production in healthy man Introduction

Transcription

1 Journal of Internal Medicine 1997; 241: Effects of short-term insulin-like growth factor-i or growth hormone treatment on bone turnover, renal phosphate reabsorption and 1,25 dihydroxyvitamin production in healthy man T. BIANDA, M. A. HUSSAIN, Y. GLATZ, R. BOUILLON,a E. R. FROESCH & C. SCHMID From the Department of Internal Medicine, Division of Endocrinology and Metabolism, University Hospital, Zu rich, Switzerland; and alaboratorium voor Experimentele Geneeskunde en Endocrinologie, Katholieke Universitet Leuven, Leuven, Belgium Abstract. Bianda T, Hussain MA, Glatz Y, Bouillon R, Froesch ER, Schmid C (University Hospital, Zu rich, Switzerland; and Katholieke Universitet, Leuven, Belgium). Effects of short-term insulin-like growth factor-i or growth hormone treatment on bone turnover, renal phosphate reabsorption and 1,25 dihydroxyvitamin production in healthy man. J Intern Med 1997; 241: Objectives. To find out whether insulin-like growth factor-i (IGF-I) mimics the stimulatory effects of growth hormone (GH) on bone turnover and renal tubular phosphate reabsorption. Design. Randomized, crossover study. Setting. University Hospital, Zu rich, Switzerland. Subjects. Seven young healthy male subjects. Interventions. Each subject was studied three times at 2-week intervals, treated with saline 0.9% (S), IGF-I [8 µg kg h ] by a continuous subcutaneous infusion and finally with GH (6 U. twice daily s.c.) for 5 days. Main outcome measures. 36 h after the start of treatment, IGF-I, biochemical markers of bone turnover, calcium, calcium regulating hormones, kidney function and phosphate reabsorption were measured in serum and in 2 h urine in fasting state. Results. Serum levels of IGF-I were (S), (IGF-I) (P 0.02) and nmol L (GH) (P 0.02), respectively. Serum osteocalcin and carboxyterminal propeptide of type I collagen (PICP) as well as the urinary deoxypyridinoline creatinine and the calcium creatinine ratios were all significantly higher after IGF-I (P 0.02) or GH (P 0.02) than after saline treatment. PTH levels did not change in response to treatment. Total albumin-corrected calcium increased only after GH treatment (P 0.05). The free calcitriol index rose from (S) to (IGF-I) (P 0.03) and (GH), respectively. Serum phosphate and maximal tubular reabsorption divided by glomerular filtration rate (TmP GFR) were significantly raised by GH (P 0.03) but not by IGF-I as compared to saline 0.9%. Conclusions. (i) Similar to GH, IGF-I rapidly activates bone turnover. (ii) IGF-I does not mimic the effect of GH on renal phosphate reabsorption in spite of comparable effects on renal blood flow and glomerular filtration rate. (iii) IGF-I increases free calcitriol index in face of unchanged serum levels of calcium, phosphate and PTH, consistent with a direct stimulatory effect on 25-OHD-1a-hydroxylase. Keywords: 1,25-dihydroxyvitamin, bone turnover, growth hormone, insulin-like growth factor-i, phosphate reabsorption. Introduction Growth hormone (GH) stimulates longitudinal bone growth in childhood and adolescence. GH enhances the tubular reabsorption of phosphate in rats and in humans [1] as well as the intestinal calcium absorption [2, 3] during growth. Insulin-like growth factor-i (IGF-I) stimulates longitudinal (endochondral) bone formation and increases type I procollagen mrna levels in bone [4, 5] as well as the 1997 Blackwell Science Ltd 143

