| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
School of Rehabilitation Science (L.J.W.), McMaster University, Hamilton, Ontario, Canada L8S 1C7; Department of Endocrinology (A.M., S.M.S.), Christie Hospital, Manchester M20 4BX, United Kingdom; and Faculty of Medicine (S.E.), The University of Toronto, Toronto, Ontario, Canada M5S 2W6
Correspondence: Address all correspondence and requests for reprints to: Dr. S. Ezzat, University of Toronto, Mt. Sinai Hospital, 600 University Avenue, No. 437, Toronto, Ontario, Canada M5G-1X5. E-mail: sezzat{at}mtsinai.on.ca
 |
Abstract |
|---|
 Top Abstract I. IntroductionII. Physical ImpairmentsIII. Functional LimitationsIV. Health-Related Qualify of...V. Summary and ConclusionsReferences |
|---|
Several studies have documented that administration of GH or somatostatin analogs to adults with GHD or GH excess, respectively, ameliorates abnormal biochemical profile and the associated somatic impairments. However, whether these improvements translate into improved physical function in adults with GHD or GH excess remains largely unknown, and their impact on HR-QoL controversial. Review of placebo-controlled trials to date suggests that GH and somatostatin analogs have greater effects on gas exchange and aerobic performance than as anabolic agents on skeletal muscle mass and function.
Future investigations should include dose-response studies to establish the optimal combination of pharmacological agents plus exercise required to improve not only biochemical markers but also physical function and HR-QoL in adults with GHD or GH excess.
|
I. Introduction |
|---|
|
TopAbstractI. IntroductionII. Physical ImpairmentsIII. Functional LimitationsIV. Health-Related Qualify of...V. Summary and ConclusionsReferences |
|---|
Development and use of objective and self-report measures of physical function and QoL are crucial to evaluate therapeutic interventions in chronic diseases such as GHD and GH excess. Traditional approaches previously evaluated efficacy and effectiveness of therapeutic interventions based on amelioration of clinical signs and symptoms or reduced morbidity. Today, clinical practice and research trials must show improvements in physical function and QoL and must demonstrate that the interventions impact physical, emotional, social, and perceived health status, in addition to symptoms of the disease. Increased longevity alone, in the absence of demonstrated improvement in an individual’s level of physical function and QoL, is no long sufficient justification for therapeutic interventions.
There is general agreement that administration of GH to adults with GHD reduces impairment by increasing circulating levels of its target IGF-I, skin thickness, bone mineral content, and lean body mass (LBM), and reducing fat mass. Whether amelioration of these impairments translates into clinically meaningful improvements in physical function and HR-QoL remains more controversial. The effects of GH excess and its treatment on such physiological endpoints have been less well studied. We review the impact of perturbed GH status and the effects of normalizing GH status on the physical impairments (body functions and structure), functional limitations, and disability, including HR-QoL, using the clinical paradigms of GHD or GH excess in adults.
The World Health Organization definition of health to be a "state of complete physical, mental, and social well-being and not merely absence of disease or infirmity" has remained unchanged since 1948 (1). Although mortality was previously the measure of choice to reflect population health, the importance of "nonfatal" health outcomes (i.e., functioning and disability in various aspects of life) has recently been recognized. National mortality statistics, reported on the basis of the international classification of diseases (ICD) system, was useful for tracking life expectancy and causes of death but failed to reflect health status among the living population. This led to the development of the International Classification of Impairments, Disabilities, and Handicaps (ICIDH) (2) to classify the consequences of disease. The ICIDH provides a unifying framework linking cellular adaptations in specific tissues to impairment and functional limitations. Nagi adapted the ICIDH and proposed a model of disablement that is widely used to evaluate effectiveness of therapeutic interventions designed to improve physical function in those with neurological and orthopedic diseases (3, 4) (Fig. 1
). We use Nagi’s model here to provide a common set of terms to describe limitations in physical function that are associated with endocrine health conditions, namely GHD and GH excess. "Impairments" refers to any loss or abnormality of psychological, physiological, or anatomical structure or function at the tissue, organ, or whole body system level (e.g., low IGF-I or reduced muscle strength). "Functional limitations" refers to any restriction or inability (resulting from an impairment) to perform an activity in the manner or within the range considered normal for a human being (for example, limited ability to walk). "Disability" has been subclassified into four categories, including physical, mental, social, and emotional disability. Disability represents any restrictions or limitations in the fulfillment of a person’s normal (depending on their age, gender, social, and cultural factors) socially defined roles and tasks at work, school, or recreation, or for personal care. We review the effects of GHD and GH excess on perceived disability using a variety of HR-QoL measures. Traditionally, impairment refers to a problem with a body structure or organ, functional limitations reflect difficulty with respect to performing a particular activity, and disability refers to a disadvantage in filling a role in life relative to a comparable peer group. This model was used because it is felt to be more meaningful to a multidisciplinary audience than the revised ICIDH model titled International Classification of Functioning, Disability and Health (ICIDH-2), where the domains include impairment, activity, and participation. To the extent that data are available, the relationships among the various measures of impairment, function, and QoL are discussed.
|
|
II. Physical Impairments |
|---|
|
TopAbstractI. IntroductionII. Physical ImpairmentsIII. Functional LimitationsIV. Health-Related Qualify of...V. Summary and ConclusionsReferences |
|---|
Unlike other mitogenic growth factor signals, IGF-I stimulates both skeletal muscle cell proliferation and differentiation with increased creatine kinase and fusion of myoblasts into myotubes (12). Stimulation of muscle cell differentiation by IGF-I appears to be a biphasic response in which differentiation is stimulated at low concentrations of IGF-I but diminishes at concentrations above 100 ng/ml (13). There is also a temporal separation between IGF-I stimulation of the opposing processes of cell proliferation and differentiation. In most cell cultures, the proliferative response to IGF-I lasts 2–3 d, followed by myogenic differentiation (14). IGF-I increases myogenin mRNA (8). Because myogenin is one of a family of myogenic genes that transforms other cell types into the myogenic lineage (15), the effects of GH and IGF-I on myogenin action warrant further investigation.
More recently, studies have explored the effects of GH on myostatin as a possible mechanism of myogenic regulation. Myostatin is a cytokine that has recently been shown to selectively and potently inhibit myogenesis. There is evidence that myostatin-induced inhibition of myoblast differentiation is mediated through Smad 3 by interference with myogenic factor MyoD activity (16, 17). Myostatin inhibits cell proliferation and protein synthesis in C2C12 muscle cells and inhibits myoblast differentiation by down-regulating MyoD expression (18). These data are consistent with the notion that myostatin plays a critical role in myogenic differentiation and that the hypertrophy seen in the absence of this factor is possibly the result of increased proliferation and disrupted differentiation of myoblasts. However, myostatin might also have additional effects on muscle protein synthesis (17). The in vitro effects of GH on myostatin regulation were studied to investigate the mechanisms of anabolic actions of GH on skeletal muscle growth. GH treatment significantly reduced myostatin expression by myotubes, whereas GH receptor (GHR) antagonism resulted in up-regulation of myostatin in myoblasts. Given the potent catabolic actions of myostatin, these data support the theory that myostatin represents a potential key target for GH-induced anabolism (19).
