The success of liver transplantation has required a redefinition of interim management of end-stage liver disease with a view to predicted outcome and timely intervention (1,2). This includes a more aggressive approach to supportive treatment to maintain, if not improve, the patient's clinical status if the treatment option of transplantation is decided on. Studies of children referred for orthotopic liver transplantation (3-6) have indicated that malnutrition, growth failure, and rapidly declining nutritional status, are major factors adversely affecting outcome. Thus, strategies to maintain or improve nutritional status and growth while awaiting transplantation deserve further study. We have previously documented the nature of malnutrition in children with end-stage liver disease, evaluated the role of nutritional therapy (4,6), and suggested that malnutrition is one area of pretransplant care in which improvements can be made.
In this double-blind, randomized, controlled study, we examined the effects of growth hormone (GH), as an adjunct to nutritional therapy, on body composition (height, weight, total body potassium [TBK], total body fat), thermogenic response (resting energy expenditure and respiratory quotient), the somatotropin-somatomedin axis, and biochemical indices of liver disease. Growth hormone is important for optimal growth in children (7,8) and has significant anabolic effects mediated by insulin-like growth factor-I (IGF-I) (9). Although patients with severe liver disease may not produce sufficient IGF-I in response to exogenous GH to promote anabolism (10), the effects of GH in patients with liver disease have not been fully tested, to our knowledge.
MATERIALS AND METHODS
The study was designed as a randomized, double-blind, placebo-controlled, crossover study of 12 weeks' duration (one visit every 2 weeks, totalling seven hospital visits). Patients were screened at the first visit for eligibility.
- Male or female children less than 14 years old
- Biopsy-proven cirrhosis of the liver
- Ability and willingness to follow a dietary regimen directed by the dietitian
- A parent or guardian willing to give written informed consent for the child's participation
- Symptomatic hyperglycemia
- Significant renal impairment
- Significant cardiac impairment
- Current treatment with corticosteroid hormones, immunosuppressive drugs, or insulin
- Major operation within 1 month
- Paracentesis-dependent ascites
- Concurrent chronic infection
- Treatment with any form of growth hormone during the past 3 months
Patients for whom parental or guardian consent had been obtained were randomized at the second visit to receive either growth hormone (recombinant human growth hormone, GH) 0.2 U/kg daily by subcutaneous injection, or placebo in equivalent volume by the same route, for 4 weeks (28 injections). Growth hormone was presented as a sterile powder with solvent in a two-compartment cartridge, the reconstituted solution containing recombinant 16 IU GH (somatropin), 2 mg glycine, 41 mg mannitol, 0.28 mg anhydrous sodium phosphate, 0.29 mg anhydrous sodium acid phosphate, 3 mg M-cresol, and 1 ml water for injection. Placebo was presented in the same way: The reconstituted solution had identical composition with the exception of GH, which was excluded. Parents or caregivers were carefully trained in injection technique. Compliance was assessed by reviewing a daily diary and vial accounting. At the end of the 4-week treatment period, a 2-week washout was followed by another 4-week period of treatment with the complementary drug(GH or placebo, whichever was not received during the first treatment period). Visits were scheduled every 2 weeks, at the beginning and end of each treatment period for evaluation of study parameters, and during the study periods for safety evaluation.
At the beginning and end of each treatment period, patients were evaluated for height, weight, total body fat (11), four-site skinfold thickness, mid-upper arm circumference, girth, blood pressure, concomitant medications, body cell mass by TBK (12,13) scanning, resting energy expenditure, respiratory quotient by indirect calorimetry (14,15), and lidocaine metabolism test (16).
Total Body Potassium
The body cell mass represents the oxygen-exchanging, glucose-oxidizing and work-performing tissue and is the best reference for expressing rates of metabolic processes (12). It is composed principally of muscle and viscera, and is therefore an excellent nutritional parameter, because these tissues are affected by periods of nutritional deprivation. Body cell mass is closely related to TBK, in that most body potassium is contained intracellularly (13,17,18). Total body potassium was measured, using a whole body counter (Accuscan, Canberra Industries, Boston, MA, U.S.A.), which detects the naturally occurring isotope 40K.