2 144 T. BIANDA et al. renal tubular transport of phosphate and the synthesis of 1,25-(OH) of hypophysectomized rats [6]. GH and IGF-I continue to exert an important role after cessation of longitudinal growth in the remodelling of adult bone [7]. Both GH [1] and IGF-I [8] increase glomerular filtration rate and renal plasma flow in normal subjects and are also involved in the control of activity of 25-OHD-1a-hydroxylase. GH increases 1,25-(OH) levels and the renal threshold for phosphate excretion [9, 10], independent of PTH [11]. However, it remains controversial whether GH and IGF-I have similar effects on renal 1,25-(OH) production and renal phosphate reabsorption in humans. In order to address the question whether IGF-I mimics and possibly mediates the effects of GH on bone remodelling, renal phosphate and calcium handling, and 1,25-(OH) production, we have measured serum levels of osteocalcin, carboxyterminal propeptide of type I procollagen (PICP), pyridinoline cross-linked carboxyterminal telopeptide of type I collagen (ICTP) and of calciotropic hormones as well as urinary excretion of deoxypyridinoline, calcium and phosphate in healthy male volunteers treated with saline, IGF-I or GH for 36 h. Subjects, materials and methods Subjects Seven healthy, normal weight male volunteers (age 27 4 years; body mass index kg m ) were studied after obtaining written informed consent. None had any evidence of somatic or mental illness as assessed by history, clinical and routine laboratory examination and chest radiograph. The study protocol had been approved by the ethics committee of the University Hospital of Zu rich. Experimental design (Fig. 1) The study consisted of three periods of 5 days each during which the subjects received continuous subcutaneous infusion of 8 µg kg m rhigf-i (Ciba- Geigy AG, Basel, Switzerland) or 0.9% saline via a portable mini-pump (MRS-1, Disetronic AG, Burgdorf, Switzerland) in a crossover, randomized fashion and finally subcutaneous injections of GH (6 U twice daily). The subjects were always studied 2 weeks after the end of the first treatment period. A sucrosefree diet of 30 kcal kg day (50% carbohydrates, 30% lipids, 20% proteins; 30% of calories at each main meal at 08.00, and hours, and 10% as bedtime snack at hours) was started 5 days before the study and was maintained throughout the study period. Light physical exercise for not more than 30 min day was allowed during this time. Immediately before the start of each treatment blood samples were drawn to check baseline serum values for IGF-I, glucose, calcium, phosphate, creatinine, osteocalcin, PICP and ICTP. After 36 h of treatment and a 10-h overnight fast, blood samples were drawn at hours for determinations of total serum IGF-I, glucose, calcium, albumin, phosphate, creatinine, osteocalcin, PICP, ICTP, PTH, 1,25-(OH) and vitamin D binding protein (DBP); 2 h fasting urinary calcium, deoxypyridinoline, phosphate and creatinine excretion were also determined. On day 3, 36 h after the start of treatment, an intravenous GH pulse was given and its acute effects on systemic and forearm (muscle) metabolism were studied [12]. On day 4, a 24 h study on lipid metabolism was performed (report in preparation). Analytical methods Plasma glucose was measured in duplicate immediately after blood sampling using an automated glucose-oxydase method (Glucose Analyser 2, Beckman Instruments Inc., Fullerton CA). Total IGF-I levels in serum were measured by RIA according to a modification of a previously described method [13, 14]. Serum total 1,25-(OH) was determined by RIA after extraction and purification by high performance liquid chromatography, as described previously [15]. Serum DBP was determined using radial immunodiffusion [16] and the free calcitriol index calculated as the molar ratio of the measured concentrations of total 1,25-(OH) and DBP [15]. Intact PTH and osteocalcin levels in serum were measured using two-site immunoradiometric assays (IRMA; Nichols Institute, San Juan Capistrano, CA). Serum levels of PICP and ICTP were measured with recently developed RIA kits (Orion Diagnostica, Espoo, Finland) [17, 18]. The lower detection limit of the tests is 1.2 µg L for PICP and 0.34 µg L for ICTP, respectively. Serum calcium, phosphate, albumin and crea-