B. In vitro effects of GH/IGF-I on differentiated skeletal muscle
It is difficult to completely dissociate the biological effects of GH from those mediated through its target growth factor IGF-I. Nevertheless, in contrast to the well-characterized effects of GH on differentiating skeletal muscle myoblasts, there is currently little evidence to support direct GH action on differentiated myocytes.
Technical difficulties have precluded the definitive demonstration of functional GHRs in differentiated skeletal muscle cells using conventional binding assays. Indirect evidence comes from GHR mRNA detection using RT-PCR-based approaches that have proven to be more feasible (20). Quantitative RT-PCR (20) suggests that the level of GHR in human skeletal muscle is low and highly variable (4–33% of that in human liver tissue) (21). Immunolocalization has been used to demonstrate that both the GHR and its posttranslational cleaved GH-binding protein (GHBP) product were present in a variety of tissues, including skeletal muscle (22). The interindividual differences in relative expression of the two isoforms of the GHR (GHR and GHBP) are reported to be greater than are levels between different tissues within a given individual (23). Immunohistochemical studies have confirmed the translation of both GHR and GHBP in a variety of rat tissues, including skeletal muscle (24).
Although substantial evidence exists that IGFs are important myogenic signals that mediate the effects of GH, this does not preclude the possibility that GH also has direct effects on skeletal muscle. Nevertheless, cell culture studies to date have had difficulty demonstrating convincingly any direct effect of GH on muscle (25, 26, 27). Two studies (28, 29) have reported direct effects of GH on isolated muscle cells in culture using cell lines that are not of embryonic origin. One of these studies found that BC3H1 cells expressed specific, functional GHRs that had metabolic responses to GH (29), whereas the other demonstrated that treating myoblasts (from a 10T1/2 cell line) with GH induced formation of myotubes (28). However, neither of these studies totally excluded the possibility of an autocrine and/or paracrine role for IGF-I in mediating these GH effects. Additional investigations are needed to confirm the presence of functional GHRs in human skeletal muscle to corroborate the hypothesis that GH may exert direct effects on skeletal muscle tissue. The underlying mechanisms by which skeletal muscle responds to GH must be delineated from those of locally synthesized IGF-I (or other mediators of growth such as myostatin) that may operate in an autocrine and/or paracrine fashion.
C. In vivo effects of GH/IGF-I status on differentiated skeletal muscle: animal studies
The majority of early studies investigating the effects of GH administration on skeletal muscle used isolated diaphragm muscle preparations in hypophysectomized rats (5). These studies all used high doses of GH and generally found an increase in amino acid uptake and protein synthesis, likely from early insulin-like effects of GH, rather than any longer term growth response of the target cells. Whole-animal experiments have confirmed that administration of GH to hypophysectomized animals increases muscle mass (30), whereas treating intact rats with a polyclonal antiserum to rat GH that decreased circulating IGF-I levels by 80–90% resulted in a reduction of weight, total protein, and RNA content of hind-limb musculature (31). Intact farm animals including pigs, lambs, and steer have been shown to be even more responsive to exogenous GH administration than laboratory rats (32, 33, 34), with the responses being largely dose-dependent. Similar to findings from isolated muscle preparations, the increase in muscle mass was due to increased rates of protein synthesis, rather than decreased rates of protein degradation as is typically seen with other anabolic agents.
Skeletal muscle fibers are generally classified as type I (slow oxidative) or type II (fast glycolytic) fibers. There are marked differences in speed of contraction, metabolism, and susceptibility to fatigue. Type I fibers are rich in mitochondria, produce energy using mainly oxidative metabolism yielding a long-lasting supply of ATP, and are fatigue-resistant. Conversely, type II fibers are comprised of three subtypes, IIa, IIx, and IIb. Type IIb fibers are at the opposite end of the spectrum from type I fibers in that they have the lowest levels of mitochondrial content and oxidative enzymes, rely on glycolytic metabolism as their major energy source, and are readily fatigued. The oxidative and contraction functions of type IIa and IIx lie somewhere between type I and IIb (35, 36, 37). Strength or resistance training typically results in increased muscle size. This increase is due to an increase in the size of individual muscle fibers (hypertrophy) and/or an increase in the number of muscle fibers (hyperplasia). The majority of evidence suggests that hypertrophy is the mechanism for increased muscle size with exercise training and anabolic interventions. Intrinsic muscle strength is higher in muscles with relatively more type II fibers. Thus, resistance exercise, anabolic steroids, and GH would be expected to induce greater increases in type II than type I fibers. Conversely, aerobic or endurance training would be expected to induce greater increases in the oxidative capacity of type I muscle fibers.
Combining GH treatment with exercise has been shown to result in larger medial gastrocnemius muscles than does exercise alone. Moreover, the combination of GH or IGF-I plus exercise in hind limb-suspended rats resulted in an increase in size of each predominant fiber type including types I, I + IIa, and IIa + IIx in the medial gastrocnemius musculature compared with untreated hind limb-suspended rats (38). These earlier studies suggested that IGF-I, as well as GH, when combined with exercise interact to maintain the medial gastrocnemius musculature, at least in the hypophysectomized rat model (38). However, more recent studies using genetic disruption of the GHR revealed no significant differences in motorneuron survival, differentiated skeletal muscle fiber diameter, or body weight in fetal mice (39). These recent data provide evidence more in favor of possible compensation by other signaling mechanisms of neurotrophic functions on differentiated skeletal muscle.
Chow et al. (40) demonstrated that tyrosine phosphorylation of JAK2 and STAT5 occurred in skeletal muscle as well as in liver of rats after systemic administration of GH. Although suggestive of a direct effect of GH on skeletal muscle, this does not confirm or refute the possibility that growth or myogenesis may be mediated by induction of local IGF-I expression in muscle. Although cultured myoblasts can release IGF-I (41), the relationship between GH administration and IGF-I expression in skeletal muscle remains unclear. For example, IGF-I mRNA levels in extrahepatic tissues have been shown to be normal in GHD dwarf chickens (42). Furthermore, skeletal muscle levels of IGF-I mRNA were unresponsive to GH administration in sheep, pigs, and cattle, despite increased muscle growth in the latter two species (43, 44, 45). Finally, endogenous production of muscle IGF-I mRNA was not impaired in hypophysectomized rats (46). These studies all suggest that skeletal muscle IGF-I expression may be at least partly GH-independent. Conversely, other animal studies have shown that circulating GH can influence IGF-I mRNA gene expression in skeletal muscle (47, 48) and that these effects appear to be more pronounced after GH than IGF-I administration in hypophysectomized rats (49). However, these effects of GH could also be mediated by autocrine or paracrine actions of locally produced IGF-I. Tissue-specific ablation of the hepatic IGF-I gene in adult mice supports the role of autocrine and paracrine-derived IGF-I. These mice exhibit normal body weight and tissue morphology, despite a 75% reduction in circulating IGF-I levels (50, 51). These data suggest that normal postnatal peripheral skeletal muscle tissue growth can occur under the influence of locally derived IGF-I. There are also data to suggest that muscle-specific expression of IGF-I plays a permissive role in muscle hypertrophy in aging and in regeneration of skeletal muscle fibers (52).