Resting Energy Expenditure
Resting energy expenditure was measured by open-circuit, indirect calorimetry, using a canopy to collect expired respiratory gases. Subjects were studied in fasting (4 hours) and resting states after a 3-minute stabilization (steady state) period for 10 minutes in a humidity- and temperature-controlled environment. Resting energy expenditure is calculated using Weir's equation from minute-by-minute measurement of oxygen consumption(Vo2) and carbon dioxide output (Vco2) (14,15).
Lignocaine is rapidly cleared from circulation by the hepatic cytochrome P450-dependent pathway and metabolized to monoethylglycinexylidide (MEGX). The rate of appearance of MEGX has been used as a quantitative measure of hepatic function. After a baseline blood sample had been obtained, 0.75 mg/kg lignocaine was infused intravenously. Samples for MEGX assay were taken 15 and 30 minutes after lignocaine administration (16,19).
Blood was taken for routine assessment of hematologic and multiple biochemical profile. Additional blood was taken for the purposes of the study of IGF-I. If sufficient blood was available, insulin-like growth factor-II(IGF-II), GH binding protein (GHBP), insulin-like growth factor binding protein-3 (IGFBP-3), insulin-like growth factor binding protein-1 (IGFBP-1), plasma amino acids, thyroid hormones, β-hydroxybutyrate, high density lipoprotein, insulin, C-peptide, and glucagon were measured. Number of albumin infusions in the previous 2 weeks was recorded at each visit. Assays for IGF-I and -II, (20) GHBP, (21), and IGFBP3 and IGFBP1 (22,23) were performed by Dr. P. Owens, (Commonwealth Scientific and Industrial Research Organization, Adelaide), Dr. R. Barnard, (Queensland University of Technology, Brisbane), and Dr. R. Baxter (Kolling Institute, Sydney), respectively. MEGX assays(19) were performed by Dr. J. Potter, Princess Alexandra Hospital, Brisbane. All other assays were performed by Royal Brisbane Hospital Pathology Department using standard commercial methods.
Adverse events were defined as any undesirable event occurring during the trial, whether or not they were considered to be related to the study drug. A serious adverse event was defined as any event that required or prolonged hospitalization, that caused permanent disability, that was fatal or life threatening (cancer or overdose), or that resulted from a congenital anomaly.
All patients had end-stage liver disease with biopsy-confirmed hepatic cirrhosis, jaundice, biochemical abnormalities, and hepatic synthetic dysfunction and had been accepted for liver transplantation by the Queensland Liver Transplantation Committee. All patients with biliary atresia had portal hypertension. Of 15 patients who were randomized to receive GH/placebo treatment, 10 completed the entire study. Three were withdrawn because liver transplantation was performed before study completion, one for deteriorating health and one because baseline values were not obtained. Only the 10 patients who completed the study are reported here. Clinical characteristics are described in Table 1. Age at study entry was 3.06± 1.15 years (range, 0.47-11.65 years). Eight patients had biliary atresia and 2 had Alagille's syndrome. Of those with biliary atresia, 7 underwent a Kasai procedure at a mean age of 0.19 ± 0.02 years (range, 0.12-0.23 years); the remaining patient proceeding directly to the option of liver transplant waiting list. Of those 7 who underwent Kasai procedures, one had a "successful" Kasai, which resulted in her acceptance for orthotopic liver transplantation at 11.53 years. The others were accepted for the transplant list at a mean age of 0.98 ± 0.14 years (range, 0.51-1.79 years).
Nine patients were regarded as compliant, defined as more than 80% of injections received. One patient, in whom more than 20% of injections were missed during the placebo period, was considered noncompliant but is included. All patients received standard nutritional and conventional therapy, as previously described (6), including enteral nutrition and branched-chain amino acid supplement. Albumin infusions were administered to maintain serum albumin at more than 30 g/l (regarded by clinicians as a conservative lower level that minimizes such clinician problems as ascites and edema). We have previously shown that branched-chain amino acid supplementation in children awaiting liver transplantation significantly improves weight, height, TBK, mid-upper arm circumference, and skin-fold thickness, and that it significantly reduces the number of albumin infusions required (6). Branched-chain amino acid supplementation is now a standard nutritional protocol applied to all patients awaiting liver transplantation. Patients are seen weekly in the clinic, and routine hematology and biochemistry are reviewed.
