- Email: [email protected]
8 Acquired growth hormone resistance in adults
8 Acquired growth hormone resistance in adults R. C. JENKINS
BMedSci, MB ChB, MRCP
Clinical Research Fellow
R. J. M. ROSS* MD, FRCP Senior Lecturer and Consultant in Endocrinology
Department of Medicine, UniversiO, o[" Sheffield, Clinical Sciences Centre, Northern General Hospital, Herries Road, Sheffield, $5 7AU, UK
Acquired growth hormone resistance (AGHR) may be defined as the combination of a raised serum growth hormone (GH) concentration, low serum insulin-like growth factor-1 (IGF-1) concentration and a reduced anabolic response to exogenous GH. A wide range of conditions exhibit the syndrome to a variable degree, including sepsis, trauma, bums, AIDS, cancer, and renal or liver failure. The primary defect seems to be a reduction in IGF- 1 concentration which then leads to increased GH concentration by a loss of negative feedback. It is not clear whether IGF-1 concentration falls because of decreased production or increased clearance from the circulation, or both. Treatment to reverse the biochemical defect by restoring IGF-1 levels, either by the administration of GH or IGF-1, has resulted in improvements in a wide range of metabolic parameters and, more recently, to definite clinical benefit in well-defined groups, such as patients with AIDS. These results cannot be extrapolated to other groups with AGHR as a recent unpublished report suggested increased mortality in critically ill patients treated with GH. Research needs to focus on the molecular basis of AGHR if we are to develop therapies for catabolism. Key words: growth hormone; insulin-like growth factor-I; catabolism,
In 1931 Cuthbertson described increased urinary nitrogen excretion following trauma and similar, more refined, observations continue to be made (Arnold et al, 1993). Analogous phenomena have been observed in patients suffering from diverse conditions such as sepsis, burns, renal failure, liver disease, AIDS, cancer or after surgery. It has become apparent that the ensuing protein-calorie malnutrition both adversely affects prognosis and is difficult to correct by improving nutritional intake alone. Improvements in the management of patients with these and other diseases * Address correspondence to Dr R.J.M. Ross.
Baillibre's Clinical Endocrinology and Metabolism-315 Vol. 12, No. 2, July 1998 Copyright 9 1998, by Baitli~re Tindall ISBN 0-7020-2465-1 All rights of reproduction in any form reserved 0950-351 X/98/020315 + 15 $12.00/00
316
R. C. J E N K I N S A N D R. J. M. ROSS
have led to prolongation of survival and an increasing need to develop ways to correct the associated chronic malnutrition and thus to optimize recovery (Wilmore, 1991). The concept of acquired growth hormone resistance results from a recognition of a similar pattern of disturbance of the growth hormone/ insulin-like growth factor- 1 (GH/IGF- 1) system in a range of disparate but related conditions sharing the common feature of catabolism. The combination of high levels of growth hormone (GH), low levels of IGF-1 and a reduced anabolic effect of exogenous GH provides a simple biochemical definition of the condition. It is becoming increasingly clear that this is an over-simplification and takes no account of the effects of the growth hormone-binding protein (GHBP), insulin-like growth factor bindingproteins (IGFBP), free IGF-1 or the half-lives of GH and IGF-1. In health, GH is secreted in a pulsatile fashion by the somatotroph ceils of the anterior pituitary gland with secretion being stimulated by growth hormone-releasing hormone (GHRH) and the as yet unknown endogenous GH-releasing peptide (GHRP), and inhibited by somatostatin and negative feedback by IGF-1. GH is carried in the circulation by a specific binding protein, GHBP, levels of which are thought to reflect levels of GH receptor expression as both proteins are derived from the same gene. Circulating GH has direct and indirect actions (Berneis and Keller, 1996). It acts directly at the liver to stimulate IGF-1 secretion, which mediates the indirect anabolic actions of GH, and also production of IGFBP-3 and the acid-labile subunit (ALS). IGF-1 is not only controlled by GH, other factors such as fasting and nutritional status are also important determinants. Clearly, a simple measure of the levels of GH and IGF-1 gives only a crude assessment of the state of this complex hormone system. This chapter describes the evidence for the acquired GH resistance syndrome, and outlines the disturbances seen in the GH/IGF-1 system in the associated conditions. The possible mechanisms underlying the condition, the clinical significance and the effects of specific treatments will be discussed.
EVIDENCE FOR THE EXISTENCE OF THE ACQUIRED GH RESISTANCE SYNDROME GH concentration Elevated interpulse levels of GH have been reported in critically ill subjects, although pulsatile GH secretion is reduced (Ross et al, 1991b; Voerman et al, 1992; van den Berghe et al, 1994). In chronic renal failure (CRF), fasting GH is high (Wright et al, 1968; Ross et al, 1991b) and in patients with colorectal carcinoma fasting GH levels of 2.9 ng/ml were found--sixfold higher than in controls (Tayek et al, 1990). In trauma, GH concentration was elevated for the first few days after injury (Frayn et al, 1984) and patients with HIV infection also have high levels of GH (Rondanelli et al, 1997). Not all reports have found GH levels to be high in ill patients. Critically
ACQUIRED GROWTH
H O R M O N E R E S I S T A N C E IN A D U L T S
317
ill patients with multiple trauma have been described as having low concentrations of GH, although neither study had controls (Jeevanandam et al, 1992; van den Berghe et al, 1994). Dopamine, frequently used in critically ill patients, has also been found to lower GH concentration (van den Berghe et al, 1994). IGF-1 concentration IGF-1 concentration has been reported to be low in critically ill patients (Phillips and Unterman, 1984; Ross et al, 1991b) with reduced bioactivity and to fall after elective surgery (Cotterill et al, 1996). Patients with bums have low levels of IGF-1 which gradually rise to normal during recovery (Abribat et al, 1993); the degree of fall is greater in patients with more severe burns. Similarly, IGF-1 bioactivity falls after trauma to about 50% of normal values before rising over a period of a week (Frayn et al, 1984). HIV patients without wasting also have low concentrations of IGF-1 (Rondanelli et al, 1997). Total IGF-1 represents free IGF-1 and IGF-1 bound to one of a number of insulin-like growth factor-binding proteins (IGFBP-1 to -6). The two major circulating members of this family are IGFBP-1 and IGFBP-3. The latter binds IGF-I and an acid-labile subunit to form a ternary complex which acts as a carrier and pool of circulating IGF-1 and may augment its actions by both increasing the circulating half-life and delivering IGF-I to the tissues. It appears likely, as in other hormone systems, that it is the free hormone which is metabolically active. Calculation of a free IGF-1 index ([total IGF-I]/[IGFBP-1] may allow an estimate of free IGF-1 levels and permit a re-assessment of the definition of acquired growth hormone resistance (AGHR)). Some support for this concept comes from the finding that exogenous GH improves this ratio and also leads to increased anabolism (T6nshoff et al, 1990). CONDITIONS ASSOCIATED WITH AGHR AGHR is likely to be a common metabolic state produced by different mechanisms in different diseases (cf. insulin resistance). A variety of conditions exhibit the AGHR syndrome to some degree. Table 1 lists these and summarizes which criteria they have been shown to satisfy. In some conditions, such as CRF, there is extensive and consistent characterization of the axis, and this condition will be discussed in detail as a possible paradigm for the other conditions which are less well researched. Table 2 describes the alterations found in many parameters of the GH/IGF-1 system in renal failure in comparison to normals. The range of conditions associated with AGHR are diverse, and in each there are many pathophysiological mechanisms at play, such as altered organ perfusion, abnormal acid-base balance, abnormal electrolytes, changes in other hormone systems (e.g. elevation of cortisol or the sick euthyroid syndrome), drug effects, cytokine activation and hypoxia, which may all contribute towards the syndrome.
318
R. C. JENKINS AND R. J. M. ROSS Table 1. Conditions which satisfy criteria for AGHR. Condition
GH
IGF- 1
GH effect
Renal failure Sepsis AIDS Surgery Cancer Burns Trauma Chronic liver disease Fasting
1" 3` ? 3" 3" 1" 1"$ 3` 3`
$ $ $ ,[, $ $ $ $ ,I.
$ $ ,[, NK $ NK NK NK $
1"= increased; ,l, = decreased; NK = not known.
Table 2. Changes in parameters of the GH/IGF-I system in renal failure. Parameter
Change in renal failure
Fasting GH concentration GH half-life GH receptor numbers GH binding protein concentration Total IGF- 1 concentration IGF-I half-life IGFBP- 1 concentration IGFB P-3 concentration Acid-labile sub-unit concentration Protease
3` 1" NK $ -,1, $ "[" 3` ~ +-~
Reference Ross et al (1991b) Garcia-Mayor et al (1993)
Maheshwari et al (1992) Jacob et al (1990) Fouque et al (1995a) T6nshoff et al (1996) Tr et al (1996) Haffner et al (1997) Rabkin et al (1996)
1"= increased; ,[, = decreased; ~ = unchanged; NK = not known.
POSSIBLE MECHANISMS UNDERLYING AGHR The processes leading to the gross changes in GH and IGF-1 concentrations are discussed below with an emphasis on the changes in specific parameters of the GH/IGF- 1 system. Decreased IGF-1 concentration IGF-1 concentration represents a balance between synthesis and breakdown, modified by binding-protein levels, and so changes in any of these parameters could result in low levels of IGF-i.
IGF-1 secretion Direct measurement of IGF-1 secretion is difficult, and so inferences must be drawn from the available evidence on IGF-I concentration and metabolism. Essentially, in the presence of a low concentration of IGF-1 and decreased IGF-I half-life, IGF-1 secretion must be inadequate to restore normal IGF-1 levels, but it is not clear whether secretion is increased, normal or decreased in comparison to normal.
ACQUIRED GROWTH HORMONE RESISTANCE 1N ADULTS
319
IGF-1 half-life IGF-1 half-life is decreased after surgery (Miell et al, 1992), in critical illness (Yarwood et al, 1997) and in CRF treated by renal replacement therapy it has been reported to be reduced (Fouque et al, 1995a) or normal (Rabkin et al, 1996). The mechanism for this is not established. IGF-1 in the circulation exists as a ternary complex with IGFBP-3 and the ALS; it is freed to act by the action of a specific protease. There is experimental evidence that IGF-1 half-life is approximately 12-15 hours when in the ternary complex and only 10 minutes when free in the circulation (Giudice, 1995). Clearly, an alteration in the activity of IGFBP-3 protease would have a major effect on IGF-I half-life. Evidence to support this mechanism for the reduction in IGF-I half-life comes from the finding that levels of the protease are increased in illness (Davies et al, 1991) or after surgery (Cwyfan-Hughes et al, 1992; Davenport et al, 1992), but not in renal failure (Rabkin et at, 1996) or cirrhosis (Ross et al, 1996). This may lead to liberation of IGF-1 to the tissues; however, if this is occurring, then it also seems necessary for there to be resistance to the actions of IGF-1 in order for protein catabolism to persist. Evidence that this is the case comes from the finding of a decreased metabolic response to IGF-I in renal failure; amino acid and insulin concentrations fell less in patients than in controls after exogenous IGF-I (Fouque et al, 1995b).