There are reports suggesting a transient increase in GHBP mRNA levels in skeletal muscle, heart, and liver, but not adipose tissues, in fasting rats after 1 d of GH treatment (53). Others have reported that increased dietary intake results in increased liver GHR mRNA, whereas reduced temperature increases skeletal muscle GHR mRNA levels (54). Taken together, these findings suggest that GHR and GHBP mRNAs are differentially regulated in various tissues and under different conditions. Some (14) have suggested that GH refractoriness in conditions of sustained fasting may help prevent anabolic effects at a time when maintenance of brain metabolism is more critical for survival.
Biopsy data have revealed a substantial reduction (50%) in the proportion of type I skeletal muscle fibers in the hind limb, including soleus and extensor digitorum longus muscles, of hypophysectomized rats (55). This reduction in the proportion of type I muscle fibers was accompanied by an increase in the number of transitional fibers (types IIC and IB). Treatment with GH results in rapid (11 d) normalization of the proportion of both the type I and transitional fibers in these hypophysectomized rats (55). At the other end of the spectrum, human GH transgenic mice (56) and rats with pituitary tumors that secrete excess GH (57) have also been shown to have an unexpected increased proportion of type I fibers, similar to the response seen with endurance training or aging. GH is also considered to play a role in regeneration of skeletal muscle after injury. GHR mRNA levels are increased early after ischemic injury and decline after skeletal muscle maturation, a response that is delayed in hypophysectomized compared with intact animals (58).
Data from animal studies are consistent and convincing that suppression or enhancement of the GH/IGF-I axis results in a respective reduction or increase in muscle mass, largely due to alterations in protein synthesis. The altered GH/IGF-I status affects not only the quantity but also the predominant type of skeletal muscle tissue present, where GH levels relate directly to the proportion of type I muscle fibers present. Type I muscle fibers are more oxidative and fatigue resistant but slower in contractile speed (59, 60, 61). Thus, hypophysectomized animals with a reduced proportion of type I fibers would be expected to be more readily fatigued, whereas those with GH excess and a higher proportion of type I fibers should be weaker but more fatigue resistant. The role that the GH/IGF-I axis plays in eliciting this phenotype transition and the extent to which it is fiber-type specific require further investigation. The effects of combining resistance exercise with GH or IGF-I appear to be additive, resulting in larger increases in skeletal muscle fiber size than any of these treatments alone. Taken together, these data suggest that skeletal muscle responses to exercise, GH, and IGF-I may be differentially regulated.
D. In vivo effects of GH/IGF-I status on differentiated skeletal muscle: human studies
1. Adults with GHD.
In contrast to animal studies, initial histochemical studies of thigh muscle biopsies from adults with untreated GHD revealed qualitatively normal muscle, with no obvious features of myopathy and no significant difference in mean fiber size or proportion (type I vs. type II fibers) compared with age- and gender-matched healthy adult controls (62, 63). The lack of between-group difference was suspected to be due to a lack of statistical power, given the small sample sizes. However, closer examination of these data (62) reveals that, although not statistically significant, there was a trend toward type I fibers being larger than the type II fibers both before and after 6 months of GH treatment. We have also demonstrated an increase in type I fiber size in adults with untreated GHD (64). These findings are in contrast to healthy control subjects who typically have larger type II muscle fibers that are recruited during brief high-intensity exercise and smaller type I fibers that are recruited during more prolonged lower intensity activities (65). Because the proportion of fiber types is a major influence on the contractile properties of skeletal muscle, speculation exists that the altered fiber composition may be partly responsible for the impaired muscle strength in adults with untreated GHD. However, findings of larger type I fibers are counter-intuitive and do not explain the increased perception of fatigue in adults with untreated GHD, because type I fibers are typically more fatigue resistant.
In terms of neuromuscular effects of GHD, Webb et al. (66) recently reported abnormal electromyographic (EMG) and biopsy findings suggestive of abnormal motor unit innervation in all seven biopsies of the biceps muscle of adults with GHD, albeit four of whom had childhood onset of disease. They hypothesized that axonal sprouting occurs during normal denervation and reinnervation of skeletal muscle in adults with GHD. This would result in larger motor units because muscle fibers are reinnervated by neighboring motor terminals that would explain the abnormal EMG and biopsy findings of abnormal clustering of type II fibers.
Difficulty in unraveling the true impact of GH treatment on skeletal muscle stems from the discrepancy between early muscle biopsy findings of minimal change in muscle fiber size and those of subsequent imaging studies documenting impressive increases in muscle mass. GH treatment of adults with GHD resulted in significant increases in thigh muscle mass, as determined by computerized tomography (CT) (62, 63, 67) and magnetic resonance imaging (MRI) (68). In contrast, earlier muscle biopsies of vastus lateralis muscle had shown no significant change in fiber size or proportion (62, 63). Conversely, others demonstrated significant and substantial increases in the size of both type I and type II fibers, with no change in fiber type distribution, after GH replacement in adults with GHD compared with placebo (64). Measurement of cross-sectional areas (CSAs) of muscle using CT and MRI scans can be affected by alterations in tissue hydration known to occur with GH replacement in adults with GHD. Thus, whether muscle fiber size is truly increased remains equivocal, with small biopsy studies reporting no change (62, 63) and a more recent trial reporting significant increases in mean fiber area of both type I and type II fibers (64). Whether the discrepancy in findings between earlier (62, 63) and later (64, 66) biopsy studies is a function of methodological differences or a true difference in muscle responses of the upper (biceps) vs. lower (vastus lateralis) extremities requires further investigation in larger controlled studies.
2. Adults with GH excess.
A single case study of a patient with acromegaly of 20-yr duration demonstrated mild muscular weakness and atrophy (69). Electromyography revealed normal firing and normal motor unit densities. Muscle biopsy revealed normal type I muscle fiber size and a range of type II fibers; some were hypertrophied, some were atrophied, and some were normal in size. It was hypothesized that excess GH hypertrophied some fibers, whereas disturbance of other endocrine functions resulted in atrophy of other type II fibers (69). In a larger study, muscle biopsies from 18 adults with acromegaly revealed hypertrophy of type I fibers in 50% of the individuals, with atrophy most often present in type II fibers (70). These data led to the generally accepted impression that GH excess results in muscles that appear larger, but are functionally weaker.
Taken together with the biopsy results for adults with GHD, these albeit limited data suggest that there is a window of optimal GH/IGF-I beyond which either increased or decreased levels of GH result in deleterious effects on skeletal muscle in humans. Biopsy data for adults with untreated GHD are different from those of animal studies, suggestive of large type I fatigue-resistant fibers and abnormal clustering of type II fibers. Despite being at the other end of the spectrum with untreated GH excess, adults with acromegaly also appear to have larger type I fibers, with atrophy most often present in the type II fibers that are characteristically capable of faster, more forceful contractions.