Changes in TBK for height and IGF-I response to GH were regarded as major outcomes, and statistical analysis for these two variables has been presented for both "all patients" (n = 10) and "compliant patients"(n = 9). For other variables, all available data were used. Changes in continuous normal parameters during the two treatment periods were compared by paired t-test. All continuous raw data were tested for normality; skewness was insufficient to warrant the use of nonparametric tests. Results are presented as mean values ± SEM, unless otherwise stated. Nonparametric data (number of albumin infusions) were analyzed using the Kruskal-Wallis test. Statistical analysis was performed independently by Medical Biostatistics Pty Ltd, Brisbane.
The study was approved by the Ethics Committee of the Royal Children's Hospital. Informed, written consent was obtained from all parents or guardians.
Body Composition and Nutrition
At baseline, mean z-score for height was -0.99 ± 0.54(range, -3.76-1.15), mean weight-for-age z-score was -0.05± 0.5 (range, -2.49-1.58), and mean weight for heightz-score was 0.75 ± 0.43 (range, -1.41-3.43). There were no significant differences between the effect of GH treatment or placebo on change in height, weight, TBK (figure 1), or body fat percentage. Mean TBK adjusted for height decreased from 0.277 ± 0.03 g/cm to 0.273 ± 0.04 g/cm during placebo treatment and increased from 0.243 ± 0.03 g/cm to 0.278 ± 0.03 g/cm during GH treatment(p = 0.112). Mean body fat as a percentage of body weight increased from 19.14 ± 1.45% to 19.27 ± 1.64% during placebo treatment and from 18.9 ± 1.98% to 19.47 ± 1.81% during GH treatment (p = 0.727). Thus, GH had no effect on body composition compared with placebo. Similarly, GH had no effect on resting energy expenditure or respiratory quotient.
Growth hormone administration was associated with a significant decrease in mean serum bilirubin concentration, which increased from 199.37 ± 49.06 µmol/l to 217.62 ± 50.02 µmol/l during placebo treatment and decreased from 236.13 ± 59.47 µmol/l to 201.5± 46.62 µmol/l during GH treatment (p = 0.02). There was no significant effect of the number of albumin infusions, results of hematologic analysis, prothrombin time, or serum enzymes (lactate dehydrogenase, aspartate amino-transferase, γ-glutamyl transferase and alanine amino-transferase.
Growth hormone had no significant effect on serum IGF-I(figure 2) compared with placebo treatment. Mean IGF-I increased from 32.5 ± 6.94 ng/ml to 24.5 ± 7.54 ng/ml during placebo treatment and from 38.2 ± 7.34 ng/ml to 43.1 ± 8.08 ng/ml during GH treatment (p = 0.645).
Similarly, there was no effect on serum IGF-II, GHBP, IGFBP-3, or IGFBP-1(Table 2). Mean GHBP levels were in the low to normal range compared with published values (25), and on the order of one tenth of normal values for adult women from our laboratory(normal range for adult women, 0.35-1.0 nmol/l). Four of 10 patients had undetectable levels of GHBP on one or more occasions. Mean IGFBP-3 ranged from 0.65 to 0.82 µg/ml, approaching the low to normal range for age from our laboratory. (Normal ranges from our laboratory are 0.4 to 2.4µg/ml for infants aged 1 to 12 months, and 1.1 to 3.5 µg/ml for children aged 1 to 5 years). Normal values for IGFBP-1 are not available for this age group. Mean ranges were below normal ranges available from our laboratory for Tanner stage I pubertal adolescents (273-448 µg/l); higher values would be expected in pre-pubertal children(26). Growth hormone had no significant effect on plasma amino acids, thyroid hormones, β-hydroxybutyrate, high density lipoprotein, insulin, C-peptide, or glucagon (data not shown).
Adverse Side Effects
There was no effect on lidocaine metabolism tests, blood pressure, blood glucose, or occurrence of ascites or edema. No category of serious adverse event other than hospitalization occurred during the course of this study, and the number of adverse events was statistically similar during the GH and placebo treatment phases. Monoethylglycinexylidide (MEGX) recovery decreased from 11 ± 4.22 µg/l to 6 ± 4.25 µg/l during GH treatment and from 4.43 ± 2.4 µg/l to 2.57 ± 1.67µg/l during placebo treatment. The differences were not significant.