IGFBP Low levels of IGF-I may just represent low levels of IGFBP and mask normal free levels of IGF-1. This does not appear to be the case in renal failure where levels of IGFBP-1 and IGFBP-3 are high (T6nshoff et al, 1996); however, some of this elevation may relate to accumulation of immunoreactive fragments (Rabkin et al, 1996). The normal regulation of IGFBP-l by insulin is preserved in critical illness (Ross et al, 1991a). Increased GH concentration
Increased concentration of GH could be due to either a prolonged GH halflife, increased secretion of GH or increased concentration of GHBP.
GH secretion GH secretion over 24 hours in patients with chronic liver disease is approximately doubled (Cuneo et al, 1995). The reported temporal patterns of GH secretion in critical illness differ according to the group studied. In liver cirrhosis more frequent secretory pulses of GH occur, with individual pulses being similar to normals in mass (Cuneo et al, 1995). In intensive care unit (ICU) patients GH secretory pulses are attenuated in amplitude but GH trough levels between pulses are high (Ross et al, 1991b; van den Berghe et al, 1994). In contrast, critically ill trauma victims were found to have less frequent secretory bursts of GH,
320
R. C. J E N K I N S A N D R. J. M. R O S S
although the sampling frequency of 60 minutes in the latter study was probably insufficient (Melarvie et al, 1995). Clearly, if the primary defect in AGHR is a low IGF-1 concentration then an increase in GH secretion could be caused by a loss of the usual negative feedback of IGF-1. This is supported by the findings in chronic liver disease of a relationship between the degree of rise in GH and the deficit in IGF-1 levels (Cuneo et al, 1995). This mechanism is analogous to the very high levels of GH reported in a child with an inherited defect in the IGF-1 gene who effectively had no circulating IGF-1 (Woods et al, 1996).
GH half-life Increased secretion of GH is unlikely to be the whole story, however, as GH half-life (of exogenously administered GH) has been reported to be increased by 25-50% (Haffner et al, 1994) or more (Garcia-Mayor et al, 1993; Schaefer et al, 1996) in CRE A similar finding, for endogenous GH, has been reported in chronic liver disease (Cuneo et al, 1995) and critical illness (Ross et al, 1991b). Whether the prolonged half-life is due to a reduction in GH binding to the liver or an increased GHBP is not clear, although, as discussed below, the latter seems unlikely.
GHBP Decreased concentrations of GHBP have been reported in patients with CRF (Maheshwari et al, 1992; Schaefer et al, 1996), chronic liver disease (Baruch et al, 1991) and critical illness (Ross et al, 1991a) and so the elevation in GH concentrations is likely to underestimate the increase in unbound GH in these conditions. The degree of reduction in GHBP in liver cirrhosis was greater in patients with more severe cirrhosis. GH receptor function The biochemical profile of acquired GH resistance is qualitatively similar to that of congenital GH resistance or Laron syndrome (Woods and Savage, 1996) in which, in the classical form, there is an inherited defect of the GH receptor (GH-R). Thus, attention in the acquired syndrome has also been directed towards the function of the GH-R. However, it must be noted that this hormonal profile could also be produced by a primary reduction in the concentration of IGF-1, by increased clearance for example, and a secondary increase in GH concentrations without there being a defect at the level of the GH-R. The translation of the GH signal to IGF-1 production may be interfered with by other circulating factors. Elevated prolactin concentrations, for example, appear to lead to lower than expected IGF-1 levels for a given GH concentration in acromegalic patients (Hulting et al, 1984) although the mechanism responsible is obscure. In uraemia there is no evidence of a circulating inhibitory serum factor (Maheshwari et al, 1992).
ACQUIRED
GROWTH
HORMONE
RESISTANCE
321
IN ADULTS
$ <---
W
g 8
E-
-+
$
6--
5 6-
W -+
=_ e~ - 9 --> - 9 ---->---->---->-~. --). --~ - 9 -+ r~
k3
II ~,,
-9
5_2
" " ~ " Zz - + - - , g $ =~
+--~ Z Z
It
=_ -< ="
=
3
;q 4. o
2-W S
W
322
R. C. J E N K I N S A N D R. J. M. ROSS
GH-R numbers Low levels of a normally functioning receptor could lead to AGHR but it is difficult to measure receptor numbers directly as appropriate liver tissue is unavailable from humans with AGHR. However, expression of the GH-R gene in skeletal muscle has been shown to be decreased after elective surgery (Hermansson et al, 1997). As the GHBP is derived from the same gene as the GH-R, either by alternative splicing or proteolytic cleavage, it is possible that concentrations of the GHBP could act as a surrogate marker for GH-R numbers. Animal studies may not be relevant to the human situation as in rats, at least, the GHBP is probably produced by a different mechanism than in humans. The rationale for this approach is supported by the finding of low levels of GHBP in most cases of congenital GH resistance. GHBP concentrations, as discussed above, have been found to be low in some groups with AGHR.
GH binding Direct assessment of the function of the GH-R is difficult. Clearly, the combination of raised GH and lowered IGF-1 concentrations suggests that the GH signal is not leading to normal IGF-1 levels, but this could merely reflect a defect at a post-receptor site. Binding of GH to the GHBP provides an analogy for GH binding to its receptor, and this has been reported to be normal in CRF (Maheshwari et al, 1992; Schaefer et al, 1996). Similarly, patients with varying degrees of liver failure also had normal binding of GH to the GHBP (Baruch et al, 1991).
Effect of pH Most receptors and their ligands bind optimally over a narrow pH range. It has been suggested that an altered pH in CRF may be responsible for AGHR (Carlin and Carlin, 1994) although we know of no experimental evidence to support this hypothesis in humans. In a number of the conditions associated with AGHR-A, pH changes as a result of renal failure, respiratory failure or poor tissue perfusion. However, experimental studies have determined that GH binds to its receptor over quite a large range of pH (6.6-8.2) (Murphy et al, 1983; Hung et al, 1985), and in renal failure GH-R binding affinity has been reported to be normal. Whether pH alters events after receptor binding is not known.
GH-R signalling If GH-R numbers and function are normal then reduced IGF-1 production must represent a post-receptor defect but no data are available to answer this question.
ACQUIRED GROWTH HORMONE RESISTANCE IN ADULTS
323
Nutritional status The nutritional state is known to modulate the function of the GHflGF-I axis, and fasting in normals leads to a state reminiscent of AGHR. For example, fasting for 24 hours led to a fivefold increase in GH production due to a doubling in the frequency of GH secretory bursts and an increase in the mass of each burst (Hartman et al, 1992). IGF- 1 levels and GH halflife in this study did not alter after 56 hours of fasting but the former has been reported to fall following calorie malnutrition (Isley et al, 1983; Musey et al, 1993). Institution of nutritional support is associated with a rise in IGF-I concentration in proportion to nitrogen balance, which is then reversed by withdrawal of support (Donahue and Phillips, 1989; Minuto et al, 1989). In patients with CRF the serum IGF-I concentration has been found to be a sensitive marker of nutritional state (Jacob et al, 1990) and to correlate with anthropometric measures (Sanaka et al, 1994). A prospective study of critically ill ICU patients found that IGF-1 was more strongly correlated with nitrogen balance than were conventional indicators of nutritional status, such as pre-albumin or albumin (Hawker et al, 1987). GHBP is also modulated by nutritional status, with low levels in malnutrition and high levels in obesity (Postel-Vinay et al, 1995). IGFBP-1 is negatively regulated by insulin, and levels rise during fasting (Musey et al, 1993). The abnormalities in fasting are not confined to the basal hormone concentrations--administration of GH (10mg daily for 5 days) to fed subjects led to a fourfold higher IGF-I concentration than when the treatment was given in the fasting state (Merimee et al, 1982), i.e. resistance to the actions of GH. Fasting can therefore lead to increased levels of GH, decreased IGF-I, decreased GHBP and high IGFBP-1 as in AGHR, and although it seems certain that inadequate nutrition contributes to the syndrome it is unlikely to be the whole story. The relative contributions of fasting and illness to the syndrome are illustrated by the finding that GH concentrations in critically ill subjects were similar to those of fasting normals, but IGF-I concentrations were much lower in patients than in controls (Ross et al, 1991b).
The effect of cytokines Cytokines are released in many of the conditions associated with AGHR, such as sepsis, trauma and burns. Complex and, at times, contradictory evidence implicates interleukin-l, interleukin-6 and tumour necrosis factor-c~ as modulating levels of somatostatin and GHRH (MandrupPoulsen et al, 1995). Recombinant interleukin-6 stimulates acute secretion of GH in humans (Tsigos et al, 1997), although the main effects of cytokines in AGHR are likely to be via effects on IGF-1 production. Interleukin-1-]3 and tumour necrosis factor-c~ inhibit GH receptor mRNA, IGF-1 mRNA and IGF-I protein production in tissue culture (Wolf et al, 1996). Interleukin-6 reduces IGF-1 levels in mice and possibly in humans also (De Benedetti et al, 1997), although the mechanism is not clear.