3 EFFECTS OF IGF-I OR GH ON BONE TURNOVER 145 Fig. 1 Schematic diagram of study protocol. tinine were analysed according to standard laboratory methods. Serum calcium was corrected for individual variations in serum albumin using the formula: corrected serum calcium [mmol L ] measured serum calcium [mmol L ] 0.02 (40 measured albumin [g L ]). Urinary deoxypyridinoline was measured using a competitive enzyme immunoassay (Pyrilinks-D kit; Metra Biosystem Inc., Mountain View, CA) and values were expressed relative to creatinine excretion and given as nmol per mmol. Urinary calcium, phosphate and creatinine were analysed according to standard laboratory methods. The renal tubular maximal phosphate reabsorption capacity per litre glomerular filtrate (TmP GFR) was calculated according to the nomogram of Bijvoet [19]. Fasting urinary calcium excretion (U Ca ), expressed as: U Ca V GFR (mmol L ) U Ca P Cr U Cr, was calculated from 2 h fasting urine. The ratios of deoxypyridinoline creatinine (dpyr creatinine) and calcium creatinine were also calculated from 2 h fasting urine. Statistics Results are expressed as mean SD. The results were analysed using the two-tailed Wilcoxon s rank-sum test for paired difference [20]. A P-value of 0.05 was considered statistically significant. Results Clinical observations All subjects tolerated the treatment very well. No episode of hypoglycaemia and no significant changes in blood pressure were observed. In no case was clinical oedema encountered, and all volunteers completed the study. Body weight and fat free mass did not change during the study [12]. Fig. 2 Fasting venous blood levels of osteocalcin and PICP in 7 healthy subjects after 36 h of treatment with saline, IGF-I and GH. *P **P Baseline values Baseline values of serum IGF-I were comparable at the start of each treatment period, (S) vs (IGF-I) vs (GH) nmol L. There were no significant differences in the baseline values for glucose, phosphate, albumin, albumin-corrected calcium, urea, creatinine, phosphate, osteocalcin, PICP, ICTP in each treatment period (not shown).

4 146 T. BIANDA et al. Serum ICTP was not significantly different after treatment with saline ( ), IGF-I ( ) or GH ( µg L ), respectively. The urinary dpyr creatinine ratio increased from (S) to (IGF-I) (P 0.02) and nmol mmol (GH), respectively. Likewise, urinary calcium creatinine ratio rose from (S) to (IGF-I) (P 0.02) and mmol mmol (GH) (P 0.02), respectively, and U Ca V GFR from (S) to (IGF-I) (P 0.02) and mmol L (GH) (P 0.02), respectively. Fig. 3 Fasting urinary deoxypyridinoline creatinine ratio and calcium creatinine ratio in 7 healthy subjects after 36 h of treatment with saline, IGF-I and GH. *P Laboratory values after 36 h Serum IGF-I and glucose. Serum IGF-I level rose significantly from (S) to (IGF-I) (P 0.02) and nmol L (GH) (P 0.02), respectively. Glucose fell from (S) to (IGF-I) (P 0.03) and rose to mmol L (GH) (P 0.02), respectively. Biochemical markers of bone turnover. The effects of either treatment on biochemical markers of bone formation and resorption are shown in Figs 2 and 3. A significant increase in serum osteocalcin levels was observed in both IGF-I ( µg L ) (P 0.02) and GH treatment ( µg L ) (P 0.02) as compared to saline ( µg L ). Likewise, serum PICP rose from (S) to (IGF-I) (P 0.03) and µg L (GH) (P 0.03). Serum calcium and calcium regulating hormones. Serum albumin remained unchanged during IGF-I ( g L ) as compared with saline ( ) but rose significantly during GH treatment ( g L ) (P 0.02). Total calcium increased significantly during GH, even when corrected for albumin, from mmol L (P 0.05), whereas no change was noted during IGF-I treatment ( mmol L ). PTH levels remained unchanged during either treatment. Total 1,25-(OH) tended to increase, from (S) to (IGF-I) (P 0.06) and ng L (GH) (P 0.09), respectively. DBP remained unchanged during IGF-I and increased from (S) to mg L during the GH treatment (P 0.05). Consequently, the free calcitriol index rose from (S) to (IGF- I) (P 0.03) and was not significantly changed during GH treatment (Fig. 4). Kidney function and phosphate reabsorption. 