Unfortunately, there are extremely limited data and no controlled clinical trials systematically examining the impact of GH/IGF-I status on skeletal muscle morphometry in patients with acromegaly. Electron microscopy of the skeletal muscle in a single case study of a patient with acromegaly revealed amelioration of ultrastructural changes (altered mitochondria, glycogen granule infiltration, inclusion bodies, and vesicular dilatations) 9 months after surgical removal of the pituitary tumor that resulted in lowering of serum GH levels (71). In a consecutive series of 17 acromegalic patients who were evaluated for evidence of neuromuscular dysfunction before and 1 yr after pituitary adenomectomy, nine patients presented with myopathy evidenced by mild, strictly proximal weakness and flabbiness of muscles. EMG studies revealed typical myopathic abnormalities. However, serum muscle enzymes and muscle biopsy findings were essentially normal. Although the presence of myopathy was not correlated with the magnitude of GH elevation or any secondary endocrine derangement, it was associated with longer disease duration. Although the myopathic symptoms improved after surgical treatment of the acromegaly, some were still detectable 1 yr later (72). Clearly, studies using current approaches to the strict biochemical control of acromegaly will permit better elucidation of the impact of the GH/IGF-I axis on the muscular system structure and function.
E. Effects of GH/IGF-I status on protein metabolism
1. Adults with GHD.
Protein stores in humans are essential not only for use as contractile proteins in muscle but also to serve as a reservoir during illness when nitrogen must be mobilized from muscle to provide amino acids to the immune system, the liver, and other organs. Reduced protein mass, such as that seen in adults with GHD, limits adaptive functions, thereby compromising the ability of the body to withstand acute insult.
Adults with GHD have reduced LBM and skeletal muscle mass compared with healthy control subjects, suggesting an underlying abnormality in protein metabolism. The loss of lean tissue with untreated GHD is due to a negative nitrogen balance resulting from reduced protein synthesis and/or increased proteolysis. Whole-body protein turnover studies, using infusions of isotopically labeled leucine, have consistently demonstrated that adults with GHD have reduced protein synthesis compared with healthy controls (73, 74). However, the effects of untreated GHD on protein breakdown are less clear. Beshyah et al. (73) reported reduced rates of protein breakdown, whereas others (74) have since reported that the rate of protein breakdown is so diminished that the net protein loss is essentially normal. Reduced protein synthesis in combination with a normal rate of proteolysis would result in a net increase in protein oxidation and an ensuing loss of protein and lean body tissue. Because the loss in LBM and muscle mass cannot continue indefinitely, metabolic adaptations must occur so that LBM and muscle mass stabilize, albeit at a reduced level, in adults with GHD. Hoffman et al. (74) demonstrated that a decline in the rate of protein breakdown occurs to offset the initial fall in protein synthesis so that net protein loss is minimized. They suggested that normalization of protein oxidation may be a homeostatic mechanism that helps to partially restrain protein loss in adults with GHD. These findings remain to be confirmed in larger prospective studies.
The fundamental question with respect to GH replacement is whether it increases protein synthesis, resulting in increased muscle mass, or whether GH simply increases water, providing a false impression of enhanced LBM. The bulk of the evidence to date suggests that acute or short-term administration of GH results in increased protein synthesis, whereas more chronic administration appears to result in reduced proteolysis.
Adults with GHD who receive GH replacement have been reported to increase total LBM by as much as 11% (75) and thigh muscle mass by 5–8% (67, 76, 77), whereas indirect measurement of muscle mass (24-h urinary creatinine excretion) revealed a much larger 18% increase (75). Because skeletal muscle accounts for the majority (
50%) of total LBM, these data suggest that GH plays a major role in the regulation of whole-body protein metabolism in adults with GHD. The majority of studies to date suggest that, whereas insulin and IGF-I mediate a reduction in protein degradation, administration of GH results in enhanced protein synthesis.
Because all nitrogen in the human body is found in protein molecules, the amount of nitrogen can be used to estimate the amount of protein, where an increase in total body nitrogen (TBN) reflects an increase in protein content. Studies have consistently demonstrated that GH administration enhances nitrogen retention that is associated with an increase in LBM (75, 78). Nitrogen balance improves as urea production is diminished. This results in decreased urinary nitrogen and positive nitrogen balance, the net effect of which is increased LBM and, in particular, proteins and muscle. The increased protein accretion is accompanied by an increase in potassium and sodium retention. It has been suggested that GH induces nitrogen retention by shifting glutamine nitrogen away from urea synthesis to the periphery and by stimulating amino acid uptake and protein synthesis in skeletal muscle (79).
Isotopic studies have consistently demonstrated that short-term administration of pharmacological doses of GH in healthy control subjects (80, 81) and replacement doses in adults with GHD (82, 83) enhances protein synthesis but does not affect proteolysis. Thus, the increase in LBM, and therefore muscle mass, with short-term administration of GH appears to be due to enhanced protein synthesis. It remains possible that short-term GH administration may also reduce muscle proteolysis; however, this hypothesis remains to be convincingly demonstrated.
As in the situation with untreated GHD, there must also be a homeostatic adaptation to protein anabolism, because LBM cannot continue to accrue indefinitely. The time-course of this metabolic adjustment remains controversial. Johannsson et al. (84) demonstrated a sustained increase in TBN in patients with GHD who were treated with GH for 2 yr. Others have reported only a transient increase in TBN that peaked at 6 months and returned to baseline values by 1 yr (85). The majority of placebo-controlled trials report that the positive effects of GH administration on protein synthesis and body composition usually occur within the first few months of treatment (67, 75, 82, 83, 86). Results of a small (n = 4) metabolic ward study of adults with GHD suggest that the initial response of acute nitrogen retention, and thus protein accretion, occurs within 2–5 d and subsides after 1–2 wk of GH administration (87). Protein kinetics during the initial few months of GH administration in adults with GHD, as their LBM is expanding, appears to be different from that during chronic GH treatment when gains in LBM have stabilized. The duration of enhanced nitrogen retention after GH treatment remains controversial and warrants further investigation.
The anabolic effects of GH therapy may result from both systemic and local changes in metabolism. The underlying mechanisms responsible for the increased protein accretion remain to be delineated. Although there is evidence from animal models that the effects of GH may be mediated through circulating and/or local production of IGF-I, there is additional evidence to suggest that the mechanisms by which IGF-I and GH promote protein anabolism are distinct. The effects of modulating systemic levels of GH or IGF-I on regulation of muscle physiology have not been well investigated, especially in humans. It is known that up-regulation of local IGF-I expression is associated with increased muscle protein and DNA synthesis. Local production of IGF-I stimulates satellite nuclei to proliferate, differentiate, and fuse with myofibers to maintain, or re-establish, the myonucleus-to-myofiber size ratios of the enlarged muscle fibers (88). That systemic levels of GH or IGF-I can effect local changes in muscle is suggested by data from forearm infusions of GH or IGF-I where increased local muscle protein synthesis occurred (80, 89, 90). Fryburg and Barrett (89) demonstrated that systemic infusion of GH resulted in acute stimulation of muscle, but not whole-body protein synthesis. These data suggest that systemic GH may acutely, and specifically, regulate muscle protein metabolism.