324
R. C. J E N K I N S A N D R. J. M. ROSS
EFFECTS OF AGHR
Is there any practical use of the concept of AGHR or is it merely a biochemical epiphenomenon? Is there evidence that the AGHR syndrome plays an active and deleterious role or does it just reflect the effect of other disease processes? The evolution of such a response to illness suggests that it has survival advantage but whether this holds true in patients with prolonged illness, such as those in the ICU, is uncertain. It is possible that, in the short term, it serves to conserve nutritional substrates, for example by allowing amino acids from the breakdown of muscle protein to be used for other purposes, and thus could be beneficial, but as illness becomes chronic and malnutrition supervenes it could become harmful by preventing anabolism. If AGHR is a beneficial metabolic reflex designed to protect against substrate use in stress then pharmacological strategies to 'correct' it may be harmful. On the other hand, if it is a deleterious state, perhaps by being permissive to catabolism, then treatment should be directed to reversing it. One way to resolve this uncertainty is to examine the evidence regarding the effects of exogenous GH and/or IGF-1 in AGHR (see below). The combination of high GH and low IGF-1 concentrations would be expected to lead to a predominance of the direct anti-insulin actions of GH over the anabolic effects mediated by IGF-1. Patients with advanced colorectal carcinoma have been found to have high hepatic glucose output in keeping with antagonism of insulin actions by GH (Tayek et al, 1990). Treatment to 'correct' AGHR
Assuming that AGHR is harmful, then how might therapy be directed to ameliorate or correct it? Other than by treating the underlying condition and optimizing nutritional support, correction of the hormonal imbalance can be approached in three ways--administration of exogenous GH, exogenous IGF-I, or both together, and all these avenues have been explored. These treatment regimens have been studied in a number of ways, predominantly using surrogate metabolic end-points to imply that there would be beneficial clinical effects; as yet there are few published studies with hard clinical end-points. As each of the conditions associated with AGHR exhibit different pathophysiological mechanisms, it is unlikely that evidence of favourable or adverse effects of treatment in one group could be safely extrapolated to other groups. Treatment of catabolism with GH and IGF-I has been recently reviewed (Jenkins and Ross, 1996), and this chapter provides only an overview of relevant results. The use of GH
In principle, it should be possible to overcome GH resistance by giving high doses of exogenous GH; this hypothesis is supported by the study of fasting obese subjects who exhibited an increase in IGF-1 concentration after GH treatment (Snyder et al, 1990). Metabolic improvements have
A C Q U I R E D G R O W T H H O R M O N E R E S I S T A N C E 1N A D U L T S
325
been seen in patients with sepsis or trauma given GH (20 units daily for 3 days) who had significantly reduced net protein loss compared to a control period (Douglas et al, 1990). Patients with gastrointestinal or pancreatic disease on parenteral nutrition had significantly greater retention of nitrogen, potassium and phosphate when treated with GH (10 mg daily for 14 days) (Ziegler et al, 1992). Immobilized, neurogically injured patients did not improve their nitrogen balance with GH treatment (0.2 mg/kg daily for 7-13 days) but did have improvements in albumin and transferrin concentrations (Behrman et al, 1995). Children with CRF have GH resistance and respond to exogenous GH by increasing growth (T6nshoff et al, 1990), but in adults it is more difficult to demonstrate benefit. Definite clinical benefits of GH treatment in adults have been reported more rarely. A large study of patients after elective cholecystectomy found reduced wound infections, possibly related to improved cell-mediated immunity, and shorter hospital stay (Vara-Thorbeck et al, 1993). Patients with major burns have been reported to have decreased mortality (11 versus 37%) following treatment with GH (mean dose 0.11 mg/kg) (Knox et al, 1995). Negative clinical studies have found no anabolic effect of GH (0.125mg/kg) in patients with cancer (Tayek and Brasel, 1995) or in patients with sepsis or trauma receiving huge doses of GH (1.5 unit/kg daily) (Gottardis et al, 1991). More recently, two large multicentre trials of the use of GH in intensive care patients have been terminated early due to significantly increased mortality in the patients treated with GH (41.7 versus 18.2%) (Pharmacia & Upjohn communication). The analysis of this study is yet to be published, but the trial included 532 patients with either open-heart surgery, abdominal surgery, multiple trauma or acute respiratory failure and used GH doses of either 16 or 24 IU/day. In contrast, Serono have found no increase in mortality in an interim analysis of 450 patients, predominantly with HIV-associated wasting, who received a similar dose of GH. Similarly, Lilly have also found no increase in mortality in a study of 166 patients with hip fracture, liver transplant or severe bums. Firm conclusions will be possible only when these trials have been fully analysed and published, but it is possible that GH could be beneficial in some groups of patients and harmful in others. The use of lGF-1
IGF- 1 has been given to patients with AIDS, but the results were disappointing (Lieberman et al, 1994). An impaired metabolic response to IGF-1 was also seen in patients with CRF (Fouque et al, 1995b). The long-term effects may have been limited by feedback suppression of endogenous GH and thus reduced IGFBP-3 and ALS leading to tachyphylaxis. In addition, IGF-1 has actions similar to those of insulin and so can lead to hypoglycaemia. The use of combined GH and 1GF-1
The combined use of these agents may be the most potent approach. The potential advantages of this approach are that it will avoid the
326
R. C. JENKINS AND R. J. M. ROSS
hypoglycaemia sometimes seen when IGF-I is given alone and that exogenous GH may correct for the inhibition of endogenous GH caused by the administration of IGF-1 and allow the production of IGFBP-3 and ALS to be maintained or increased--thus preserving the anabolic actions of IGF-1. Use of this combination in fasting subjects produced a potent nitrogen-retaining effect, no hypoglycaemia and increased concentration of IGFBP-3 and ALS (Kupfer et al, 1993). However, combined treatment of HIV-infected patients did not produce significant sustained anabolic effects (Ellis et al, 1996; Lee et al, 1996).
SUMMARY There is now abundant evidence that in a range of acute and chronic illnesses associated with catabolism there is a consistent perturbation in the GH/IGF-1 system characterized by raised levels of GH and low levels of IGF-1. It remains unclear whether this biochemical finding is merely a nonspecific consequence of illness or whether it contributes adversely to the nutrition and survival of the patient. A similar profile occurs in fasting, and this may underlie some, but not all, of the syndrome. Different conditions manifest AGHR to different extents. In many, there is induction of a protease which reduces IGF-1 half-life, and GH concentrations probably rise due to the removal of IGF- 1 negative feedback. The varying therapeutic strategies which have used exogenous recombinant human GH and/or IGF-1 to correct the syndrome have met with limited clinical success, such as in patients with AIDS, but more usually have been successful only at altering metabolic parameters. It is possible that AGHR represents a protective metabolic reflex designed to deal with protein-calorie malnutrition, and so treatment to overcome it could be detrimental. This question may be answered by ongoing clinical studies.
Acknowledgements This work was supported by the University of Sheffield, the Northern General Hospital NHS Trust and Pharmacia and Upjohn Limited.
REFERENCES Abribat T, Brazeau P, Davignon I & Garrel DR (1993) Insulin-like growth factor-I blood levels in severely burned patients: effects of time post injury, age of patient and severity of burn. Clinical Endocrinology 39: 583-589. Arnold J, Campbell IT, Samuels TA et al (1993) Increased whole body protein breakdown predominates over increased whole body protein synthesis in multiple organ failure. Clinical Science 84: 655-661. Baruch Y, Amit T, Hertz Pet al (1991) Decreased serum growth hormone-binding protein in patients with liver cirrhosis. Journal of Clinical Endocrinology and Metabolism 73: 777-780. Behrman SW, Kudsk KA, Brown RO et al (1995) The effect of growth hormone on nutritional markers in enterally fed immobilised trauma patients. Journal ~)f Parenteral and Enteral Nutrition 19: 41-46.
ACQUIRED GROWTH HORMONE RESISTANCE IN ADULTS
327
De Benedetti F, Alonzi T, Moretta Aet al (1997) lnterleukin-6 causes growth impairment in transgenic mice through a decrease in insulin-like growth factor-I. A model for stunted growth in children with chronic inflammation. Journal of Clinical Investigation 99: 643-650. *van den Berghe, G, De Zegher E Lauwers P & Veldhuis JD (1994) Growth hormone secretion in critical illness: effect of dopamine. Journal of Clinical Endocrinology and Metabolism 79: 1141 - 1146.
Berneis K & Keller U (1996) Metabolic actions of growth hormone: direct and indirect. BailliOre's Clinical Endocrinology and Metabolism 10: 337-352. Carlin K & Carlin S (1994) A possible explanation for growth hormone resistance in chronic renal failure. Medical Hypotheses 43: 193-194. Cotterill AM, Mendel P, Holly JMP et al (1996) The differential regulation of the circulating levels of the insulin-like growth factor binding proteins (IGFBP) I, 2 and 3 after elective abdominal surgery. Clinical Endocrinology 44: 91-101. Cuneo RC, Hickman PE, Wallace JD et al (1995) Altered endogenous growth hormone secretory kinetics and diurnal GH-binding protein profiles in adults with chronic liver disease. Clinical Endocrinology 32: 265-275. Cuthbertson DP (1931) Observations on the disturbance of metabolism produced by injury to the limbs. Quarterly Journal qf Medicine 2: 233-246. Cwyfan-Hughes SC, Cotterill AM, Molloy AR et al (1992) The induction of specific proteases for insulin-like growth factor-binding proteins following major heart surgery. Journal of Endocrinology 135:135-145. Davenport ML, lsley WL, Pucilowska JB et al (1992) Insulin-like growth factor-binding protein-3 proteolysis is induced after elective surgery. Journal of Clinical Endocrinology and Metabolism 75:590 595. *Davies SC, Wass JA, Ross RJ et al (1991) The induction of a specific protease for insulin-like growth factor binding protein-3 in the circulation during severe illness. Journal of Endocrinology 130: 469-473. Donahue SP & Philips LS (1989) Response of IGF-I to nutritional support in malnourished hospital patients: a possible indicator of short-term changes in nutritional status. American Journal c{f Clinical Nutrition 50: 962-969. *Douglas RG, Humberstone DA, Haystead A & Shaw JHF (1990) Metabolic effects of recombinant human growth hormone: isotopic studies in the postabortive state and during total parenteral nutrition. British Journal ~f Surgery 77: 785-790. Ellis KJ, Lee PDK, Pivarnik JM et al (1996) Changes in body composition of human immunodeficiency virus-infected males receiving insulin-like growth factor I and growth hormone. Journal r Clinical Endocrinology and Metabolism 81: 3033-3038. Fouque D, Peng SC & Kopple JD (1995a) Pharmacokinetics of recombinant human insulin-like growth factor-1 in dialysis patients. Kidney International 47: 869-875. Fouque D, Peng SC & Kopple JD (1995b) h'npaired metabolic response to recombinant insulin-like growth factor-1 in dialysis patients. Kidney h~ternationa147: 876-883. Frayn KN, Price DA, Maycock PF & Carroll SM (1984) Plasma somatomedin activity after injury in man and its relationship to other hormonal and metabolic changes. Clinical Endocrinology 20: 179-187. Garcia-Mayor RVG, Perez A J, Gandara Aet al (1993) Metabolic clearance rate of biosynthetic growth hormone after endogenous growth hormone supp~'ession with a somatostatin analogue in chronic renal failure patients and control subjects. Clinical Endocrinology 39: 337-3~.3. Giudice LC (1995) IGF binding protein-3 protease regulation: how sweet it is! Journal qf Clinical Endocrinology and Metabolism 80: 2279-2281. Gottardis M, Benzer A, Koller W e t al (1991) Improvement of septic syndrome after administration of recombinant human growth hormone? Journal c)fTraumu 31:81-86. Haffner D, Schaefer F, Girard J e t al (1994) Metabolic clearance of recombinant human growth hormone in health and chronic renal failure. Journal Of Clinical Investigation 93: 11631171.