36 h after beginning treatment serum creatinine levels were decreased only during IGF-I from (S) to (IGF-I) (P 0.02), whereas they remained unchanged during GH, µmol L. Serum urea levels fell in both treatments, from (S) to (IGF-I) (P 0.02) and to mmol L (GH) (P 0.02), respectively. The urinary urea excretion decreased from (S) to (IGF-I) (P 0.02) and mmol 2 h (GH) (P 0.02), respectively. Serum phosphate increased from (S) to mmol L during GH (P 0.02), but was not significantly changed during IGF-I treatment

5 EFFECTS OF IGF-I OR GH ON BONE TURNOVER 147 Fig. 4 Fasting venous blood levels of total albumin-corrected calcium, PTH, phosphate, total 1,25-(OH), free calcitriol index and TmP GFR in 7 healthy subjects after 36 h of treatment with saline, IGF-I and GH. *P **P P ( mmol L ). In agreement with the increase in serum phosphate during GH treatment, we found an increase of TmP GFR only after GH ( mmol L ) (P 0.03) as compared to saline ( mmol L ), whereas TmP GFR did not change after IGF-I treatment ( mmol L ) (Fig. 4). Discussion The present study was undertaken to compare the acute effects of IGF-I and GH on bone metabolism, renal calcium and phosphate handling and production of 1,25-(OH), mostly to estimate whether any of the GH effects may be mediated by IGF-I. It has been found that IGF-I opposes rather than mediates the effects of GH on insulin resistance [13]. Therefore, the current study was designed to evaluate whether IGF-I treatment alters the metabolic response to a GH-pulse, as reported separately [12]. Numerous studies have demonstrated that GH treatment is associated with increased bone remodelling in GH-deficient adults [7, 21 24]. Whether the action of GH on bone results from the hormonal effects of circulating IGF-I or from autocrine paracrine effects of IGF-I or from separate IGF-I-independent GH effects is unknown. S. C. administration of a pharmacological dose of IGF-I for 7 days activated bone turnover in a male patient with idiopathic osteoporosis [25] and stimulatory effects of IGF-I on bone turnover were also demonstrated in elderly [26, 27] and in fasting young women [28]. In a recent cross-over study comparing the effects of IGF-I and those of GH in 12 men with idiopathic osteoporosis, Johansson et al. showed that IGF-I and GH similarly stimulated bone turnover [29]. As assessed by biochemical markers of bone resorption as well as of bone formation, we found that treatment with rhigf-i rapidly increased bone

6 148 T. BIANDA et al. turnover in young healthy men. Our subjects were cross-over randomized with a 2 week interval between the treatment periods (Fig. 1). Because serum values of osteocalcin and PICP started from identical baseline levels 2 weeks later, we conclude that the stimulatory effect of IGF-I does not persist for 2 weeks. These findings indicate that the stimulation of bone formation by short-term IGF-I application may be due to an increase in the activity and not necessarily in the number of osteoblasts. The stimulatory effects of short-term IGF-I application on parameters of bone formation disappeared more readily in our young healthy adults than in an osteoporotic male patient [25]; in a subsequent study, Johansson et al. found that the effects of IGF- I were less lasting than those of GH [29]. Ebeling et al. and more recently Ghiron et al. did not address the question for how long the stimulatory effects of IGF- I on bone turnover persisted in elderly women [26, 27]. After 36 h of treatment with IGF-I or GH, urinary deoxypyridinoline excretion and calcium excretion were significantly increased, suggesting that both hormones rapidly activate, directly or indirectly (via raised serum level of 1,25-(OH) or other factors) bone resorption in young healthy humans. ICTP did not significantly change during either treatment. Therefore, compared to urinary