In a well-designed study, Copeland and Nair (91) demonstrated the leucine-sparing action of GH in whole-body metabolism and evidence for acute stimulation of muscle protein synthesis in 15 young healthy males. They investigated the effects of an acute dose of GH on amino acid metabolism while controlling for the effects of other confounding hormones (i.e., via simultaneous infusion of somatostatin and replacement of insulin, glucagon, and GH). They demonstrated direct effects of GH in terms of increased anabolic effects via inhibition of amino acid oxidation and stimulation of whole-body protein synthesis. The direct effect of GH for this action was supported by the fact that insulin, glucagon, cortisol, IGF-I, catecholamine, and glucose concentrations were not different between the control and GH treatment groups. Fryburg et al. (80, 92) also reported increased arm muscle protein synthesis after chronic GH infusion into the brachial artery. However, this increase occurred only after longer exposure to GH than the 7-h study by Copeland and Nair (91). Comparison between whole-body and leg kinetic data suggests that the acute GH-induced increase in whole-body protein synthesis occurs primarily in nonskeletal muscle tissues. Yarasheski et al. (93) reported an increase in whole-body protein synthesis but not muscle protein synthesis of the quadriceps during chronic administration of GH, combined with exercise. However, these latter studies were conducted in healthy, non-GHD young men. It has been hypothesized that the increased muscle mass after longer-term GH treatment of adults with GHD (67, 94, 95) is a chronic effect of inhibited proteolysis, mediated by IGF-I.
It has also been suggested that muscle IGF-I and one of its binding proteins (IGFBP-4) may regulate muscle protein anabolism. Specifically, increasing systemic GH may enhance the local effects of IGF-I and IGFBP-4 in skeletal muscle. In vitro, IGF-I stimulates satellite myoblasts to express myogenin, which mediates the differentiation of myoblasts to myotubes and to mature myocytes (8, 14, 96, 97). It is possible that GH may directly increase the total number of myocytes, thereby increasing protein synthesis and local production of IGF-I mRNA in muscle, and/or that GH may stimulate mature myocytes to increase autocrine expression of IGF-I, as occurs in vitro when differentiated skeletal muscle cells secrete more IGF-I in response to a stretch stimulus (98). Because IGFBP-4 inhibits the mitogenic actions of IGF-I (99), it is also plausible that GH treatment may decrease levels of IGFBP-4, further facilitating IGF-I action. In hypogonadism, induced by administration of GnRH (GnRH) analogs, treatment with GH significantly increases skeletal muscle androgen receptors and IGF-I while decreasing IGFBP-4. This suggests that systemic therapy with GH has the potential to up-regulate muscle protein synthesis (98). Finally, GH has been shown by some (100), but not others (101), to increase myosin heavy chain expression in skeletal muscle.
The effects of systemic administration of GH on the change in local regulators of skeletal muscle protein anabolism (i.e., local IGF-I, IGFBPs, myostatin) and muscle catabolism (ubiquitin and proteasome activity) remain largely unknown. It has been suggested that myostatin, a member of the TGF-ß superfamily, may possibly be a negative regulator of skeletal muscle growth in adults (for review, see Refs. 102 and 103). Myostatin is expressed uniquely in human skeletal muscle as a 26-kDa mature glycoprotein, is present in both type I and type II muscle fibers, and is secreted in serum. Marcell et al. (104) also demonstrated a negative correlation between GHR and myostatin mRNA levels in elderly, healthy men. This led them to suggest that age-related deficits in GH may result in increased myostatin expression and a disassociation in autocrine IGF-I effects on muscle protein synthesis that may contribute to age-related sarcopenia. As previously discussed, myostatin appears to be a potent, selective inhibitor of myogenesis. Moreover, significant inhibition of myostatin after 18 months of GH treatment of adults with GHD has recently been demonstrated (19).
From the aforementioned studies, it appears that acute effects of GH infusion or short-term use appear to result from increased protein synthesis by GH itself. Conversely, the anabolic effects associated with long-term use of GH on skeletal muscle appear to be the result of inhibition of proteolysis, more likely mediated through IGF-I, resulting in reduced protein degradation.
2. Adults with GH excess.
Although few studies have examined abnormalities of protein metabolism in patients with acromegaly, their physical characteristics and mesomorphic presentation suggest possible increased protein metabolism. Excess GH results in a volume expansion due to sodium and water retention (105). Isotopic dilution studies can be used to estimate total body water (TBW), extracellular water (ECW), and plasma volume; however, intracellular water is calculated indirectly by combining measurements of TBW and ECW. The fundamental question is whether adults with acromegaly have increased protein synthesis, and therefore LBM, or simply increased water and bone mineral content. Studies combining dual energy x-ray absorptiometry (DEXA) and sodium dilution techniques have demonstrated increased LBM in adults with acromegaly that was due to increased ECW but not body cell mass (BCM) (106). Few studies have used total body potassium (TBK) counting to assess BCM, which is the metabolically active and protein-rich intracellular tissue (107), in adults with acromegaly. Adults with acromegaly have been shown to have increased BCM compared with healthy control subjects (108, 109). However, in a study of 18 acromegalic patients, TBK and BCM were elevated compared with predicted values (based on age and height) but not compared with healthy control subjects (110). Further studies are required to delineate the specific effects of excess GH on body cell and protein mass because the modulation of local protein dynamics remains poorly defined.
Successful treatment of patients with acromegaly has provided an opportunity for the examination of the effects of excess GH on skeletal muscle metabolism. GH excess results in insulin resistance with diminished peripheral glucose uptake (111, 112, 113). Whether this antagonism of insulin action affects protein and amino acid metabolism after surgical cure for acromegaly was recently investigated (114). Protein kinetics was evaluated in both the postabsorptive and insulin-stimulated states using leucine tracer infusion and euglycemic insulin clamp. Protein kinetics was normal in the postabsorptive state; however, there was marked resistance to the antiproteolytic effects of insulin during the insulin clamp. Although insulin did not reduce proteolysis to the same extent as in healthy controls, leucine oxidation was reduced, with a shift of amino acids toward protein synthesis during hyperinsulinemia. Six months after pituitary surgery, insulin resistance was normalized in terms of glucose disposal, whereas dysfunctional protein metabolism persisted. Additionally, short-term fasting studies suggest a shift toward increased fatty acid oxidation as the preferred fuel in patients with acromegaly, thereby sparing amino acids from oxidation and conserving protein (115, 116). These findings support the notion that enhanced lipolysis may underlie the protein-sparing effect associated with GH excess (116).
F. Effects of GH/IGF-I status on lean body mass
1. Adults with GHD.
Body composition in adults with untreated GHD has been measured using a variety of methods, including anthropometry (skinfold thickness, ratio of waist to hip circumference), measurement of TBW using dilution of tritiated water or deuterium oxide, nuclear techniques measuring TBK, bioelectrical impedance analysis (BIA), and, more recently, radiographic measures including DEXA, CT (117), and MRI (118). These methods vary in their accuracy, ease of use, and comparability with the historic reference standard because each method is based on different assumptions that may or may not be valid in adults with GHD (119, 120). Thus, interpretation of results of body composition measurements must take into account the limitations of the methods of measurement used. Nonetheless, over the last 15 yr, extensive study of body composition in adults with GHD has established a pattern of abnormalities that include increased sc and visceral fat mass with reduced LBM compared with healthy control subjects (121, 122, 123, 124, 125, 126, 127). The reduction in LBM is greater in untreated adults with childhood-onset GHD compared with age-matched GH naive adults with adult-onset GHD (128), suggesting that GH plays a role in somatic development. Skeletal muscle mass remains difficult and impractical to quantify, although models are now being used to attempt to address this issue (129). Nonetheless, a strong correlation has been noted between LBM (measured as TBK) and total thigh muscle area in healthy controls and adults with GHD, both before and after GH treatment (76). Because skeletal muscle represents a relatively fixed proportion of LBM of around 50%, measurement of LBM and its response to GH replacement in adults with GHD reflects changes in the skeletal muscle compartment.