Mean TBK adjusted for height and IGF-I values were statistically similar at the beginning of GH and placebo treatment periods (p > 0.2), suggesting that the washout period was adequate and that patients returned to baseline after cessation of GH therapy. There was no effect of order of treatment on outcome-that is, mean changes in TBK and IGF-I were similar, regardless of whether GH was administered in the first or second treatment period.
Our study confirms previous studies in which results have shown apparent biochemical GH insensitivity in children with end-stage liver disease(25,27), further indicating that GH therapy has no major clinical benefits, although total bilirubin levels showed significant reduction in association with GH administration.
Body Composition and Nutrition
The children in this study had better nutrition compared with previous cohorts of end-stage liver disease patients (24). When IGF-I levels were measured in previous studies, mean z-score for height was -2.7 ± 0.5 and z-score for weight was -2.0± 0.5 (24). Bucuvalas et al.(25) reported a group of 5 patients in whom heightz-score was -2.4 ± 0.39 and weight z-score was-2.1 ± 0.25. Children in the current study had normalz-score for weight and weight-for-height, and although weight is not a good measure of nutritional status in liver disease because of alterations in body fluid dynamics, this is still an improvement on the status of previously reported children. Growth hormone administration failed, however, to have any measurable effect on height, weight, TBK, or total body fat.
Despite the lack of GH response, somatomedin axis is evaluated in detail in this study and is worthy of comment. IGF-I levels were not significantly different (p > 0.1) when compared with levels measured in the same laboratory in a group of cystic fibrosis patients (mean IGF-I, 51± 8.0 ng/ml) (20). Similarly, IGF-I levels appear higher than those reported by other investigators of children with end-stage liver disease (25), although levels from different laboratories are not directly comparable. In the study subjects, IGF-I production, possibly from peripheral tissues, may have been induced by other factors, most probably enteral nutrition, in that circulating IGF-I levels are predominantly regulated by nutritional intake and GH (28,29). Although most of the circulating IGF-I is traditionally regarded as originating in the liver, many tissues express the IGF-I gene in a GH-independent manner (30); thus, GH resistance may not preclude at least some IGF-I production. However, IGF-I synthesis in nonliver tissues may also be GH dependent (31). It is possible that children with liver disease may produce IGF-I from nonliver tissues, a process that may or may not be GH dependent. IGF-II levels in the current series were much lower than those previously reported by us, both in children with end-stage liver disease and in cystic fibrosis patients(24). The low observed levels of IGF-II may related to high levels of circulating GH; it has been shown in growing pigs that GH increases IGF-I and decreases IGF-II (32). Results of previous studies have shown that endogenous GH levels are high in patients with liver disease (33,34), and exogenously administered hormone would render circulating levels even higher. It is difficult to reconcile this with decreased levels of circulating binding protein observed in the current study, which has also been reported in other patients with liver disease (35). A possible explanation may be that these children have upregulated peripheral and/or hepatic cellular GH receptors, although increased numbers of hepatocyte membrane receptors would seem unlikely. Growth hormone receptor mRNA has demonstrated in rat adipose tissue, growth plate, kidney, intestine, lung, muscle, pancreas, brain, testis, ovary adrenal, skin, heart, and stomach, as well as liver (36). Such peripheral receptor upregulation would be conceivable in the face of a catabolic disease, although cell receptor up-regulation implies failure of intracellular transcription or translation of IGF-I-that is, a postreceptor defect. The extracellular domain of the GH receptor is the same molecular identity as circulating GHBP(37), but it is not known whether these proteins are expressed concordantly with each other. There is some evidence that they are under separate control (37). Most likely, these patients with end-stage liver disease have deficits in both GH receptor and GHBP production due simply to failure of hepatic synthesis. Levels of IGFBP-3 appear to correspond with the observed relatively improved IGF-I levels. The liver appears to be the source of the majority of circulating GHBP-1(38,39), so decreased levels measured in these may be directly caused by liver disease. Additionally, pertubations in insulin secretion may affect IGFBP-1, which is related to circulating insulin levels (26).