Haffncr D, Trnshoff B, Blum WF et al (1997) Insulin-like growth factors (IGFs) and IGF binding proteins, serum acid-liable subunit and growth hormone binding protein in nephrotic children. Kidney hlternational 52:802-8 I0. *Hartman ML, Veldhuis JD, Johnson ML et al (t992) Augmented growth hormone (GH) secretory burst frequency and amplitude mediate enhanced GH secretion during a two-day fast in normal men. Journal qf Clinical Endocrinology and Metabolism 74: 757-765.
328
R. C. JENKINS AND R. J. M. ROSS
Hawker FH, Stewart PM, Baxter RC et al (1987) Relationship of somatomedin-C/insulin-like growth factor I levels to conventional nutritional indices in critically ill patients. Critical Care Medicine 15: 732-736. *Hermansson M, Wickelgren RB, Hammarqvist F et al (1997) Measurement of human growth hormone receptor messenger ribonucleic acid by a quantitative polymerase chain ~reaction-based assay: demonstration of reduced expression after elective surgery. Journal of Clinical Endocrinology and Metabolism 82:421-428. Hulting AL, Hall K & Werner S (1984) Somatomedin A increments are diminished in acromegaly with concomitant hyperprolactinaemia. Acta Endocrinologica 106: 305-310. Hung CH, Kover K & Moore WV (1985) Characteristics of somatotropic growth hormone binding to homologous liver plasma membranes. Molecular and Cellular Endocrinology 39:189-196. IsIey WL, Underwood LE & Clemmons DR (t983) Dietary components that regulate serum somatomedin-C concentrations in humans. Journal of Clinical Investigation 71:175-182. Jacob V, Le Carpentier JE, Salzano S et al (1990) IGF-I, a marker of undernutrition in hemodialysis patients. American Journal of Clinical Nutrition 52: 39-44. Jeevanandam M, Ramias L, Shamos RF & Schiller WR (1992) Decreased growth hormone levels in the catabolic phase of severe injury. Surgery 111: 495-502. Jenkins RC & Ross RJM (1996) Growth hormone therapy for protein catabolism. Quarterly Journal of Medicine 89: 813-819. Knox J, Demling R, Wilmore D et al (1995) Increased survival after major thermal injury: the effect of growth hormone therapy in adults. Journal of Trauma 39: 526-530. *Kupfer SR, Underwood LE, Baxter RC & Clemmons DR (1993) Enhancement of the anabolic effects of growth hormone and insulin-like growth factor-I by use of both agents simultaneously. Journal of Clinical Investigation 91:391-396. Lee PDK, Pivarnik JM, Bukar JG et al (1996) A randomized, placebo-controlled trial of combined insulin-like growth factor 1 and low dose growth hormone therapy for wasting associated with human immunodeficiency virus infection. Journal of Clinical Endocrinology and Metabolism 81: 2968-2975. Lieberman SA, Butterfield GE, Harrison D & Hoffman AR (1994) Anabolic effects of recombinant insulin-like growth factor-I in cachetic patients with the acquired immunodeficiency syndrome. Journal of Clinical Endocrinology and Metabolism 78: 404-410. Maheshwari HG, Rifkin I, Butler J & Norman M (1992) Growth hormone binding protein in patients with renal failure. Acta Endocrinologica 127: 485-488. Mandrup-Poulsen T, Nerup J, Reimers Jl et al (1995) Cytokines and the endocrine system. I. The immunoendocrine network. European Journal of Endocrinology 133: 665-666. Melarvie S, Jeevanadam M, Holaday NJ & Petersen SR (1995) Pulsatile nature of growth hormone levels in critically ill trauma victims. Surge O' 117: 402-408. Merimee TJ, Zapf J & Froesch ER (1982) Insulin-like growth factors in the fed and fasted states. Journal of Clinical Endocrinology and Metabolism 55: 999-1002. *Miell JP, Taylor AM, Jones J e t al (1992) Administration of human recombinant insulin-like growth factor-I to patients following major gastrointestinal surgery. Clinical Endocrinology 37:542-551. Minuto E Barreca A, Adami GF et at (1989) Insulin-like growth factor-I in human malnutrition: relationship with some body composition and nutritional parameters. Journal of Parenteral and Enteral Nutrition 13: 392-39@ Murphy LJ, Vrhovsek E & Lazarus L (1983) Characterisation of specific growth hormone binding sites in mouse fibmblasts. Endocrinology 113: 750-757. Musey VC, Goldstein S, Farmer PK et al (1993) Differential regulation of IGF-I and [GF-binding protein-1 by dietary composition in humans. American Journal of Medical Science 305: 131-138. Phillips LS & Unterman TG (1984) Somatomedin activity in disorders of nutrition and metabolism. Clinical Endocrinology and Metabolism 13:145-189. Postel-Vinay M-C, Saab C & Gourmelen M (1995) Nutritional status and growth hormone-binding protein. Hormone Research 44: 177-181. Rabkin R, Fervenza FC, Maidment H et al (1996) Pharmacokinetics of insulin-like growth factor-I in advanced chronic renal failure~ Kidney h~ternationa149: l 134-1140. Rondanelli M, Solerte SB, Fioravanti Met al (1997) Circadian secretory pattern of growth hormone, insulin-like growth factor type I, cortisol, adrenocorticotropic hormone, thyroid-stimulating hormone, and prolactin during HIV infection. Aids Research and Human Retroviruses 13: 1243 - 1249.
ACQUIRED GROWTH HORMONE RESISTANCE IN ADULTS
329
Ross RJM, Goodwin FJ, Houghton BJ & Boucher BJ (1985) Alteration of pituitary thyroid function in patients with chronic renal failure treated by haemodialysis or continuous ambulatory peritoneal dialysis. Annals of Clinical Biochemistry 22: 156-160. Ross RJ, Miell JE Holly JM et al (1991a) Levels of GH binding activity, IGFBP-1, insulin, blood glucose and cortisol in intensive care patients. Clinical Endocrinology 35: 361-367. *Ross RJM, Miell J, Freeman E et al (1991b) Critically ill patients have high basal growth hormone levels with attenuated oscillatory activity associated with low levels of insulin like growth factorI. Clinical Endocrinology 35: 47-54. Ross RJ, Chew SL, D'Souza L et al (1996) Expression of IGF I and IGF-binding protein genes in cirrhotic liver. Journal of Endocrinology 149: 209-216. Sanaka T, Shinobe M, Ando Met a[ (1994) IGF-I as an early indicator of malnutrition in patients with end-stage renal disease. Nephron 67: 73-81. Schaefer F, Baumann G, Haffner D et al (1996) Multifactorial control of the elimination kinetics of unbound (free) growth hormone in the human: regulation by age, adiposity, renal function, and steady state concentrations of GH in plasma. Journal rff Clinical Endocrinology and Metabolism 81:22 31. Snyder DK, Underwood LE & Clemmons, DR (1990)Anabolic effects of growth hormone in obese diet-restricted subjects are dose dependent. American Journal Of Clinical Nutrition 52:431-437. *Tayek JA & Brasel JA (1995) Failure of anabolism in malnourished cancer patients receiving growth hormone: a clinical research center study. Journal o/Clinical Endocrinology and Metabolism 80: 2082-2087. Tayek JA, Bulcavage L & Chlebowski RT (1990) Relationship of hepatic glucose production to growth hormone and severity of malnutrition in a population with colorectal carcinoma. Cancer Research 50:2119 2122. TOnshoff B, Blum WF & Mehls O (1996) Serum of insulin like growth factors and their binding proteins in children with end-stage renal disease. Pediatric Nephrology 10: 269-274. TOnshoff B, Mehls O, Heinrich U et al (1990) Growth-stimulating effects of recombinant human growth hormone in children with end-stage renal disease. Journal ~['Pediatrics 116:561-566. Tsigos C, Papanicolaou DA, Defcnsor R et al (1997) Dose effects of recombinant human interleukin6 on pituitary hormone secretion and energy expenditure. Neuroendocrinology 66: 54-62. Vara-Thorbeck R, Guerrcro JA, Rosell J e t al (1993) Exogenous growth hormone: effects on the catabolic response to surgically produced acute stress and on post-operative immune function. World .lournal qf Surgel3' 1 7 : 5 3 0 - 5 3 8 Voerman H J, Strack van Schijndcl RJM, Groenevcld ABJ et al (1992) Pulsatile hormone secretion during scvere sepsis: accuracy of different blood sampling regimens. Metabolism 9: 934-940. Wihnore, DW (1991) Catabolic illness. Stratcgies for enhancing recovery. New England Journal (ff Medicine 325:695 702. Wolf M, Bohm S, Brand M & Kreymann G (1996) Proinflammatory cytokines interleukin 1 beta and turnout necrosis factor alpha inhibit growth hormone stimulation of insulin-like growth factor I synthesis and growth hormone receptor mRNA levels in cultured rat liver cells. European Journal c)f Endocrinology 135: 729-737. Woods KA & Savage MO (1996) Laron syndrome: typical and atypical forms. Baillikre~ Clinical Fndocrim~logy and Metabolism 10:371 387. Woods KA, Camacho-Hubner C, Savage MO & Clark AJL (1996) Intrauterine growth retardation and postnatal growth failure associated with deletion of the insulin-like growth factor-I gene. New E~tgland Journal (~fMedicine 335:1363 1367. Wright AD, Lowy C, Fraser TR et al (1968) Serum growth hormone and glucose intolerance in renal failure. Lancet ii: 798-801. Yarwood G[), Ross RJM, Medbak S e t al (1997) Administration of human recombinant insulin tike growth l~tctor-I to critically ill pafieuts. Critical Ca~e Medwine 25: 1352-1361. *Ziegler TR, Rombeau JL, Young LS el a] (1992) Recombinant human growth hormone enhances the metabolic efficacy of parenteral nutrition: a double-blind, randomized controlled sludy Journal ~! Clinical Endocrinology .nd Metabolism 74: 865- 873.