The vast majority of cross-sectional studies have compared the body composition of adults with untreated GHD to that of either historical (123) or age-, gender-, height-, weight-, and BMI-matched healthy controls (122, 130, 131, 132, 133, 134, 135, 136), using a two-compartment model of fat mass and LBM. Although studies using DEXA (133, 134) and TBW (123) revealed no difference, those using BIA (122, 130) demonstrated significantly reduced LBM (–3 to –9 kg) in adults with GHD. It has been argued that this discrepancy may be a function of the method used to measure LBM, because BIA is based on the assumption that electrical conductance is related to the water content of the body, which is primarily located in fat-free tissues, and is more affected by altered hydration.
Although there is general agreement that fat mass is higher and LBM slightly lower in adults with GHD (137), there is no definitive evidence as to whether the reduced LBM associated with GHD reflects a reduction in metabolically active muscle or simply reduced water content of muscle. Using BIA as an indirect measure of both TBW and LBM, two early studies demonstrated that the decreased conductance in GHD adults reflected reduced total water and ECW content that was out of proportion to an independent estimate of LBM (121, 122). The indirect techniques used in these studies did not take into account the components of LBM and BCM. Furthermore, conductance through the skin may be decreased by the reduced sweat content of skin known to occur in GHD. Rosen et al. (123) compared body composition (weight, TBW, and TBK) in adults with GHD to normative values predicted from their height, weight, age, and gender. They reported a reduction of ECW and TBW, in the order of magnitude of 2.4 kg in males and 3.2 kg in females, with no change in BCM in adults with GHD compared with predicted values. However, measurement of fluids using total potassium counts is based on the assumption that the intracellular concentration of potassium is normal in adults with GHD, which has not been confirmed. Hoffman et al. (125) used a four-compartment model to compare body composition in adults with GHD to age-, gender-, height-, and weight-matched controls. Using DEXA and isotopic sodium dilution techniques (a direct measure of ECW), they demonstrated that although fat-free soft tissue mass was clearly reduced, the ECW proportions of fat-free soft tissue mass were similar in the GHD and healthy controls. Jorgensen et al. (132) assessed hydration state using BIA and body composition using CT and DEXA. This group also demonstrated normal lean tissue hydration in untreated GHD adults. Findings of these two studies infer that TBW is reduced proportionately to the reduced LBM in adults with untreated GHD. Thus, it appears that LBM is reduced in adults with untreated GHD.
Regional tissue comparisons using single-slice CT studies of the midthigh support the reduced LBM because they have consistently demonstrated that adults with GHD have reduced thigh muscle mass and increased thigh fat mass, compared with their healthy control counterparts (67, 138). In adults with GHD, the muscle-to-fat ratio is 65:35% (67), whereas healthy normal controls typically have a ratio of 85:15% (139, 140). The reduced skeletal muscle and increased thigh fat mass would lead one to expect reduced overall LBM and absolute muscle strength in adults with GHD.
Although it was demonstrated in 1959 that GH promoted nitrogen retention (141), the first placebo-controlled clinical trials of GH to establish increased LBM in adults with GHD (67, 75) were not done until some 30 yr later. Both trials reported altered body composition abnormalities, including 7–8% ( 4 kg) lower LBM and 7% higher fat mass in adults with GHD. These abnormalities were ameliorated following replacement doses of GH. Since then, a variety of different methods have been used in randomized, placebo-controlled studies to examine the effects of GH replacement on body composition in GHD adults. Essentially, GH replacement in adults with GHD has been shown nearly to normalize body composition over time (137).
The main target for the anabolic effects of GH is the LBM, which is comprised of skeletal muscle and visceral organs. After GH replacement in adults with GHD, the gains in LBM are generally reported to offset the loss of fat mass, so that overall body weight remains unchanged. LBM has been shown to increase by a mean of 2–5.5 kg in adults with GHD (75, 77, 117, 124, 137, 142, 143, 144, 145, 146, 147, 148). The effects of GH replacement on LBM apply to men and women with GHD (149, 150) and to the elderly (151), and have been equally observed in patients with adult- and childhood-onset of GHD (128), with a particularly profound effect on somatic maturation in young adults (152, 153, 154). In terms of regional tissue changes, GH replacement has been shown to result in significant increases in thigh muscle CSA (68, 76, 155).
Studies of GH replacement in adults with GHD have consistently shown TBW replenishment of around 2–4 kg (117, 142, 144, 147, 155, 156, 157, 158), which appears to occur rapidly within the first few weeks of treatment (156, 159, 160). In healthy subjects, muscle contains approximately six times the total water and twice the ECW content of fat mass (161). Because GH administration results in an increase in LBM and a reduction in fat mass, the rise in TBW accompanying GH administration may be an epiphenomenon, reflecting these changes. Conversely, the increase in TBW documented after GH replacement in GHD adults may represent replenishment of water to previously dehydrated muscle cells, without an increase in muscle fiber hypertrophy. The study by Jorgensen et al. (132), in which hydration state was assessed by BIA and body composition by CT and DEXA in GHD adults before and during GH replacement, demonstrated normal lean tissue hydration in untreated GHD adults but alterations in lean tissue hydration during long-term GH replacement. The authors attributed this to overhydration due to supraphysiological GH dosing regimens used. Early studies of GH replacement in adults with GHD used high dosing regimens based on weight, because the dose was derived largely from the pediatric experience in which GH was used to achieve increased stature in children. It rapidly became apparent that GH dose requirements in adults were even lower than had been anticipated and significantly less than those used in childhood. Moreover, the supraphysiological IGF-I levels that resulted from these regimens (162) brought into question whether some of the observed effects of GH on skeletal muscle represented pharmacological effects related to its antinatriuretic actions. However, subsequent studies using lower GH dose regimens, resulting in IGF-I levels within the age-adjusted normal range, have confirmed that the effects of GH on LBM and skeletal muscle occur with physiological as well as pharmacological treatment (163). Moreover, long-term follow-up studies of prolonged GH substitution, of up to 7-yr duration, in severely GHD, hypopituitary adults have demonstrated sustained acquisition of muscle mass (164, 165).
2. Adults with GH excess.
GH is an anabolic hormone with the capacity to induce a positive nitrogen balance, stimulate protein synthesis, and increase lipolysis in adipose tissue. Given the excessive secretion of GH and IGF-I in acromegalic patients, one would expect increased LBM and a corresponding reduction in fat mass in these patients.
Skeletal muscle abnormalities associated with acromegaly (70, 71, 166) were observed before current techniques for evaluation of body composition. Few studies have systematically investigated the characteristic body composition changes in patients with acromegaly. Nonetheless, an increase in body weight, LBM, and ECW with a corresponding reduction in fat mass have been consistently observed (106, 167, 168, 169, 170). Only one study has investigated body composition and energy expenditure in acromegalic patients and compared them with age-, gender-, weight-, and height-matched healthy control subjects (106). The authors found that the increase in LBM observed in patients with acromegaly was due to a corresponding expansion of ECW but not BCM (106), in contrast to the findings of others (108, 167). The latter, taken together with the finding of increased TBN (167), suggest that the impact of GH excess in acromegaly is primarily on the synthesis of extracellular protein (i.e., collagen), with little effect on skeletal muscle tissue itself. It should be emphasized that acromegaly is a chronic disease often accompanied by several endocrine derangements including hypogonadism. The latter feature may, at least partically, contribute to interruption of the putative anabolic effects of GH on skeletal muscle.
