THE ONSET of puberty in healthy children is accompanied

Transcription

1 X/97/$03.00/0 Vol. 82, No. 10 Journal of Clinical Endocrinology and Metabolism Printed in U.S.A. Copyright 1997 by The Endocrine Society Decreased Insulin Sensitivity and Compensatory Hyperinsulinemia after Hormone Treatment in Children with Short Stature* RUBINA A. HEPTULLA, SUSAN D. BOULWARE, SONIA CAPRIO, DAVID SILVER, ROBERT S. SHERWIN, AND WILLIAM V. TAMBORLANE Department of Pediatrics and the Children s Clinical Research Center, Yale University School of Medicine, New Haven, Connecticut ABSTRACT To assess the effects of GH treatment on carbohydrate and protein metabolism, we studied eight patients with short stature before and after the commencement of GH treatment. The hyperglycemic clamp procedure was employed to produce a hyperglycemic stimulus of 50 mg/dl above fasting levels for 120 min. These patients were then treated with 0.3 mg/kg week GH for 6 months, after which they were restudied. Patients were compared to eight healthy control children matched for age, sex, and Tanner stage. Fasting plasma glucose did not change significantly, but fasting plasma insulin levels were higher after GH therapy (P 0.005). Despite identical glucose increments during the glucose clamp procedure, both first and second phase insulin responses were markedly greater after instituting GH treatment. Even in the face of higher mean plasma insulin levels after GH Received April 4, Revision received June 16, Accepted June 26, Address all correspondence and requests for reprints to: Rubina A. Heptulla, M.D., Department of Pediatrics, Division of Endocrinology, Yale University School of Medicine, 333 Cedar Street, New Haven, Connecticut * This work was supported by NIH Grants HD-30671, RR-06022, and RR Recipient of a postdoctoral fellowship award from the Juvenile Diabetes Foundation International. treatment, the rate of insulin-stimulated glucose metabolism did not differ during the last 60 min of both studies. Hence, the rate of insulin-stimulated glucose metabolism/mean plasma insulin ratio (an index of insulin sensitivity) was sharply reduced after GH treatment (P 0.01). During the clamp, the fall in circulating branched chain amino acid levels was significantly greater after GH therapy (P 0.02). We conclude that glucose-stimulated insulin responses are increased in short children treated with GH and that such hyperinsulinemic responses compensate for reductions in insulin sensitivity. The compensatory hyperinsulinemic responses induced by GH therapy may serve a beneficial role by augmenting insulin s anabolic effects on protein metabolism. (J Clin Endocrinol Metab 82: , 1997) THE ONSET of puberty in healthy children is accompanied by hormonal changes that lead to the development of secondary sexual characteristics and rapid acceleration in linear growth. Increases in circulating concentrations of sex steroids, GH, and insulin-like growth factor I (IGF-I) are central to normal pubertal development. Concomitant to these dramatic changes in the hormonal milieu are agerelated changes in both insulin action and insulin secretion that occur during adolescence. Lean healthy adolescents have decreased insulin sensitivity and increased insulin secretion compared to lean healthy prepubertal children and young adults (1, 2). However, the hormonal changes responsible for these puberty-associated alterations in insulin action and insulin secretion have not been established. The pubertal rise in gonadal and adrenal steroids has been suggested to be a contributing factor, but such steroid levels remain elevated in normal adults despite increased insulin sensitivity (3, 4). Studies from our laboratory have shown that insulin resistance and hypersecretion of insulin in healthy adolescents are correlated with elevated mean 24-h GH and IGF-I levels (1, 5), suggesting that the rise in circulating GH concentrations that occurs during the pubertal growth spurt may mediate changes in insulin sensitivity during adolescence. Moreover, there is evidence that the insulin resistance that occurs during puberty may be greater for glucose than for amino acid metabolism (5). We have hypothesized that selective insulin resistance and compensatory hyperinsulinemia induced by the pubertal rise in GH serve to amplify insulin s effects on amino acid metabolism and to facilitate protein anabolism during the adolescent growth spurt. If this hypothesis is correct, a similar sequence of metabolic events should occur in children before or during puberty who are treated with exogenous GH for short stature. The present study was undertaken to examine this question. Subjects Subjects and Methods The clinical characteristics of the study populations are shown in Table 1. The patients included four males and four females (aged 12 1 yr) who were about to begin treatment with exogenous GH. Five patients had idiopathic GH deficiency, two patients had nondeficient short stature, and one patient had postsurgical GH deficiency. Five of the eight patients were prepubertal. Bone age was delayed in all patients except the patient with postsurgical GH deficiency. Controls consisted of eight healthy, nonobese children who were age, sex, and Tanner stage matched to the study subjects. All of the children and their parents gave written informed consent before taking part in the study, which was approved by the human investigation committee of the Yale University School of Medicine. Hyperglycemic clamp procedure Each patient had a hyperglycemic clamp study performed twice: before and approximately 6 months after the start of GH treatment (