BMedSci, MB ChB, MRCP
Clinical Research Fellow
R. J. M. ROSS* MD, FRCP Senior Lecturer and Consultant in Endocrinology
Department of Medicine, UniversiO, o[" Sheffield, Clinical Sciences Centre, Northern General Hospital, Herries Road, Sheffield, $5 7AU, UK
Acquired growth hormone resistance (AGHR) may be defined as the combination of a raised serum growth hormone (GH) concentration, low serum insulin-like growth factor-1 (IGF-1) concentration and a reduced anabolic response to exogenous GH. A wide range of conditions exhibit the syndrome to a variable degree, including sepsis, trauma, bums, AIDS, cancer, and renal or liver failure. The primary defect seems to be a reduction in IGF- 1 concentration which then leads to increased GH concentration by a loss of negative feedback. It is not clear whether IGF-1 concentration falls because of decreased production or increased clearance from the circulation, or both. Treatment to reverse the biochemical defect by restoring IGF-1 levels, either by the administration of GH or IGF-1, has resulted in improvements in a wide range of metabolic parameters and, more recently, to definite clinical benefit in well-defined groups, such as patients with AIDS. These results cannot be extrapolated to other groups with AGHR as a recent unpublished report suggested increased mortality in critically ill patients treated with GH. Research needs to focus on the molecular basis of AGHR if we are to develop therapies for catabolism. Key words: growth hormone; insulin-like growth factor-I; catabolism,
In 1931 Cuthbertson described increased urinary nitrogen excretion following trauma and similar, more refined, observations continue to be made (Arnold et al, 1993). Analogous phenomena have been observed in patients suffering from diverse conditions such as sepsis, burns, renal failure, liver disease, AIDS, cancer or after surgery. It has become apparent that the ensuing protein-calorie malnutrition both adversely affects prognosis and is difficult to correct by improving nutritional intake alone. Improvements in the management of patients with these and other diseases * Address correspondence to Dr R.J.M. Ross.
Baillibre's Clinical Endocrinology and Metabolism-315 Vol. 12, No. 2, July 1998 Copyright 9 1998, by Baitli~re Tindall ISBN 0-7020-2465-1 All rights of reproduction in any form reserved 0950-351 X/98/020315 + 15 $12.00/00
316
R. C. J E N K I N S A N D R. J. M. ROSS
have led to prolongation of survival and an increasing need to develop ways to correct the associated chronic malnutrition and thus to optimize recovery (Wilmore, 1991). The concept of acquired growth hormone resistance results from a recognition of a similar pattern of disturbance of the growth hormone/ insulin-like growth factor- 1 (GH/IGF- 1) system in a range of disparate but related conditions sharing the common feature of catabolism. The combination of high levels of growth hormone (GH), low levels of IGF-1 and a reduced anabolic effect of exogenous GH provides a simple biochemical definition of the condition. It is becoming increasingly clear that this is an over-simplification and takes no account of the effects of the growth hormone-binding protein (GHBP), insulin-like growth factor bindingproteins (IGFBP), free IGF-1 or the half-lives of GH and IGF-1. In health, GH is secreted in a pulsatile fashion by the somatotroph ceils of the anterior pituitary gland with secretion being stimulated by growth hormone-releasing hormone (GHRH) and the as yet unknown endogenous GH-releasing peptide (GHRP), and inhibited by somatostatin and negative feedback by IGF-1. GH is carried in the circulation by a specific binding protein, GHBP, levels of which are thought to reflect levels of GH receptor expression as both proteins are derived from the same gene. Circulating GH has direct and indirect actions (Berneis and Keller, 1996). It acts directly at the liver to stimulate IGF-1 secretion, which mediates the indirect anabolic actions of GH, and also production of IGFBP-3 and the acid-labile subunit (ALS). IGF-1 is not only controlled by GH, other factors such as fasting and nutritional status are also important determinants. Clearly, a simple measure of the levels of GH and IGF-1 gives only a crude assessment of the state of this complex hormone system. This chapter describes the evidence for the acquired GH resistance syndrome, and outlines the disturbances seen in the GH/IGF-1 system in the associated conditions. The possible mechanisms underlying the condition, the clinical significance and the effects of specific treatments will be discussed.
EVIDENCE FOR THE EXISTENCE OF THE ACQUIRED GH RESISTANCE SYNDROME GH concentration Elevated interpulse levels of GH have been reported in critically ill subjects, although pulsatile GH secretion is reduced (Ross et al, 1991b; Voerman et al, 1992; van den Berghe et al, 1994). In chronic renal failure (CRF), fasting GH is high (Wright et al, 1968; Ross et al, 1991b) and in patients with colorectal carcinoma fasting GH levels of 2.9 ng/ml were found--sixfold higher than in controls (Tayek et al, 1990). In trauma, GH concentration was elevated for the first few days after injury (Frayn et al, 1984) and patients with HIV infection also have high levels of GH (Rondanelli et al, 1997). Not all reports have found GH levels to be high in ill patients. Critically
ACQUIRED GROWTH
H O R M O N E R E S I S T A N C E IN A D U L T S
317
ill patients with multiple trauma have been described as having low concentrations of GH, although neither study had controls (Jeevanandam et al, 1992; van den Berghe et al, 1994). Dopamine, frequently used in critically ill patients, has also been found to lower GH concentration (van den Berghe et al, 1994). IGF-1 concentration IGF-1 concentration has been reported to be low in critically ill patients (Phillips and Unterman, 1984; Ross et al, 1991b) with reduced bioactivity and to fall after elective surgery (Cotterill et al, 1996). Patients with bums have low levels of IGF-1 which gradually rise to normal during recovery (Abribat et al, 1993); the degree of fall is greater in patients with more severe burns. Similarly, IGF-1 bioactivity falls after trauma to about 50% of normal values before rising over a period of a week (Frayn et al, 1984). HIV patients without wasting also have low concentrations of IGF-1 (Rondanelli et al, 1997). Total IGF-1 represents free IGF-1 and IGF-1 bound to one of a number of insulin-like growth factor-binding proteins (IGFBP-1 to -6). The two major circulating members of this family are IGFBP-1 and IGFBP-3. The latter binds IGF-I and an acid-labile subunit to form a ternary complex which acts as a carrier and pool of circulating IGF-1 and may augment its actions by both increasing the circulating half-life and delivering IGF-I to the tissues. It appears likely, as in other hormone systems, that it is the free hormone which is metabolically active. Calculation of a free IGF-1 index ([total IGF-I]/[IGFBP-1] may allow an estimate of free IGF-1 levels and permit a re-assessment of the definition of acquired growth hormone resistance (AGHR)). Some support for this concept comes from the finding that exogenous GH improves this ratio and also leads to increased anabolism (T6nshoff et al, 1990). CONDITIONS ASSOCIATED WITH AGHR AGHR is likely to be a common metabolic state produced by different mechanisms in different diseases (cf. insulin resistance). A variety of conditions exhibit the AGHR syndrome to some degree. Table 1 lists these and summarizes which criteria they have been shown to satisfy. In some conditions, such as CRF, there is extensive and consistent characterization of the axis, and this condition will be discussed in detail as a possible paradigm for the other conditions which are less well researched. Table 2 describes the alterations found in many parameters of the GH/IGF-1 system in renal failure in comparison to normals. The range of conditions associated with AGHR are diverse, and in each there are many pathophysiological mechanisms at play, such as altered organ perfusion, abnormal acid-base balance, abnormal electrolytes, changes in other hormone systems (e.g. elevation of cortisol or the sick euthyroid syndrome), drug effects, cytokine activation and hypoxia, which may all contribute towards the syndrome.
318
R. C. JENKINS AND R. J. M. ROSS Table 1. Conditions which satisfy criteria for AGHR. Condition
GH
IGF- 1
GH effect
Renal failure Sepsis AIDS Surgery Cancer Burns Trauma Chronic liver disease Fasting
1" 3` ? 3" 3" 1" 1"$ 3` 3`
$ $ $ ,[, $ $ $ $ ,I.
$ $ ,[, NK $ NK NK NK $
1"= increased; ,l, = decreased; NK = not known.
Table 2. Changes in parameters of the GH/IGF-I system in renal failure. Parameter
Change in renal failure
Fasting GH concentration GH half-life GH receptor numbers GH binding protein concentration Total IGF- 1 concentration IGF-I half-life IGFBP- 1 concentration IGFB P-3 concentration Acid-labile sub-unit concentration Protease
3` 1" NK $ -,1, $ "[" 3` ~ +-~
Reference Ross et al (1991b) Garcia-Mayor et al (1993)
Maheshwari et al (1992) Jacob et al (1990) Fouque et al (1995a) T6nshoff et al (1996) Tr et al (1996) Haffner et al (1997) Rabkin et al (1996)
1"= increased; ,[, = decreased; ~ = unchanged; NK = not known.
POSSIBLE MECHANISMS UNDERLYING AGHR The processes leading to the gross changes in GH and IGF-1 concentrations are discussed below with an emphasis on the changes in specific parameters of the GH/IGF- 1 system. Decreased IGF-1 concentration IGF-1 concentration represents a balance between synthesis and breakdown, modified by binding-protein levels, and so changes in any of these parameters could result in low levels of IGF-i.
IGF-1 secretion Direct measurement of IGF-1 secretion is difficult, and so inferences must be drawn from the available evidence on IGF-I concentration and metabolism. Essentially, in the presence of a low concentration of IGF-1 and decreased IGF-I half-life, IGF-1 secretion must be inadequate to restore normal IGF-1 levels, but it is not clear whether secretion is increased, normal or decreased in comparison to normal.
ACQUIRED GROWTH HORMONE RESISTANCE 1N ADULTS
319
IGF-1 half-life IGF-1 half-life is decreased after surgery (Miell et al, 1992), in critical illness (Yarwood et al, 1997) and in CRF treated by renal replacement therapy it has been reported to be reduced (Fouque et al, 1995a) or normal (Rabkin et al, 1996). The mechanism for this is not established. IGF-1 in the circulation exists as a ternary complex with IGFBP-3 and the ALS; it is freed to act by the action of a specific protease. There is experimental evidence that IGF-1 half-life is approximately 12-15 hours when in the ternary complex and only 10 minutes when free in the circulation (Giudice, 1995). Clearly, an alteration in the activity of IGFBP-3 protease would have a major effect on IGF-I half-life. Evidence to support this mechanism for the reduction in IGF-I half-life comes from the finding that levels of the protease are increased in illness (Davies et al, 1991) or after surgery (Cwyfan-Hughes et al, 1992; Davenport et al, 1992), but not in renal failure (Rabkin et at, 1996) or cirrhosis (Ross et al, 1996). This may lead to liberation of IGF-1 to the tissues; however, if this is occurring, then it also seems necessary for there to be resistance to the actions of IGF-1 in order for protein catabolism to persist. Evidence that this is the case comes from the finding of a decreased metabolic response to IGF-I in renal failure; amino acid and insulin concentrations fell less in patients than in controls after exogenous IGF-I (Fouque et al, 1995b).