Alterations in body composition toward normal have been observed in patients with acromegaly in a number of studies using different treatment modalities and different body composition measurement methods (106, 167, 171, 172).
Modern treatment algorithms for acromegaly include modalities with widely differing mechanisms of action and degree and speed of modification of GH/IGF-I status (173). However, the ability of these therapies to normalize body composition (i.e., reduce LBM and increase fat mass) appears to be dependent on the resulting GH/IGF-I status rather than mode of treatment (172). The impact of treatment on body composition occurs after radiotherapy (167), medical therapies (106, 174), or successful pituitary surgery (109, 175). A recent study failed to demonstrate any effect of 3-month treatment with a long-acting somatostatin analog on LBM (176), whereas a study of withdrawal of long-term bromocriptine therapy demonstrated reversal of body composition parameters back toward that expected in active acromegaly (177). Data relating to the intermediate-term effects of the GHR antagonist pegvisomant on body composition suggest that its effects are similar to those of the GH inhibitor octreotide (178).
Biochemical criteria for remission of acromegaly have become more stringent with the development of highly sensitive and specific GH assays and their use in conjunction with measurement of serum IGF-I levels (179). Such advances make more powerful, tailored, and therefore effective treatment of acromegaly possible. More aggressive treatments such as surgical adenomectomy and the GHR antagonist could result in depression of circulating GH/IGF-I levels consistent with a state of functional or actual GHD in formerly acromegalic patients. Such marked reduction in IGF-I levels could be associated with a theoretical risk of development of the adverse body composition associated with adult GHD (125). However, to date, no data to confirm or refute this hypothesis in the acromegalic population have been demonstrated.
G. Effects of GH/IGF-I status on skeletal muscle performance
1. Adults with GHD.
Based on reduced absolute lean body and muscle and the tissue changes seen with biopsy data from adults with GHD, a reduction in absolute maximum isometric force generating capacity, isokinetic torque production (particularly at slower velocities), and improved local muscular endurance would be expected. Indeed, maximum voluntary isometric quadriceps force production has consistently been found to be reduced in adults with GHD, compared with healthy controls (Table 1) (68, 138, 180, 181). Isokinetic results for the knee flexors and extensors have been more variable, with the majority of studies reporting torque values in the low normal range in adults with GHD compared with healthy controls (68, 181, 182).
|
Adults with GHD have increased subjective complaints of fatigue. This led to the speculation that untreated GHD may result in an increased percentage of type II muscle fibers, which are more readily fatigueable than slower type I fibers. Indeed, findings from a small (n = 14) open-labeled study demonstrated that reduced maximal isometric strength was accompanied by contractile changes suggestive of a shift toward an increased percentage of type II muscle fibers that may contribute to the increased fatiguability in adults with GHD (180). However, a subsequent study reported that although absolute values of strength and fiber CSA of the quadriceps musculature were significantly lower in adults with childhood-onset GHD compared with healthy controls, once strength and fiber CSA were normalized for quadriceps CSA and subject height, respectively, any differences disappeared (183). Furthermore, there was no difference in the quadriceps muscle twitch characteristics, fatigue index, or fiber type distribution between those with GHD and healthy controls. Their results suggest that weakness and fatigability, at least in childhood-onset GHD, do not have a peripheral or skeletal muscle origin but may originate more centrally, perhaps involving cardiovascular and/or neural structures (183). Although these findings remain to be confirmed, the suggestion that there are two entities with very different clinical presentations and response to GH treatment, one developmental of childhood onset and one metabolic of adult onset, cannot be discounted (148).
Another possible explanation for reduced intrinsic muscle strength is abnormal neuromuscular function in adults with GHD. EMG and biopsy studies of the biceps musculature of 20 adults with GHD, of whom 10 had childhood-onset of disease, revealed altered motor unit recruitment compared with healthy controls (66). Adults with untreated GHD had abnormal clustering of type II fibers, suggestive of abnormal motor unit innervation. The authors hypothesized that axonal sprouting occurred during normal motor unit remodeling in adults with untreated GHD. As motor neurons become denervated, muscle fibers may have become reinnervated by neighboring motor terminals, resulting in larger motor units. This would explain the abnormal EMG and biopsy findings and the clinical symptoms of increased fatigue. The fact that stretch and tissue overload models in rabbits have increased expression of IGF-I mRNA in type I muscle fibers (184) led Webb et al. (66) to further suggest that neural trophism in adults with GHD may be reduced due to decreased IGF-I synthesis specifically in type I muscle fibers that may promote reinnervation by type II fiber nerve terminals.
In summary, adults with GHD have reduced absolute maximal isometric, and possibly isokinetic, muscle strength that is at least partly due to reduced muscle mass. Although findings from earlier studies suggested that intrinsic muscle strength might be reduced in these patients, more recent data refute this notion.
Isometric quadriceps muscle strength correlates with quadriceps muscle volume in adults with GHD (67). However, there is no definitive evidence that an increase in LBM, and therefore muscle mass, after short-term GH replacement in adults with GHD translates into improved muscle strength. Several studies have demonstrated increased thigh muscle size after short-term GH administration in adults with GHD that was not accompanied by improved muscle strength (62, 63, 67, 68, 76, 77, 182). The only placebo-controlled trial to report improved isometric strength after short-term GH treatment evaluated strength of the proximal hip flexor musculature (76). Placebo-controlled trials (Table 2) revealed no change or a slight decrease in isometric and isokinetic quadriceps strength after GH treatment of 6-month duration or less (67, 76, 132, 144, 181, 185, 186). Only trials prone to experimental bias, that is open-labeled trials (180, 181) and those studies where a subset of participants continued on active GH treatment in an open-labeled fashion after the completion of the placebo-controlled trial (144, 185, 186, 187), reported improved maximal isometric strength after treatment. Janssen et al. (68) reported improved isokinetic, but not isometric, quadriceps muscle strength. Besides being open-labeled, another common feature of the trials reporting increased isometric (144, 180, 181, 185, 186, 187) or isokinetic (68, 181) strength is that they were of longer duration (i.e., exceeding 1 yr). Of interest, Johannsson et al. (181) reported a transient decrease in isometric knee extensor strength after 6 months of treatment and found that strength gains were greater in those who were younger or initially weaker than predicted values for their age and gender at baseline (181).
|
|
Interpretation of these findings is complicated by differences in severity of GHD, the dose of GH administered, and the fact that the majority of the placebo-controlled trials used a shorter treatment period (4–6 months) than the open-labeled studies (1–3 yr). However, given that protein synthesis, muscle fiber size, and LBM increase within the first few months whereas increases in muscle strength take much longer (more than 6 months), it is unlikely that the reported increases in muscle strength can be attributed to the anabolic effects of GH.