In summary, exogenous GH had no effect on body composition or somatomedin axis in the study children with end-stage liver disease. Levels of IGF-I were not significantly different from those in cystic fibrosis children with normal liver function, and IGFBP-3 levels seemed concordant with the observed IGF-I levels. These results, which appear improved when compared with those in previous cohorts of study patients with end-stage liver disease, may be related to improved nutritional status because of earlier referral with aggressive nutritional management. We conclude that administration of exogenous GH is of limited clinical value in these patients.
We hypothesize that GH resistance that is still apparent, despite reasonable nutritional status, is a primary consequence of liver failure rather than a secondary consequence of end-stage liver disease or malnutrition. Levels of IGF-I may be correlated with nutritional status, despite GH resistance, and early intensive nutritional therapy may improve IGF-I levels to within normal limits, bypassing the apparent GH unresponsiveness in these children with end-stage liver disease. These findings further indicate the potential benefits of adequate nutritional support in these patients (4,6).
Acknowledgment: The authors thank the medical and nursing staff of the Gastroenterology and Transplant Care Units of the Royal Children's Hospital, Brisbane, Australia.
Recombinant human growth hormone (Genotropin) and partial funding for this study were provided by Pharmacia and Upjohn Pty Ltd., New South Wales, Australia.
1. Gartner JC, Zitelli BJ, Malatack JJ et al. Orthotopic liver transplantation in children: Two-year experience with 47 patients.Pediatrics
2. Shepherd RW. The treatment of end stage liver disease in childhood. Aust Pediatr J
3. Shepherd RW, Chin S, Patrick M, et al. Malnutrition in children with chronic liver disease accepted for liver transplantation.J Pediatr Child Health
4. Chin SE, Shepherd RW, Cleghorn GJ, et al. Pre-operative nutritional support in children with end stage liver disease accepted for liver transplantation: An approach to management. J Gastroenterol Hepatol
5. Chin SE, Shepherd RW, Thomas BJ, et al. The nature of malnutrition in children with end stage liver disease awaiting orthotopic liver transplantation. Am J Clin Nutr
6. Chin SE, Shepherd RW, Thomas BJ, et al. Nutritional support in children with liver disease: A randomized crossover trial of branched chain amino acid supplement. Am J Clin Nutr
7. Isaksson OGP, Lindahl A, Nilsson A, Isgaard J. Mechanism of the stimulatory effect of growth hormone on longitudinal bone growth.Endocr Rev
8. Hindmarsh PC, Bridges NA & Brook CDG. Wider indications for treatment with biosynthetic human growth hormone in children.Clin Endocrinol
9. Isaksson OGP, Efen S, Jansson JO. Mode of action of pituitary growth hormone on target cells. Ann Rev Physiol
10. Moller S, Becker U. Insulin like growth factor I and growth hormone in chronic liver disease. Dig Dis
11. Brook CGD. Determination of body composition of children from skinfold measurements. Arch Dis Child
12. Moore FD, Olesen KH, McMurrey JD, Ball MR, Boyden CM.The body cell mass and its supporting environment
. Philadelphia: WB Saunders, 1963.