IGFBP Low levels of IGF-I may just represent low levels of IGFBP and mask normal free levels of IGF-1. This does not appear to be the case in renal failure where levels of IGFBP-1 and IGFBP-3 are high (T6nshoff et al, 1996); however, some of this elevation may relate to accumulation of immunoreactive fragments (Rabkin et al, 1996). The normal regulation of IGFBP-l by insulin is preserved in critical illness (Ross et al, 1991a). Increased GH concentration
Increased concentration of GH could be due to either a prolonged GH halflife, increased secretion of GH or increased concentration of GHBP.
GH secretion GH secretion over 24 hours in patients with chronic liver disease is approximately doubled (Cuneo et al, 1995). The reported temporal patterns of GH secretion in critical illness differ according to the group studied. In liver cirrhosis more frequent secretory pulses of GH occur, with individual pulses being similar to normals in mass (Cuneo et al, 1995). In intensive care unit (ICU) patients GH secretory pulses are attenuated in amplitude but GH trough levels between pulses are high (Ross et al, 1991b; van den Berghe et al, 1994). In contrast, critically ill trauma victims were found to have less frequent secretory bursts of GH,
320
R. C. J E N K I N S A N D R. J. M. R O S S
although the sampling frequency of 60 minutes in the latter study was probably insufficient (Melarvie et al, 1995). Clearly, if the primary defect in AGHR is a low IGF-1 concentration then an increase in GH secretion could be caused by a loss of the usual negative feedback of IGF-1. This is supported by the findings in chronic liver disease of a relationship between the degree of rise in GH and the deficit in IGF-1 levels (Cuneo et al, 1995). This mechanism is analogous to the very high levels of GH reported in a child with an inherited defect in the IGF-1 gene who effectively had no circulating IGF-1 (Woods et al, 1996).
GH half-life Increased secretion of GH is unlikely to be the whole story, however, as GH half-life (of exogenously administered GH) has been reported to be increased by 25-50% (Haffner et al, 1994) or more (Garcia-Mayor et al, 1993; Schaefer et al, 1996) in CRE A similar finding, for endogenous GH, has been reported in chronic liver disease (Cuneo et al, 1995) and critical illness (Ross et al, 1991b). Whether the prolonged half-life is due to a reduction in GH binding to the liver or an increased GHBP is not clear, although, as discussed below, the latter seems unlikely.
GHBP Decreased concentrations of GHBP have been reported in patients with CRF (Maheshwari et al, 1992; Schaefer et al, 1996), chronic liver disease (Baruch et al, 1991) and critical illness (Ross et al, 1991a) and so the elevation in GH concentrations is likely to underestimate the increase in unbound GH in these conditions. The degree of reduction in GHBP in liver cirrhosis was greater in patients with more severe cirrhosis. GH receptor function The biochemical profile of acquired GH resistance is qualitatively similar to that of congenital GH resistance or Laron syndrome (Woods and Savage, 1996) in which, in the classical form, there is an inherited defect of the GH receptor (GH-R). Thus, attention in the acquired syndrome has also been directed towards the function of the GH-R. However, it must be noted that this hormonal profile could also be produced by a primary reduction in the concentration of IGF-1, by increased clearance for example, and a secondary increase in GH concentrations without there being a defect at the level of the GH-R. The translation of the GH signal to IGF-1 production may be interfered with by other circulating factors. Elevated prolactin concentrations, for example, appear to lead to lower than expected IGF-1 levels for a given GH concentration in acromegalic patients (Hulting et al, 1984) although the mechanism responsible is obscure. In uraemia there is no evidence of a circulating inhibitory serum factor (Maheshwari et al, 1992).
ACQUIRED
GROWTH
HORMONE
RESISTANCE
321
IN ADULTS
$ <---
W
g 8
E-
-+
$
6--
5 6-
W -+
=_ e~ - 9 --> - 9 ---->---->---->-~. --). --~ - 9 -+ r~
k3
II ~,,
-9
5_2
" " ~ " Zz - + - - , g $ =~
+--~ Z Z
It
=_ -< ="
=
3
;q 4. o
2-W S
W
322
R. C. J E N K I N S A N D R. J. M. ROSS
GH-R numbers Low levels of a normally functioning receptor could lead to AGHR but it is difficult to measure receptor numbers directly as appropriate liver tissue is unavailable from humans with AGHR. However, expression of the GH-R gene in skeletal muscle has been shown to be decreased after elective surgery (Hermansson et al, 1997). As the GHBP is derived from the same gene as the GH-R, either by alternative splicing or proteolytic cleavage, it is possible that concentrations of the GHBP could act as a surrogate marker for GH-R numbers. Animal studies may not be relevant to the human situation as in rats, at least, the GHBP is probably produced by a different mechanism than in humans. The rationale for this approach is supported by the finding of low levels of GHBP in most cases of congenital GH resistance. GHBP concentrations, as discussed above, have been found to be low in some groups with AGHR.
GH binding Direct assessment of the function of the GH-R is difficult. Clearly, the combination of raised GH and lowered IGF-1 concentrations suggests that the GH signal is not leading to normal IGF-1 levels, but this could merely reflect a defect at a post-receptor site. Binding of GH to the GHBP provides an analogy for GH binding to its receptor, and this has been reported to be normal in CRF (Maheshwari et al, 1992; Schaefer et al, 1996). Similarly, patients with varying degrees of liver failure also had normal binding of GH to the GHBP (Baruch et al, 1991).
Effect of pH Most receptors and their ligands bind optimally over a narrow pH range. It has been suggested that an altered pH in CRF may be responsible for AGHR (Carlin and Carlin, 1994) although we know of no experimental evidence to support this hypothesis in humans. In a number of the conditions associated with AGHR-A, pH changes as a result of renal failure, respiratory failure or poor tissue perfusion. However, experimental studies have determined that GH binds to its receptor over quite a large range of pH (6.6-8.2) (Murphy et al, 1983; Hung et al, 1985), and in renal failure GH-R binding affinity has been reported to be normal. Whether pH alters events after receptor binding is not known.
GH-R signalling If GH-R numbers and function are normal then reduced IGF-1 production must represent a post-receptor defect but no data are available to answer this question.
ACQUIRED GROWTH HORMONE RESISTANCE IN ADULTS
323
Nutritional status The nutritional state is known to modulate the function of the GHflGF-I axis, and fasting in normals leads to a state reminiscent of AGHR. For example, fasting for 24 hours led to a fivefold increase in GH production due to a doubling in the frequency of GH secretory bursts and an increase in the mass of each burst (Hartman et al, 1992). IGF- 1 levels and GH halflife in this study did not alter after 56 hours of fasting but the former has been reported to fall following calorie malnutrition (Isley et al, 1983; Musey et al, 1993). Institution of nutritional support is associated with a rise in IGF-I concentration in proportion to nitrogen balance, which is then reversed by withdrawal of support (Donahue and Phillips, 1989; Minuto et al, 1989). In patients with CRF the serum IGF-I concentration has been found to be a sensitive marker of nutritional state (Jacob et al, 1990) and to correlate with anthropometric measures (Sanaka et al, 1994). A prospective study of critically ill ICU patients found that IGF-1 was more strongly correlated with nitrogen balance than were conventional indicators of nutritional status, such as pre-albumin or albumin (Hawker et al, 1987). GHBP is also modulated by nutritional status, with low levels in malnutrition and high levels in obesity (Postel-Vinay et al, 1995). IGFBP-1 is negatively regulated by insulin, and levels rise during fasting (Musey et al, 1993). The abnormalities in fasting are not confined to the basal hormone concentrations--administration of GH (10mg daily for 5 days) to fed subjects led to a fourfold higher IGF-I concentration than when the treatment was given in the fasting state (Merimee et al, 1982), i.e. resistance to the actions of GH. Fasting can therefore lead to increased levels of GH, decreased IGF-I, decreased GHBP and high IGFBP-1 as in AGHR, and although it seems certain that inadequate nutrition contributes to the syndrome it is unlikely to be the whole story. The relative contributions of fasting and illness to the syndrome are illustrated by the finding that GH concentrations in critically ill subjects were similar to those of fasting normals, but IGF-I concentrations were much lower in patients than in controls (Ross et al, 1991b).
The effect of cytokines Cytokines are released in many of the conditions associated with AGHR, such as sepsis, trauma and burns. Complex and, at times, contradictory evidence implicates interleukin-l, interleukin-6 and tumour necrosis factor-c~ as modulating levels of somatostatin and GHRH (MandrupPoulsen et al, 1995). Recombinant interleukin-6 stimulates acute secretion of GH in humans (Tsigos et al, 1997), although the main effects of cytokines in AGHR are likely to be via effects on IGF-1 production. Interleukin-1-]3 and tumour necrosis factor-c~ inhibit GH receptor mRNA, IGF-1 mRNA and IGF-I protein production in tissue culture (Wolf et al, 1996). Interleukin-6 reduces IGF-1 levels in mice and possibly in humans also (De Benedetti et al, 1997), although the mechanism is not clear.