2 EFFECTS OF GH ON CARBOHYDRATE AND PROTEIN METABOLISM 3235 TABLE 1. Clinical features of subjects and controls Subject no. mg/kg wk in three to six divided doses); controls were studied once. Hyperglycemic clamps were performed after the subjects had fasted for h overnight at the Children s Clinical Research Center at Yale- New Haven Children s Hospital. During GH treatment, each patient received the usual dose of GH on the evening before the study. Physical examination, including height, weight, blood pressure, and assessment of pubertal development, was performed on the morning of the study. Two iv catheters were then inserted: one in an antecubital vein for administration of exogenous glucose (20%) and the other in a vein of the dorsal part of the contralateral hand for blood sampling. The hand for blood sampling was kept in a heated box (60 65 C) to arterialize venous blood samples (6). After a rest period of min, baseline fasting samples were obtained for the measurement of glucose, insulin, and IGF-I. The hyperglycemic clamp technique has been described in detail previously (7). With this procedure, plasma glucose was rapidly raised by 50 mg/dl above fasting values by infusing, in a decreasing logarithmic manner, a priming dose of exogenous glucose to achieve the desired hyperglycemic plateau quickly. Subsequently, plasma glucose (measured at 5-min intervals) was kept constant at this hyperglycemic level for 120 min by appropriate adjustment of the variable rate 20% glucose infusion. Blood samples were obtained 2, 4, 6, 8, and 10 min after the start of the glucose infusion and every 20 min thereafter for measurement of plasma insulin and C peptide. Branched chain amino acids were collected at 30 min before and 0, 60, 90, and 120 min after the start of the infusion. Urine was collected at the beginning and at the end of the procedure for determination of glucose content; all urine samples were free of glucose. Analyses Plasma glucose levels were measured by the glucose oxidase method with a Beckman glucose analyzer (Beckman Instruments, Brea, CA). Plasma insulin, C peptide, and GH levels were measured by commercially available, double antibody RIA kits (8, 9). Plasma IGF-I levels were measured by the Nichols Institute (San Juan Capistrano, CA). Glycosylated hemoglobin was measured chromatographically with a microcolumn method (Isolab, Akron, OH; nondiabetic range, %). Plasma branched chain amino acids were measured by ion exchange chromatography (10 13). Calculations Age (yr) Sex Tanner stage Ht SD score Bone age delay (yr) a During the hyperglycemic clamp procedure, plasma insulin and C peptide responses were biphasic; first phase (0 10 min) and second phase ( min) responses were calculated as the mean hormone concentration during the respective time periods. The rate of glucose metabolism (M) during the hyperglycemic clamp procedure (expressed in milligrams per m 2 body surface area/min) was calculated at 20-min intervals according to the equation M INF SC, where M is the glucose metabolism rate, INF is the glucose infusion rate, and SC is the correction for changes in the glucose space (7). For determination of the M/I ratio (an index of insulin sensitivity), the rate of glucose metabolism during the last 60 min of each clamp study (M) was divided by the mean plasma insulin level during this time period (I). Estradiol (pg/ml) Pre TABLE 2. Clinical and biochemical effects of GH therapy Pretreatment Posttreatment Growth velocity (cm/yr) a IGF-I (ng/ml) a Fasting glucose (mg/dl) Fasting insulin (pmol/l) b Fasting C peptide (pmol/l) HgA 1 (%) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) a P b P Testosterone (ng/ml) Estradiol (pg/ml) Post Testosterone (ng/ml) Diagnosis b 1 8 F IGHI 2 14 M NDSS 3 16 M IGHI 4 11 F NDSS 5 8 F IGHI 6 15 M IGHI 7 15 M IGHI 8 9 F PSGHD Controls M/4 F 1 3 a Bone age minus chronological age. b Diagnostic codes: IGHI, idiopathic GH insufficiency; PSGHD, postsurgical GH deficiency; NDSS, non-gh deficiency with short stature. Descriptive and inferential statistics were performed using Systat version 5.1 for Windows (SPSS, Chicago, IL). All data are presented as the mean sem. Patients with and without GH insufficiency responded similarly to treatment with respect to clinical and metabolic changes. Consequently, data for all eight patients were combined for analysis. Multiple group