13. Shepherd RW, Holt TL, Greer RM, Cleghorn GJ, Thomas BJ. Total body potassium in cystic fibrosis. J Pediatr Gastroenterol Nutr
14. Thomson M, Bucolo S, Quirk P, Shepherd RW. Measured v. predicted resting energy expenditure in infants: A need for reappraisal.J Pediatr
15. Thomson MA, Wilmott RW, Wainwright C, Masters B, Francis PJ, Shepherd RW. Resting energy expenditure, pulmonary inflammation and genotype in the early course of cystic fibrosis. J Pediatr
16. Gremse DA, A-Kader HH, Schroeder TJ, Balisteri WF. Assessment of lidocaine metabolite formation as a quantitative liver function test in children. Hepatology
17. Crawford DH, Shepherd RW, Halliday JW, Cooksley GW, Golding SD, Cheng WS, Powell LW. Body composition in nonalcoholic cirrhosis: The effect of disease etiology and severity on nutritional compartments.Gastroenterology
18. Greer RM, Shepherd RW, Cleghorn GJ, Bowling FG, Holt TL. Evaluation of growth and changes in body composition following neonatal diagnosis of cystic fibrosis. J Pediatr Gastroenterol Nutr
19. Potter JM, Hickman PE, Lynch SV, et al. Use of monoethylglycinexylidide as a liver function test in the liver transplant recipient. Transplantation
20. Francis GL, McNeil KA, Wallace JC, Ballard FJ, Owens PC. Sheep insulin like growth factors I and II: Sequences, activities and assays. Endocrinology
21. Bamard R, Quirk P, Waters MJ. Characterization of the growth-hormone binding protein of human serum using a panel of monoclonal antibodies. J Endocrinol
22. Baxter RC, Martin JL, Wood MH. Two immunoreactive binding proteins for insulin like growth factors in human amniotic fluid: Relationship to fetal maturity. J Clin Endocrinol Metab
23. Baxter RC, Martin JL. Radioimmunoassay of growth hormone dependent insulin like growth factor binding protein in human plasma.J Clin Invest
24. Quirk P, Owens P, Moyse K, Chin S, Wall C, Ballard J, Shepherd R. Insulin-like growth factors I and II are reduced in plasma from growth retarded children with chronic liver disease. Growth Regulation
25. Bucuvalas JC, Horn JA, Slusher J, Alfaro MP, Chernausek SD. Growth hormone insensitivity in children with biliary atresia. J Pediatr Gastroenterol Nutr
26. Holly JMP, Smith CP, Dunger DB, et al. Levels of the small insulin-like growth factor binding protein are strongly related to those of insulin in prepubertal and pubertal children but only weakly so after puberty. J Endocrinol
27. Bucuvalas JC, Cutfield W, Horn JA, et al. Resistance to the growth-promoting and metabolic effects of growth hormone in children with chronic liver disease. J Pediatr
28. Clemmons DR, Underwood LE, Dickerson RN, et al. Use of plasma somatomedin-C/insulin-like growth factor I measurements to monitor the response to nutritional repletion in malnourished patients. Am J Clin Nutr
29. Mathews LS, Norstedt G, Palmiter RD: Regulation of insulin-like growth factor I gene expression by growth hormone. Proc Natl Acad Sci
30. Ketelslegers JM, Maiter D, Maes M, Underwood LE, Thissen JP. Nutritional regulation of insulin-like growth factor I.Metabolism
1995;44(10 Suppl 4):50-7.
31. Hynes MA, Van Wyk JJ, Brooks PJ, D'Ercole AJ, Jansen M, Lund PK. Growth hormone dependence of somatomedin-C/insulin like growth factor I and insulin like growth factor II messenger ribonucleic acids.Mol Endocrinol
32. Owens PC, Johnson RJ, Campbell RG, Ballard FJ. Growth hormone increases insulin-like growth factor-I (IGF-I) and decreases IGF-II in plasma of growing pigs. J Endocrinol
33. Donaghy A, Ross R, Gimson A, Hughes SC, Holly J, Williams R. Growth hormone, insulin-like growth factor-I and insulin like growth factor binding proteins 1 and 3 in chronic liver disease.Hepatology
34. Hattori N, Kurahachi H, Ikekubo K, et al. Serum GH-binding protein, insulin-like growth factor-I, and growth hormone in patients with liver cirrhosis. Metabolism
35. Baruch Y, Amit T, Hertz P, Enat R, Youdim MB, Hochberg Z. Decreased serum growth hormone-binding protein in patients with liver cirrhosis. J Clin Endocrinol Metab
36. Tiong TS, Herington AC. Tissue distribution, characterization and regulation of messenger ribonucleic acid for growth hormone receptor and serum binding protein in the rat.Endocrinology
37. Herington AC, Tiong TS, Ymer SI. Serum binding proteins for growth hormone: Origins, regulation of gene expression and possible roles. Acta Paediatr Scand
38. Brinkman A, Groffen C, Kortleve DJ, et al. Isolation and characterisation of a cDNA encoding the low molecular weight insulin like growth factor finding protein (IGFBP-1). EMBO J
39. Julkunen M, Koistinen R, Aalto-Setala K, et al. Primary structure of human insulin-like growth factor binding protein 12 and tissue-specific expression of its mRNA. FEBS Lett