

Comparison with previous studies

Interestingly, the few controlled trials addressing the effects of GH abuse in athletes failed to document any change in muscle strength or body composition (9, 10, 11). There have been no controlled studies of the effect of GH on cardiac structure and function in normal humans. A pioneer uncontrolled study showed a significant increase of fractional shortening associated with a slight growth response in seven subjects treated for 1 wk with 3 mg GH daily (28). A modest increase of heart rate was also recorded. The data of the present study recall the cardiac abnormalities observed in young acr*****lics (29); when the duration of acr*****ly was less than 5 yr, concentric LV hypertrophy was present in association with a high cardiac output state and no abnormalities of diastolic filling.


Dose dependency of the effects of GH

Another novel finding of our study is that the effects of GH are dose-dependent. Low-dose GH exerted little effect on cardiac structural or functional parameters, despite circulating levels of IGF-I almost doubled above baseline values. Therefore, it appears that high GH doses are required to trigger specific cardiovascular changes, although longer treatment at the lower GH dose might have resulted in more pronounced effects. However, when examining cardiovascular changes induced by exogenous GH, one must take into account that circulating IGF-I levels may not represent faithful markers of biological activity, a speculation supported by the lack of correlation between percentage differences of LV mass or relative wall thickness and percentage changes in IGF-I levels.

Clinical implications

The observation that a mere 4 wk of GH treatment at high doses induces a significant increase of LV mass associated with a high cardiac output state has relevant clinical implications. In fact, although there are no published statistics for obvious reasons, GH abuse probably lasts for years (2). Such continued exposure to high GH/IGF-I levels may lead to the development of an acr*****lic syndrome. In addition to the attendant risk of malignancy, peripheral neuropathy, and glucose intolerance, the negative reverberations of chronic GH excess on the cardiovascular system have been extensively characterized: systemic hypertension, LV hypertrophy with marked histological changes, arrhythmias, abnormal LV diastolic filling, and systolic function (7). These alterations, particularly LV hypertrophy, observed after only 4 wk of GH administration in our high-dose group, and cardiac arrhythmias, in turn, are known to predispose to heart failure and sudden death (30). It must also be stressed that small increments in LV mass, as those reported in the current investigation, have been found to be associated with increased risk for sudden death in the Framingham population (31), particularly when associated with LV concentric remodeling (32). In addition, even though the high cardiac output state observed in our high-dose group may augment exercise capacity, it may be detrimental in the long term because of the attendant increase of oxygen consumption and, therefore, of cardiac energy requirements.

In a realistic scenario, the possibility of concomitant abuse of GH and anabolic steroids should be taken into account. Indeed, anabolic steroids are commonly employed among athletes and body builders and are associated with increased LV mass and altered diastolic filling (33, 34). It appears likely that the actions of both prohibited substances may be additive or even synergistic on myocardial growth, and possibly responsible for functional derangements. This speculation is supported by the recent observation that nandrolone decanoate per se increases IGF-I mRNA in rat respiratory muscles (35).

Besides the rapid cardiac structural and functional changes observed in our study, almost half of the subjects in the active treatment group developed significant side effects, compared with 12% in the placebo group. It may be anticipated that such a percentage would increase under a chronic exposure to GH/IGF-I excess.

In conclusion, short-term administration of supraphysiological doses of GH to healthy, moderately trained subjects is associated with increased LV mass and a high cardiac output state.