comparisons were made using ANOVA and repeated measures ANOVA. Dunnett s procedure for multiple comparisons was used post-hoc to localize effects. Differences were considered significant at the 0.05 level. The two-tailed paired t test was applied for paired comparisons in the patients before and after treatment. Results Table 2 illustrates clinical changes in the patients after GH therapy. After approximately 6 months of treatment, there was a 3-fold increase in IGF-I (P 0.002) and a 2-fold increase in growth velocity (P 0.002). Although there were no significant change in fasting plasma glucose or hemoglobin A 1 levels, fasting insulin concentrations increased significantly after GH treatment (P 0.005). Despite the elevation in basal insulin levels, systolic and diastolic blood pressures were unaffected by treatment. Responses during the hyperglycemic clamp study in the patients before and after treatment and in controls are shown in Fig. 1. As illustrated, plasma glucose was raised to the same hyperglycemic plateau of approximately 130 mg/dl in all three groups of studies during this procedure. Moreover, the identical hyperglycemic stimulus achieved during the clamp study produced similar time courses and biphasic patterns of insulin and C peptide responses in all three groups (Figs. 1 and 2). However, the magnitude of the insulin response was significantly greater after vs. before GH therapy (P 0.02). First and second phase plasma insulin and C peptide responses in the patients before and after treatment

3 3236 HEPTULLA ET AL. JCE&M 1997 Vol 82 No 10 FIG. 1. Changes in plasma levels of glucose, insulin, and C peptide during the glucose clamp procedure in patients before ( ) and after (E) GH treatment and in control subjects (f and shaded area). Data are the mean SEM. and in controls are summarized in Fig. 2. GH therapy in the patients led to a significant increase in both first phase (from 48 6to pmol/l; P 0.01) and second phase (from to pmol/l; P 0.02) insulin responses vs. pretreatment values, but these responses did not differ from the control values either before or after treatment. Glucosestimulated C peptide responses also increased in the patients after GH therapy, but the changes did not achieve statistical significance compared to pretreatment or control values (Fig. 2). M values during the last 60 min of the hyperglycemic clamp did not differ significantly between patients before and after treatment and controls (178 27, , and mg/m 2 body surface area min, respectively). The FIG. 2. Mean SEM first phase (0 10 min) and second phase ( min) plasma insulin and C peptide responses during clamp procedure in patients before (open bars) and after treatment (hatched bars) with GH and in control subjects (shaded bars). *, P 0.05; **, P 0.01 (vs. pretreatment). FIG. 3. Mean SEM. M/I responses during glucose clamp procedure in patients before and after GH therapy and in control subjects. **, P 0.01 (vs. pretreatment). rate of glucose metabolism was similar in the patients before and after treatment, even though plasma insulin levels were nearly 3-fold higher after GH treatment. Consequently, the M/I ratio fell markedly during therapy (P 0.01), as depicted in Fig. 3. After GH therapy, the increase in second

4 EFFECTS OF GH ON CARBOHYDRATE AND PROTEIN METABOLISM 3237 phase insulin responses ( 2- to 3-fold) and the decrease in M/I ratio ( 75%) were similar in prepubertal and pubertal subjects even though estrogen and testosterone levels remained in the prepubertal range in the Tanner I patients (Table 1). Baseline total branched chain amino acid levels did not differ between the three groups of studies. However, as shown in Fig. 4, the increased insulin response in the patients after GH treatment was associated with a significantly greater suppression in circulating branched chain amino acid levels compared to values during the pretreatment study (74 22 vs mol/l; P 0.02) but not vs. controls ( mol/l). Differences in the suppression of total amino acids, alanine, and phenylalanine concentrations during the clamp before and after treatment were not statistically significant (Table 3). The two subjects with non-gh-deficient short stature showed nearly identical 2- to 3-fold increases in basal and glucose-stimulated insulin responses, growth velocity, and IGF-I levels and similar marked reductions in M/I ratio. The changes in these parameters remained significant (P 0.05) even when the two non-gh-deficient short stature subjects were excluded from the analyses. In addition, the increased suppression of branched chain amino acids after GH therapy was similar in GH-deficient and non-gh-deficient subjects (69 12 mol/l pretreatment vs posttreatment and mol/l pretreatment vs mol/l posttreatment, respectively). Discussion In the present study, the hyperglycemic clamp technique was employed to examine the effects of 6 months of GH FIG. 4. Mean SE changes in total branched chain amino acids (BCAA) during the 120-min hyperglycemic clamp study