deoxypyridinoline excretion serum ICTP levels appear to be of limited sensitivity, as previously reported from a study of bone metabolism in postmenopausal osteoporosis [30]. It has been suggested that the effect of GH on bone remodelling is mediated at least partially by IGF-I. Our results are consistent with this hypothesis, as IGF-I given systematically leads to rapid activation of bone turnover, i.e. is active in an endocrine fashion. It is important to note that IGF-I treatment suppresses GH secretion in healthy humans; therefore, a possible direct GH effect on bone is markedly reduced during IGF-I treatment. To address the question more directly, GH and IGF-I should be compared with regard to their effects on markers of bone turnover in patients lacking of IGF-I and GH. In GH-deficient patients the IGF-I effects on bone can be tested without the interference of GH suppression. Further investigations with longer treatment periods with IGF-I and GH and long-term control of indices for bone resorption and formation as well as bone mineral density are needed to determine whether the described effects on bone turnover are sustained and whether IGF-I is an effective treatment for osteoporosis. In accordance with previous studies [2, 31] we found that renal phosphate reabsorption markedly increased in response to GH treatment, whereas such an effect was not observed during IGF-I treatment. Circulating IGF-I levels were significantly elevated during both treatment periods, even more during IGF-I treatment. Johansson et al. found a significant increase in serum phosphate in men with idiopathic osteoporosis after 7 days of treatment with GH but not after treatment with IGF-I [29]. Similarly, Ghiron et al. did not find an increase in serum phosphate in elderly women treated with IGF-I at 15 µg kg b.i.d. or at 5 µg kg b.i.d. These authors did not measure TmP GFR in their study, but speculated that GH could directly regulate renal phosphorus handling [27]. Our findings also suggest that GH raises TmP GFR independently of circulating IGF-I. In contrast, IGF-I mimics the effects of GH on glomerular filtration rate [8]. Possible explanations for this finding include a direct effect of GH on tubular transport of phosphate, or an enhancement of local production of IGF-I acting in a paracrine fashion, or GH effects via other local or systemic regulators. Acromegaly is a disease often associated with increased plasma phosphate and TmP GFR. IGF-I has been an attractive candidate as a circulating factor responsible for the increase in TmP GFR since it stimulates renal phosphate reabsorption in rats [6], but our data in healthy adult humans obviously contrast with the findings in growing rats. The GH-IGF-I axis is also involved in the renal production of 1,25-(OH). Whether IGF-I mediates the stimulatory effects of GH on renal 1,25-(OH) production is controversial. A direct stimulatory effect of IGF-I on the activity of the renal 25-OHD-1ahydroxylase has been suggested [32, 33]. In healthy human adults 36 h of treatment with both IGF-I and GH appeared to increase total 1,25-(OH) levels in serum. During IGF-I treatment, PTH, phosphate and calcium levels remained unchanged. These results are consistent with a direct stimulatory effect of IGF- I on 25-OHD-1a-hydroxylase. Similarly to previous studies we found that GH treatment for 36 h induced a rise of serum phosphate [31] and calcium [34], whereas PTH remained unchanged. Both serum phosphate and calcium elevation inhibit 25-OHD- 1a-hydroxylase activity, and thereby could mask an IGF-I-mediated stimulation of 25-OHD-1a-hydroxyl-