GH treatment increases muscle mass in adults with GHD but does not appear to translate into improved muscular performance. A hind limb-suspension atrophy model in hypophysectomized rats demonstrated that GH or high-intensity exercise alone had minimal effect on the mass of the unloaded muscle, whereas a combination of the two produces strong interactive effects (188). These data suggest that the endocrine action of GH/IGF-I, the paracrine/autocrine effects of IGF-I, and neuromuscular activity are all important factors that affect muscle size and function. The optimal GH treatment regimen for increasing muscle function in conjunction with exercise training to improve neural activation remains a critical question.
Local muscle endurance of the quadriceps remains unchanged (64) or is reduced (181) after GH administration in adults with GHD. Johannsson et al. (181) concluded that the increase in muscle mass and strength they saw with GH does not result in the maintenance of capacity to produce force during repeated contractions. They suggested a lack of parallel metabolic adaptation in muscle in terms of muscle glucose storage and utilization, oxidative enzyme activity, and capillarization as explanations for this deficit. Findings of larger and proportionally more slow twitch, fatigue-resistant, type I fibers after GH treatment remain discordant with the lack of improvement in local muscle endurance (64, 180). However, the discrepancy between improved muscle strength and reduced local muscular endurance of the quadriceps may reflect the difference in methods used to quantify local muscle endurance. Johannsson et al. (181) measured fatigue index as the percent reduction in peak torque between the first and last three repetitions during 50 repeated concentric contractions at a velocity of 180°/sec. Because the maximum force generated increased after GH, the reported "increased fatigue index" reflects that the individuals had higher mean peak torque values during the first three contractions and/or lower values during the final three contractions after GH treatment compared with values at baseline. This suggests that, after GH, participants were actually able to do more absolute work during this test. A better indicator of local muscle endurance would be to measure the area under the curve or the total volume of work done over the 50 repetitions, rather than percent decline from the initial to final contractions. Similarly, future studies should evaluate changes in isotonic local muscle endurance by establishing a relative load [i.e., percentage of 1 repetition maximum (RM)] at baseline and then using that same absolute load both before and after interventions. This eliminates any change in absolute loads that would occur should 1 RM change after the intervention.
In summary, short-term (4–6 month) GH replacement in adults with GHD induces an increase in lean body and skeletal muscle mass. The fact that the ability to functionally use this increased mass (measured in terms of muscle strength) takes so much longer suggests that the anabolic effects of GH do not include increased muscle strength. Although it is encouraging that longer-term GH treatment can partially offset the physiological effects of muscle weakness associated with GHD, longer duration randomized controlled trials with larger sample sizes are required to objectively evaluate the physiological and functional effects of this therapy. Because improved physical capacity (of which strength is one component) is influenced by both the hormonal milieu and physical training, further work is needed to evaluate the role of physical exercise, alone and in combination with GH, to achieve optimal improvements in muscle performance.
2. Adults with GH excess.
A single case study report of mild muscular weakness and atrophy (69) was corroborated by muscle biopsy findings of hypertrophy of type I and atrophy of type II fibers in adults with acromegaly (70) over two decades ago. Although these data led to the generally accepted conclusion that GH excess results in skeletal muscles that are large but functionally weak, this hypothesis remains to be tested. There are very limited studies specifically examining the impact of GH changes on skeletal muscle strength in acromegalic patients. Nevertheless, as a model for studying long-term effects of GH on muscle strength, transgenic mice for the rat phosphoenolpyruvate carboxykinase-bovine GH fusion gene were studied. This transgene was expressed predominantly in the liver and kidney, but not in skeletal muscle (189), with an associated increase in circulating GH and IGF-I levels. These animals demonstrated increased weights of forelimb and hind-limb muscles. However, muscle weight/body weight ratios were in fact nearly 20% smaller than control nontransgenic littermates. Moreover, forelimb grip strength did not increase proportionally with muscle weight (189). Whether similar findings occur in patients with GH excess remains to be investigated.
To date, there are no published placebo-controlled or open-labeled trials investigating the effects of successful treatment of acromegaly on muscle function as determined by measures of skeletal muscle performance (i.e., muscle strength or local muscular endurance). In light of the advances in the treatment of acromegaly (190, 191), it seems appropriate to evaluate the impact of these costly interventions on physical performance and functional abilities in patients with varying degrees of GH/IGF-I control. In fact, we would argue that findings from such studies may assist in defining more objective, evidence-based treatment endpoints for this disease.
H. Effects of GH/IGF-I status on aerobic performance
1. Adults with GHD.
Measurement of maximal oxygen uptake (VO2max), more commonly known as aerobic capacity or the maximum ability to take in and use oxygen, is the most widely accepted test of physical work capacity (192, 193). Oxygen consumption measured during a progressive, continuous cardiopulmonary exercise test allows for objective determination of an individual’s VO2max and evaluation of their linked cardiovascular, pulmonary, and muscular systems (194).
Studies evaluating maximal aerobic capacity have consistently demonstrated markedly reduced VO2max (18–28% deficit) in adults with GHD (77, 95, 195) compared with predicted values based on formulae that include age, weight, and height (196, 197) (Table 3). By comparison, VO2max in adults with untreated GHD is reduced to levels comparable to those observed in individuals with congestive heart failure (198). Moreover, these deficits were evident even when using the cycle ergometer for testing when, unlike for treadmill testing, the increased body weight is supported by the cycle and yields no additional work for the participant. The proposed underlying mechanisms responsible for this deficit include reduced skeletal muscle mass or altered metabolism (95), diminished cardiac capacity (137, 199, 200, 201), and reduced red blood cell volume due to impaired IGF-I mediated erythropoiesis (124, 157, 202).
|
VeT occurs at a significantly higher percent of VO2max, owing to a significantly reduced VO2max in adults with GHD (73%) (64) compared with healthy controls (45–65%) (211, 212, 213, 214, 215). This may help to explain the perception of increased fatigue in adults with untreated GHD, because execution of ordinary activities of daily living would require oxygen supplies that exceed their VeT, resulting in increased fatigue and discomfort. Conversely, healthy adults would easily be able to perform the same activities with less fatigue, because they would be working at intensities requiring oxygen supplies below their VeT that could easily be performed using predominantly aerobic energy supplies.
Jorgensen et al. (67) demonstrated increased maximal cycle work rate (WRmax) after 4 months of GH treatment in a double-blind, placebo-controlled (DBPC) crossover trial in 22 adults with GHD of childhood onset, with a 4-month washout between interventions (Table 4). A subset of participants who agreed to continue taking GH were then followed in an open study and reevaluated after 16 (187) and 38 (143) months. Similar to their findings for muscle strength, there was a further increase in WRmax after 16, but not 38 months of follow-up. Degerblad et al. (182) reported a within-group increase in WRmax after 3 months of GH replacement in adults with GHD of childhood onset; however, no comparison was made between the treatment groups. Similar findings of improved WRmax have been reported in individuals with adult-onset GHD. Several DBPC studies of 6- to 12-month duration have consistently demonstrated significant within-group increases (14–19%) in WRmax after GH, without significant change (–6 to 6%) in the placebo groups (77, 95, 132, 195, 216). Some of these studies (77, 95, 132) have also reported significant increases in WRmax after GH compared with placebo treatment. Conversely, studies measuring maximal exercise time walking on a treadmill report no significant improvement with GH over placebo during a 6-month DBPC trial, but a significant increase after a further 6, 12, and 18 months of the open-study follow-up (144).
|