324
R. C. J E N K I N S A N D R. J. M. ROSS
EFFECTS OF AGHR
Is there any practical use of the concept of AGHR or is it merely a biochemical epiphenomenon? Is there evidence that the AGHR syndrome plays an active and deleterious role or does it just reflect the effect of other disease processes? The evolution of such a response to illness suggests that it has survival advantage but whether this holds true in patients with prolonged illness, such as those in the ICU, is uncertain. It is possible that, in the short term, it serves to conserve nutritional substrates, for example by allowing amino acids from the breakdown of muscle protein to be used for other purposes, and thus could be beneficial, but as illness becomes chronic and malnutrition supervenes it could become harmful by preventing anabolism. If AGHR is a beneficial metabolic reflex designed to protect against substrate use in stress then pharmacological strategies to 'correct' it may be harmful. On the other hand, if it is a deleterious state, perhaps by being permissive to catabolism, then treatment should be directed to reversing it. One way to resolve this uncertainty is to examine the evidence regarding the effects of exogenous GH and/or IGF-1 in AGHR (see below). The combination of high GH and low IGF-1 concentrations would be expected to lead to a predominance of the direct anti-insulin actions of GH over the anabolic effects mediated by IGF-1. Patients with advanced colorectal carcinoma have been found to have high hepatic glucose output in keeping with antagonism of insulin actions by GH (Tayek et al, 1990). Treatment to 'correct' AGHR
Assuming that AGHR is harmful, then how might therapy be directed to ameliorate or correct it? Other than by treating the underlying condition and optimizing nutritional support, correction of the hormonal imbalance can be approached in three ways--administration of exogenous GH, exogenous IGF-I, or both together, and all these avenues have been explored. These treatment regimens have been studied in a number of ways, predominantly using surrogate metabolic end-points to imply that there would be beneficial clinical effects; as yet there are few published studies with hard clinical end-points. As each of the conditions associated with AGHR exhibit different pathophysiological mechanisms, it is unlikely that evidence of favourable or adverse effects of treatment in one group could be safely extrapolated to other groups. Treatment of catabolism with GH and IGF-I has been recently reviewed (Jenkins and Ross, 1996), and this chapter provides only an overview of relevant results. The use of GH
In principle, it should be possible to overcome GH resistance by giving high doses of exogenous GH; this hypothesis is supported by the study of fasting obese subjects who exhibited an increase in IGF-1 concentration after GH treatment (Snyder et al, 1990). Metabolic improvements have
A C Q U I R E D G R O W T H H O R M O N E R E S I S T A N C E 1N A D U L T S
325
been seen in patients with sepsis or trauma given GH (20 units daily for 3 days) who had significantly reduced net protein loss compared to a control period (Douglas et al, 1990). Patients with gastrointestinal or pancreatic disease on parenteral nutrition had significantly greater retention of nitrogen, potassium and phosphate when treated with GH (10 mg daily for 14 days) (Ziegler et al, 1992). Immobilized, neurogically injured patients did not improve their nitrogen balance with GH treatment (0.2 mg/kg daily for 7-13 days) but did have improvements in albumin and transferrin concentrations (Behrman et al, 1995). Children with CRF have GH resistance and respond to exogenous GH by increasing growth (T6nshoff et al, 1990), but in adults it is more difficult to demonstrate benefit. Definite clinical benefits of GH treatment in adults have been reported more rarely. A large study of patients after elective cholecystectomy found reduced wound infections, possibly related to improved cell-mediated immunity, and shorter hospital stay (Vara-Thorbeck et al, 1993). Patients with major burns have been reported to have decreased mortality (11 versus 37%) following treatment with GH (mean dose 0.11 mg/kg) (Knox et al, 1995). Negative clinical studies have found no anabolic effect of GH (0.125mg/kg) in patients with cancer (Tayek and Brasel, 1995) or in patients with sepsis or trauma receiving huge doses of GH (1.5 unit/kg daily) (Gottardis et al, 1991). More recently, two large multicentre trials of the use of GH in intensive care patients have been terminated early due to significantly increased mortality in the patients treated with GH (41.7 versus 18.2%) (Pharmacia & Upjohn communication). The analysis of this study is yet to be published, but the trial included 532 patients with either open-heart surgery, abdominal surgery, multiple trauma or acute respiratory failure and used GH doses of either 16 or 24 IU/day. In contrast, Serono have found no increase in mortality in an interim analysis of 450 patients, predominantly with HIV-associated wasting, who received a similar dose of GH. Similarly, Lilly have also found no increase in mortality in a study of 166 patients with hip fracture, liver transplant or severe bums. Firm conclusions will be possible only when these trials have been fully analysed and published, but it is possible that GH could be beneficial in some groups of patients and harmful in others. The use of lGF-1
IGF- 1 has been given to patients with AIDS, but the results were disappointing (Lieberman et al, 1994). An impaired metabolic response to IGF-1 was also seen in patients with CRF (Fouque et al, 1995b). The long-term effects may have been limited by feedback suppression of endogenous GH and thus reduced IGFBP-3 and ALS leading to tachyphylaxis. In addition, IGF-1 has actions similar to those of insulin and so can lead to hypoglycaemia. The use of combined GH and 1GF-1
The combined use of these agents may be the most potent approach. The potential advantages of this approach are that it will avoid the
326
R. C. JENKINS AND R. J. M. ROSS
hypoglycaemia sometimes seen when IGF-I is given alone and that exogenous GH may correct for the inhibition of endogenous GH caused by the administration of IGF-1 and allow the production of IGFBP-3 and ALS to be maintained or increased--thus preserving the anabolic actions of IGF-1. Use of this combination in fasting subjects produced a potent nitrogen-retaining effect, no hypoglycaemia and increased concentration of IGFBP-3 and ALS (Kupfer et al, 1993). However, combined treatment of HIV-infected patients did not produce significant sustained anabolic effects (Ellis et al, 1996; Lee et al, 1996).
SUMMARY There is now abundant evidence that in a range of acute and chronic illnesses associated with catabolism there is a consistent perturbation in the GH/IGF-1 system characterized by raised levels of GH and low levels of IGF-1. It remains unclear whether this biochemical finding is merely a nonspecific consequence of illness or whether it contributes adversely to the nutrition and survival of the patient. A similar profile occurs in fasting, and this may underlie some, but not all, of the syndrome. Different conditions manifest AGHR to different extents. In many, there is induction of a protease which reduces IGF-1 half-life, and GH concentrations probably rise due to the removal of IGF- 1 negative feedback. The varying therapeutic strategies which have used exogenous recombinant human GH and/or IGF-1 to correct the syndrome have met with limited clinical success, such as in patients with AIDS, but more usually have been successful only at altering metabolic parameters. It is possible that AGHR represents a protective metabolic reflex designed to deal with protein-calorie malnutrition, and so treatment to overcome it could be detrimental. This question may be answered by ongoing clinical studies.
Acknowledgements This work was supported by the University of Sheffield, the Northern General Hospital NHS Trust and Pharmacia and Upjohn Limited.
REFERENCES Abribat T, Brazeau P, Davignon I & Garrel DR (1993) Insulin-like growth factor-I blood levels in severely burned patients: effects of time post injury, age of patient and severity of burn. Clinical Endocrinology 39: 583-589. Arnold J, Campbell IT, Samuels TA et al (1993) Increased whole body protein breakdown predominates over increased whole body protein synthesis in multiple organ failure. Clinical Science 84: 655-661. Baruch Y, Amit T, Hertz Pet al (1991) Decreased serum growth hormone-binding protein in patients with liver cirrhosis. Journal of Clinical Endocrinology and Metabolism 73: 777-780. Behrman SW, Kudsk KA, Brown RO et al (1995) The effect of growth hormone on nutritional markers in enterally fed immobilised trauma patients. Journal ~)f Parenteral and Enteral Nutrition 19: 41-46.
ACQUIRED GROWTH HORMONE RESISTANCE IN ADULTS
327
De Benedetti F, Alonzi T, Moretta Aet al (1997) lnterleukin-6 causes growth impairment in transgenic mice through a decrease in insulin-like growth factor-I. A model for stunted growth in children with chronic inflammation. Journal of Clinical Investigation 99: 643-650. *van den Berghe, G, De Zegher E Lauwers P & Veldhuis JD (1994) Growth hormone secretion in critical illness: effect of dopamine. Journal of Clinical Endocrinology and Metabolism 79: 1141 - 1146.
Berneis K & Keller U (1996) Metabolic actions of growth hormone: direct and indirect. BailliOre's Clinical Endocrinology and Metabolism 10: 337-352. Carlin K & Carlin S (1994) A possible explanation for growth hormone resistance in chronic renal failure. Medical Hypotheses 43: 193-194. Cotterill AM, Mendel P, Holly JMP et al (1996) The differential regulation of the circulating levels of the insulin-like growth factor binding proteins (IGFBP) I, 2 and 3 after elective abdominal surgery. Clinical Endocrinology 44: 91-101. Cuneo RC, Hickman PE, Wallace JD et al (1995) Altered endogenous growth hormone secretory kinetics and diurnal GH-binding protein profiles in adults with chronic liver disease. Clinical Endocrinology 32: 265-275. Cuthbertson DP (1931) Observations on the disturbance of metabolism produced by injury to the limbs. Quarterly Journal qf Medicine 2: 233-246. Cwyfan-Hughes SC, Cotterill AM, Molloy AR et al (1992) The induction of specific proteases for insulin-like growth factor-binding proteins following major heart surgery. Journal of Endocrinology 135:135-145. Davenport ML, lsley WL, Pucilowska JB et al (1992) Insulin-like growth factor-binding protein-3 proteolysis is induced after elective surgery. Journal of Clinical Endocrinology and Metabolism 75:590 595. *Davies SC, Wass JA, Ross RJ et al (1991) The induction of a specific protease for insulin-like growth factor binding protein-3 in the circulation during severe illness. Journal of Endocrinology 130: 469-473. Donahue SP & Philips LS (1989) Response of IGF-I to nutritional support in malnourished hospital patients: a possible indicator of short-term changes in nutritional status. American Journal c{f Clinical Nutrition 50: 962-969. *Douglas RG, Humberstone DA, Haystead A & Shaw JHF (1990) Metabolic effects of recombinant human growth hormone: isotopic studies in the postabortive state and during total parenteral nutrition. British Journal ~f Surgery 77: 785-790. Ellis KJ, Lee PDK, Pivarnik JM et al (1996) Changes in body composition of human immunodeficiency virus-infected males receiving insulin-like growth factor I and growth hormone. Journal r Clinical Endocrinology and Metabolism 81: 3033-3038. Fouque D, Peng SC & Kopple JD (1995a) Pharmacokinetics of recombinant human insulin-like growth factor-1 in dialysis patients. Kidney International 47: 869-875. Fouque D, Peng SC & Kopple JD (1995b) h'npaired metabolic response to recombinant insulin-like growth factor-1 in dialysis patients. Kidney h~ternationa147: 876-883. Frayn KN, Price DA, Maycock PF & Carroll SM (1984) Plasma somatomedin activity after injury in man and its relationship to other hormonal and metabolic changes. Clinical Endocrinology 20: 179-187. Garcia-Mayor RVG, Perez A J, Gandara Aet al (1993) Metabolic clearance rate of biosynthetic growth hormone after endogenous growth hormone supp~'ession with a somatostatin analogue in chronic renal failure patients and control subjects. Clinical Endocrinology 39: 337-3~.3. Giudice LC (1995) IGF binding protein-3 protease regulation: how sweet it is! Journal qf Clinical Endocrinology and Metabolism 80: 2279-2281. Gottardis M, Benzer A, Koller W e t al (1991) Improvement of septic syndrome after administration of recombinant human growth hormone? Journal c)fTraumu 31:81-86. Haffner D, Schaefer F, Girard J e t al (1994) Metabolic clearance of recombinant human growth hormone in health and chronic renal failure. Journal Of Clinical Investigation 93: 11631171.