7 EFFECTS OF IGF-I OR GH ON BONE TURNOVER 149 ase during GH treatment. Moreover, GH treatment stimulates bone formation and increases calcium demands. Therefore, numerous factors apart from IGF-I could explain the elevated levels of serum 1,25- (OH) that we found after 36 h of treatment with GH in spite of the raised serum phosphate. Our results are compatible with the hypothesis that IGF- I mediates the stimulatory effect of GH on 25-OHD- 1a-hydroxylase [32, 33]. A significant increase of the free calcitriol index was found only during treatment with IGF-I, which did not influence the concentration of serum DBP. GH treatment did not significantly increase the free calcitriol index possibly because serum phosphate and DBP concentrations were elevated. The rise of serum albumin during GH treatment is in line with results of previous studies showing that GH stimulates hepatic production of albumin [35]. In summary, IGF-I quickly stimulates bone remodelling and appears to mediate the renal effects of GH on 1,25-(OH) production independently of PTH in young healthy adults. However, increased levels of circulating IGF-I under the influence of GH cannot account for increased renal phosphate reabsorption. This is a GH-related effect, independent of circulating IGF-I. Acknowledgements This work was supported by the Swiss National Science Foundation (Grant no ) and the Belgian National Foundation for Scientific Research (Grant no ). References 1 Corvilain J, Abramov M. Some effects of human growth hormone on renal hemodynamics and tubular Pi transport in man. J Clin Invest 1962; 41: Chipman JJ, Zerwekh J, Nicar M, Marks J, Pak CYC. Effect of growth hormone administration: reciprocal changes in serum; 1: 25 dihydroxyvitamin D and intestinal calcium absorption. J Clin Endocrinol Metab 1980; 51: Yeh JK, Aloia JF. Effect of hypophysectomy and 1:25- dihydroxyvitamin D on duodenal calcium absorption. Endocrinology 1984; 114: Guler HP, Zapf J, Scheiwiler E, Froesch ER. Recombinant human insulin-like growth factor I stimulates growth and has distinct effects on organ size in hypophysectomized rats. Proc Natl Acad Sci 1988; 85: Schmid C, Guler HP, Rowe D, Froesch ER. Insulin-like growth factor I regulates type I procollagen messenger ribonucleic acid steady state levels in bone of rats. Endocrinology 1989; 125: Caverzasio J, Montessuit C, Bonjour JP. Stimulatory effect of Insulin-like growth factor I on renal Pi transport and plasma; 1:25-dihydroxyvitamin. Endocrinology 1990; 127: Parfitt AM. Growth hormone and adult bone remodeling. Clinical Endocrinol 1991; 35: Guler HP, Eckardt KU, Zapf J, Bauer C, Froesch ER. Insulin-like growth factor I increases glomerular filtration rate and renal plasma flow. Acta Endocrinol (Copenh.) 1989; 121: Brown DJ, Spanos E, MacIntyre I. Role of pituitary hormones in regulating vitamin D metabolism in man. Br Med J 1980; 280: Lund B, Eskildsen PC, Lund B, Norman AW, Soerensen OH. Calcium and vitamin D metabolism in acromegaly. Acta Endocrinol (Copenh.) 1981; 96: Lieberman SA, Holloway L, Marcus R, Hoffman AR. Interactions of growth hormone and parathyroid hormone in renal phosphate, calcium, and calcitriol metabolism and bone remodeling in postmenopausal women. J Bone Miner Res 1994; 11: Bianda T, Hussain MA, Keller A, Glatz Y, Schmitz O, Christiansen JS et al. Insulin-like growth factor-i in man enhances lipid mobilization and oxidation induced by a growth hormone pulse. Diabetologia 1996; 39: Hussain MA, Schmitz O, Mengel A, Glatz Y, Christiansen JS, Zapf J et al. Comparison of the effects of growth hormone and insulin-like growth factor-i on substrate oxidation and on insulin sensitivity in growth hormone deficient humans. J Clin Invest 1994; 94: Zapf J, Walter H, Froesch ER. Radioimmunological determination of insulin-like growth factor-i and II in normal subjects and in patients with growth disorders and extrapancreatic tumor hypoglycemia. J Clin Invest 1981; 86: Bouillon R, Van Assche FA, Van Baelen H, Heyns W, De Moor P. Influence of the vitamin D binding protein on the serum concentration of; 1: 25-dihydroxyvitamin. Significance of the free; 1: 25-dihydroxyvitamin concentration. J Clin Invest 1981; 67: Bouillon R, Van Assche FA, De Moor P. The measurement of