in patients before and after GH therapy and in controls. *, P 0.02 (vs. pretreatment). TABLE 3. Reduction in circulating amino acid concentrations during the hyperglycemic clamp Pretreatment Posttreatment Total amino acids Alanine Phenylalanine The reduction is the mean concentration during the last 30 min minus the basal concentration. treatment on insulin secretion and insulin action in a group of short children who were generally in good health except for reduced GH concentrations in six of the eight patients. The hyperglycemic clamp technique was chosen because glucose-stimulated insulin responses are closely linked to insulin sensitivity in nondiabetic subjects and because it allowed us to examine the effects of compensatory hyperinsulinemia on circulating amino acid levels. For these purposes, an increment in plasma glucose (50 mg/dl above fasting concentrations) was selected to simulate the physiological rise in plasma glucose that occurs postprandially, whereas previous studies in normal control children have employed much greater increments in plasma glucose (i.e. 125 mg) (1). In our patients, GH treatment resulted in a sharp increase in growth velocity and IGF-I levels and was associated with the development of insulin resistance and peripheral hyperinsulinemia. Nevertheless, even after treatment, insulin sensitivity and glucose-stimulated insulin responses remained within the physiological range of normal control children. These observations support the hypothesis that the increase in circulating GH concentrations that normally occur during puberty is a major contributor to the insulin resistance and peripheral hyperinsulinemia that is seen during adolescence (5). Although the puberty-associated increase in androgens and estrogens (3, 4) may also play a role, the fact that such metabolic alterations were observed after a short course of GH therapy in both prepubertal and pubertal children in this study indicates that elevations in gonadal steroids are not required. Indeed, the antiinsulin effects of gonadal steroids may be mediated in part by their stimulatory action on GH secretion (14). The effects of GH on glucose metabolism have been examined in a large number of in vitro and in vivo studies (15 17). However, the metabolic consequences of GH therapy in currently recommended doses for treatment of children with GH deficiency and other causes of growth failure has not been extensively studied. Standard parameters have included changes in fasting plasma glucose and glycosylated hemoglobin levels, which, as in our patients, have generally not been affected by therapy (18, 19), although a rare patient has developed diabetes while receiving treatment (20). Using oral glucose tolerance testing and iv glucose administration, Walker and colleagues (21) found a similar increase in fasting and glucose-stimulated insulin levels after 1 yr of GH therapy in children with non-gh-deficient short stature. Surprisingly, using the euglycemic clamp procedure, they were unable to demonstrate a change in the rate of peripheral glucose disposal after GH treatment, suggesting that peripheral insulin sensitivity was not altered by GH therapy. It should be noted, however, that the steady state plasma insulin levels achieved in those subjects during the euglycemic

5 3238 HEPTULLA ET AL. JCE&M 1997 Vol 82 No 10 clamp were close to maximally stimulatory values and much higher than those observed in our subjects during the hyperglycemic clamp. Thus, the hyperglycemic clamp procedure used here may have been able to uncover more subtle defects in peripheral insulin action induced by GH than the euglycemic hyperinsulinemic clamp procedure involving high doses of insulin that were used by Walker et al. (21). The observation that first and second phase plasma C peptide as well as plasma insulin levels increased after GH therapy suggests that the secretion of insulin was increased with therapy. It is noteworthy, however, that the changes in plasma C peptide concentrations posttreatment were more modest than those of plasma insulin and did not achieve statistical significance vs. pretreatment values. These discrepancies may be due to the small sample size and greater variability in the C peptide responses. Alternatively, GH therapy may have reduced hepatic clearance of insulin leading to greater peripheral hyperinsulinemia or enhanced renal clearance of C peptide due to increased glomerular filtration rate (22, 23), leading to lower circulating C peptide concentrations. As urinary excretion of C peptide was not determined, we are unable to evaluate this issue. Despite the marked reduction in insulin sensitivity caused by GH therapy, the overall rate of glucose metabolism at the end of the clamp study was similar in the patients before and after treatment and in comparison to control values. Thus, the patients were able to maintain normal rates of glucose metabolism during therapy by increasing peripheral plasma insulin concentrations. The compensatory