Haffncr D, Trnshoff B, Blum WF et al (1997) Insulin-like growth factors (IGFs) and IGF binding proteins, serum acid-liable subunit and growth hormone binding protein in nephrotic children. Kidney hlternational 52:802-8 I0. *Hartman ML, Veldhuis JD, Johnson ML et al (t992) Augmented growth hormone (GH) secretory burst frequency and amplitude mediate enhanced GH secretion during a two-day fast in normal men. Journal qf Clinical Endocrinology and Metabolism 74: 757-765.
328
R. C. JENKINS AND R. J. M. ROSS
Hawker FH, Stewart PM, Baxter RC et al (1987) Relationship of somatomedin-C/insulin-like growth factor I levels to conventional nutritional indices in critically ill patients. Critical Care Medicine 15: 732-736. *Hermansson M, Wickelgren RB, Hammarqvist F et al (1997) Measurement of human growth hormone receptor messenger ribonucleic acid by a quantitative polymerase chain ~reaction-based assay: demonstration of reduced expression after elective surgery. Journal of Clinical Endocrinology and Metabolism 82:421-428. Hulting AL, Hall K & Werner S (1984) Somatomedin A increments are diminished in acromegaly with concomitant hyperprolactinaemia. Acta Endocrinologica 106: 305-310. Hung CH, Kover K & Moore WV (1985) Characteristics of somatotropic growth hormone binding to homologous liver plasma membranes. Molecular and Cellular Endocrinology 39:189-196. IsIey WL, Underwood LE & Clemmons DR (t983) Dietary components that regulate serum somatomedin-C concentrations in humans. Journal of Clinical Investigation 71:175-182. Jacob V, Le Carpentier JE, Salzano S et al (1990) IGF-I, a marker of undernutrition in hemodialysis patients. American Journal of Clinical Nutrition 52: 39-44. Jeevanandam M, Ramias L, Shamos RF & Schiller WR (1992) Decreased growth hormone levels in the catabolic phase of severe injury. Surgery 111: 495-502. Jenkins RC & Ross RJM (1996) Growth hormone therapy for protein catabolism. Quarterly Journal of Medicine 89: 813-819. Knox J, Demling R, Wilmore D et al (1995) Increased survival after major thermal injury: the effect of growth hormone therapy in adults. Journal of Trauma 39: 526-530. *Kupfer SR, Underwood LE, Baxter RC & Clemmons DR (1993) Enhancement of the anabolic effects of growth hormone and insulin-like growth factor-I by use of both agents simultaneously. Journal of Clinical Investigation 91:391-396. Lee PDK, Pivarnik JM, Bukar JG et al (1996) A randomized, placebo-controlled trial of combined insulin-like growth factor 1 and low dose growth hormone therapy for wasting associated with human immunodeficiency virus infection. Journal of Clinical Endocrinology and Metabolism 81: 2968-2975. Lieberman SA, Butterfield GE, Harrison D & Hoffman AR (1994) Anabolic effects of recombinant insulin-like growth factor-I in cachetic patients with the acquired immunodeficiency syndrome. Journal of Clinical Endocrinology and Metabolism 78: 404-410. Maheshwari HG, Rifkin I, Butler J & Norman M (1992) Growth hormone binding protein in patients with renal failure. Acta Endocrinologica 127: 485-488. Mandrup-Poulsen T, Nerup J, Reimers Jl et al (1995) Cytokines and the endocrine system. I. The immunoendocrine network. European Journal of Endocrinology 133: 665-666. Melarvie S, Jeevanadam M, Holaday NJ & Petersen SR (1995) Pulsatile nature of growth hormone levels in critically ill trauma victims. Surge O' 117: 402-408. Merimee TJ, Zapf J & Froesch ER (1982) Insulin-like growth factors in the fed and fasted states. Journal of Clinical Endocrinology and Metabolism 55: 999-1002. *Miell JP, Taylor AM, Jones J e t al (1992) Administration of human recombinant insulin-like growth factor-I to patients following major gastrointestinal surgery. Clinical Endocrinology 37:542-551. Minuto E Barreca A, Adami GF et at (1989) Insulin-like growth factor-I in human malnutrition: relationship with some body composition and nutritional parameters. Journal of Parenteral and Enteral Nutrition 13: 392-39@ Murphy LJ, Vrhovsek E & Lazarus L (1983) Characterisation of specific growth hormone binding sites in mouse fibmblasts. Endocrinology 113: 750-757. Musey VC, Goldstein S, Farmer PK et al (1993) Differential regulation of IGF-I and [GF-binding protein-1 by dietary composition in humans. American Journal of Medical Science 305: 131-138. Phillips LS & Unterman TG (1984) Somatomedin activity in disorders of nutrition and metabolism. Clinical Endocrinology and Metabolism 13:145-189. Postel-Vinay M-C, Saab C & Gourmelen M (1995) Nutritional status and growth hormone-binding protein. Hormone Research 44: 177-181. Rabkin R, Fervenza FC, Maidment H et al (1996) Pharmacokinetics of insulin-like growth factor-I in advanced chronic renal failure~ Kidney h~ternationa149: l 134-1140. Rondanelli M, Solerte SB, Fioravanti Met al (1997) Circadian secretory pattern of growth hormone, insulin-like growth factor type I, cortisol, adrenocorticotropic hormone, thyroid-stimulating hormone, and prolactin during HIV infection. Aids Research and Human Retroviruses 13: 1243 - 1249.
ACQUIRED GROWTH HORMONE RESISTANCE IN ADULTS
329
Ross RJM, Goodwin FJ, Houghton BJ & Boucher BJ (1985) Alteration of pituitary thyroid function in patients with chronic renal failure treated by haemodialysis or continuous ambulatory peritoneal dialysis. Annals of Clinical Biochemistry 22: 156-160. Ross RJ, Miell JE Holly JM et al (1991a) Levels of GH binding activity, IGFBP-1, insulin, blood glucose and cortisol in intensive care patients. Clinical Endocrinology 35: 361-367. *Ross RJM, Miell J, Freeman E et al (1991b) Critically ill patients have high basal growth hormone levels with attenuated oscillatory activity associated with low levels of insulin like growth factorI. Clinical Endocrinology 35: 47-54. Ross RJ, Chew SL, D'Souza L et al (1996) Expression of IGF I and IGF-binding protein genes in cirrhotic liver. Journal of Endocrinology 149: 209-216. Sanaka T, Shinobe M, Ando Met a[ (1994) IGF-I as an early indicator of malnutrition in patients with end-stage renal disease. Nephron 67: 73-81. Schaefer F, Baumann G, Haffner D et al (1996) Multifactorial control of the elimination kinetics of unbound (free) growth hormone in the human: regulation by age, adiposity, renal function, and steady state concentrations of GH in plasma. Journal rff Clinical Endocrinology and Metabolism 81:22 31. Snyder DK, Underwood LE & Clemmons, DR (1990)Anabolic effects of growth hormone in obese diet-restricted subjects are dose dependent. American Journal Of Clinical Nutrition 52:431-437. *Tayek JA & Brasel JA (1995) Failure of anabolism in malnourished cancer patients receiving growth hormone: a clinical research center study. Journal o/Clinical Endocrinology and Metabolism 80: 2082-2087. Tayek JA, Bulcavage L & Chlebowski RT (1990) Relationship of hepatic glucose production to growth hormone and severity of malnutrition in a population with colorectal carcinoma. Cancer Research 50:2119 2122. TOnshoff B, Blum WF & Mehls O (1996) Serum of insulin like growth factors and their binding proteins in children with end-stage renal disease. Pediatric Nephrology 10: 269-274. TOnshoff B, Mehls O, Heinrich U et al (1990) Growth-stimulating effects of recombinant human growth hormone in children with end-stage renal disease. Journal ~['Pediatrics 116:561-566. Tsigos C, Papanicolaou DA, Defcnsor R et al (1997) Dose effects of recombinant human interleukin6 on pituitary hormone secretion and energy expenditure. Neuroendocrinology 66: 54-62. Vara-Thorbeck R, Guerrcro JA, Rosell J e t al (1993) Exogenous growth hormone: effects on the catabolic response to surgically produced acute stress and on post-operative immune function. World .lournal qf Surgel3' 1 7 : 5 3 0 - 5 3 8 Voerman H J, Strack van Schijndcl RJM, Groenevcld ABJ et al (1992) Pulsatile hormone secretion during scvere sepsis: accuracy of different blood sampling regimens. Metabolism 9: 934-940. Wihnore, DW (1991) Catabolic illness. Stratcgies for enhancing recovery. New England Journal (ff Medicine 325:695 702. Wolf M, Bohm S, Brand M & Kreymann G (1996) Proinflammatory cytokines interleukin 1 beta and turnout necrosis factor alpha inhibit growth hormone stimulation of insulin-like growth factor I synthesis and growth hormone receptor mRNA levels in cultured rat liver cells. European Journal c)f Endocrinology 135: 729-737. Woods KA & Savage MO (1996) Laron syndrome: typical and atypical forms. Baillikre~ Clinical Fndocrim~logy and Metabolism 10:371 387. Woods KA, Camacho-Hubner C, Savage MO & Clark AJL (1996) Intrauterine growth retardation and postnatal growth failure associated with deletion of the insulin-like growth factor-I gene. New E~tgland Journal (~fMedicine 335:1363 1367. Wright AD, Lowy C, Fraser TR et al (1968) Serum growth hormone and glucose intolerance in renal failure. Lancet ii: 798-801. Yarwood G[), Ross RJM, Medbak S e t al (1997) Administration of human recombinant insulin tike growth l~tctor-I to critically ill pafieuts. Critical Ca~e Medwine 25: 1352-1361. *Ziegler TR, Rombeau JL, Young LS el a] (1992) Recombinant human growth hormone enhances the metabolic efficacy of parenteral nutrition: a double-blind, randomized controlled sludy Journal ~! Clinical Endocrinology .nd Metabolism 74: 865- 873.