the vitamin D binding protein in human serum. J Clin Endocrinol Metab 1977; 45: Mellko J, Niemi S, Risteli L, Risteli J. Radioimmunoassay of the carboxyterminal propeptide of human type I procollagen. Clin Chem 1990; 36: Risteli J, Elomaa I, Niemi S, Novamo A, Risteli L. Radioimmunoassay for the pyridinoline cross-linked carboxy-terminal telopeptide of type I collagen: a new serum marker of bone collagen degradation. Clin Chem 1993; 39: Walton RJ, Bijvoet OLM. Nomogram for derivation of renal threshold phosphate concentration. Lancet 1975; ii: Wilcoxon F. Individual comparisons by ranking methods. Biomed Bull 1945; 1: Schlemmer A, Johansen JS, Pedersen SA, Jo rgensen JO, Hassager C, Christiansen C. The effect of growth hormone therapy on urinary pyridinoline cross-links in GH-deficient adults. Clin Endocrinol 1991; 35: Johansen JS, Pedersen SA, Jorgensen JOL, Riis BJ, Christiansen C, Christiansen JS et al. Effects of growth hormone on plasma bone Gla protein in GH-deficient adults. J Clin Endocrinol Metab 1990; 70:

8 150 T. BIANDA et al. 23 Binnerts A, Swart GR, Wilson PJH, Hoogerbrugge N, Pols HAP, Birkenhager JC et al The effect of growth hormone administration in growth hormone deficient adults on bone, protein, carbohydrate and lipid homeostasis as well as on body composition. Clin Endocrinol 1992; 37: Balducci R, Toscano F, Pasquino, AM Mangiatini A, Municchi G, Armenise P et al. Bone turnover and bone mineral density in young adult patients with panhypopituitarism before and after long-term growth hormone therapy. Eur J Endocrinol 1995; 132: Johansson AG, Lindh E, Ljunghall S. Insulin-like growth factor I stimulates bone turnover in osteoporosis. Lancet 1992; 339: Ebeling PR, Jones JD, O Fallon WM, Janes CH, Riggs B. Short term effects of recombinant human insulin-like growth factor- I on bone turnover in normal women. J Clin Endocrinol Metab 1993; 77: Ghiron L, Thompson JL, Holloway L, Hintz RL, Butterfield GE, Hoffman AR et al. Effects of recombinant Insulin-like growth factor-i and growth hormone on bone turnover in elderly women. J Bone Min Res 1995; 12: Grinspoon SK, Baum HBA, Peterson S, Klibanski A. Effects of rhigf-i administration on bone turnover during short-term fasting. J Clin Invest 1995; 96: Johansson AG, Lindh E, Blum WF, Kollerup G, So rensen OH, Ljunghall S. Effects of growth hormone and insulin-like growth factor-i in men with idiopathic osteoporosis. J Clin Endocrinol Metab 1996; 81: Hassager C, Jensen LT, Podenphant J, Thomsen K, Christiansen C. The carboxy-terminal pyridinoline cross-linked telopeptide of type I collagen in serum as a marker of bone resorption: the effect of nandrolone decanoate and hormone replacement therapy. Calc Tissue Int 1994; 54: Marcus R, Butterfield G, Holloway L, Gilliland L, Baylink DJ, Hintz RL et al. Effects of short term administration of recombinant human growth hormone to elderly people. J Clin Endocrinol Metab 1990; 70: Condamine L, Vztovsnik F, Friedlander G, Menaa C, Garabedian M. Local action of phosphate depletion and insulin-like growth factor-i on in vitro production of; 1: 25-dihydroxyvitamin D by cultured mammalian kidney cells. J Clin Invest 1994; 94: Nesbitt T, Drezner MK. Insulin-like growth factor-i regulation of renal 25-hydroxyvitamin D-1-hydroxylase activity. Endocrinology 1993; 132: Brixen K, Nielsen HK, Bouillon R, Flyvbjerg A, Mosekilde L. Effects of short term growth hormone treatment on PTH, calcitriol, thyroid hormones, insulin and glucagon. Acta Endocrinol (Copenh.) 1992; 127: Schwander JC, Hauri C, Zapf J, Froesch ER. Synthesis and secretion of Insulin-like growth factor and its binding protein by the perfused rat liver: dependence on growth hormone status. Endocrinology 1983; 113: Received 14 August 1996; accepted 10 September Correspondence: Dr Tarcisio Bianda, Division of Endocrinology and Metabolism, Department of Internal Medicine, University Hospital, CH-8091 Zu rich, Switzerland.