hyperinsulinemia did not have adverse effects on blood pressure, findings consistent with those in a large sample of healthy Finnish adolescents in whom basal hyperinsulinemia correlated weakly, if at all, with systolic and diastolic blood pressure (24). On the other hand, the compensatory hyperinsulinemia may have contributed to the increase in growth velocity seen with therapy. We have previously hypothesized that compensatory hyperinsulinemia might serve a beneficial role if the metabolic defects induced by GH were restricted to glucose metabolism and spared other insulin-sensitive metabolic fuels (5). Therefore, in this study we were interested in determining the effects of GH treatment on changes in branched chain amino acid levels during the clamp procedure. Branched chain amino acids are essential amino acids that are primarily metabolized in muscle and under these fasting conditions, the fall in circulating branched chain amino acids is due to an inhibition of protein breakdown (25). The finding that the greater insulin response during the clamp in the patients during treatment was associated with greater suppression of branched chain amino acid levels compared to pretreatment values is particularly noteworthy. Thus, increases in plasma insulin as well as IGF-I levels may play a role in the somatotropic effects of GH therapy. References 1. Caprio S, Plewe G, Diamond MP, et al Increased insulin secretion in puberty: a compensatory response to reductions in insulin sensitivity. J Pediatr. 114: Bloch CA, Clemmons P, Sperling MA Puberty reduces insulin sensitivity. J Pediatr. 10: Oppenheimer E, Linden B, DiMartino-Nardi J Decreased insulin sensitivity of pubertal girls with premature adrenarche and acanthosis nigricans. J Clin Endocrinol Metab. 80: Shoupe D, Lobo RA The influence of androgens on insulin resistance. Fertil Steril. 41: Amiel SA, Caprio S, Sherwin RS, et al Insulin resistance of puberty: a defect restricted to peripheral glucose metabolism. J Clin Endocrinol Metab. 72: McGuire EAH, Helderman JH, Tobin JD, Andres R, Berman M Effects of arterial versus venous sampling on analysis of glucose kinetics in man. J Appl Physiol. 41: DeFronzo RA, Tobin JD, Andres R Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol. 237:E214 E Morgan CR, Lararono A Immunoassay of insulin plasma insulin levels of normal, subdiabetic and diabetic rats. Diabetes. 12: Kuzuya H, Blix PN, Horowitz DL, Steiner DF, Rubenstein AM Determination of free and total insulin and C-peptide insulin-treated diabetes. Diabetes. 26: Schoenheimer R, Ratner S, Rittenberg D Studies in protein metabolism. J Biol Chem. 130: Schwenk WF, Beaufrere B, Haymond MW Use of reciprocal pool specific activities to model leucine metabolism in humans. Am J Physiol. 249:E646 E Schwenk WF, Berg PJ, Beaufrere B, Miles JM, Haymond MW Use of t-butyldimethylsilylation in the gas chromatographic mass spectrometric analysis of physiologic compounds found in plasma using electron-impact ionization. Anal Biochem. 141: Spackman DH, Stein WH, Moore S Automatic recording apparatus for use in chromatography of amino acids. Anal Chem. 30: Moll G, Rosenfield R, Fang V Administration of low-dose estrogen rapidly and directly stimulates growth hormone production. Am J Dis Child. 140: Birnbaum RS, Goodman HM Comparison of several insulin-like effects of growth hormone. Horm Metab Res. 11: MacGorman LR, Rizza RA, Gerich JE Physiologic concentrations of growth hormone exert insulin-like and insulin antagonistic effects on both hepatic and extrahepatic tissues in man. J Clin Endocrinol Metab. 53: Rizza RA, Mandarino LJ, Gerich JE Effects of growth hormone on insulin action in man. Mechanism of insulin resistance, impaired suppression of glucose production and impaired stimulation of glucose utilization. Diabetes. 31: Price DA, Clayton PE, Crowne EH, Roberts CR Safety and efficacy of human growth hormone treatment in girls with Turner syndrome. Horm Res. 39(Suppl 2): Weise M, James D, Leitner CM, et al Glucose metabolism in Ullrich Turner syndrome. Long-term effects of therapy with human growth hormone. Horm Res. 39: Verge CF, Donaghue KC, Williams PE, Cowell CT, Silink M Insulin resistant diabetes during growth hormone therapy in a child with SHORT syndrome. Acta Paediatr. 83: Walker J, Chaussain JL, Bougnères PF Growth hormone treatment of children with short stature increases insulin secretion but does not impair glucose disposal. J Clin Endocrinol Metab. 69: Hirschberg R, Kopple JD Effects of growth hormone and IGF-I on renal function. Kidney Int. 36:S20 S Hirschberg R, Rabb H, Ronaldo B, Kopple JD The delayed effects of growth hormone on renal function in humans. Kidney Int. 35: Taittonen L, Uhari M, Turtinen J, Nuutinen M Change in blood pressure during pubertal insulin resistance. Pediatr Res. 41: Fukagawa NK, Minaker KL, Rowe JW, et al Insulin-mediated reduction of whole body protein breakdown: dose response effects on leucine metabolism in post-absorptive man. J Clin Invest. 76: