| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Pediatric Endocrinology |
Department of Pediatrics, Autonomous University, Division of Pediatric Endocrinology, Hospital of Niño Jesús (J.A., N.C., V.B., J.P., M.T.M., M.H.), E-28009 Madrid; and the Laboratory of Molecular and Cellular Neuroendocrinology, Ramón y Cajal Institute (J.A.C.), Madrid, Spain
Address all correspondence and requests for reprints to: Jesús Argente, M.D., Ph.D., Division of Pediatric Endocrinology, Department of Pediatrics, Hospital Niño Jesús, Avenida Menéndez Pelayo 65, 28009 Madrid, Spain.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In conclusion, the GH-IGF axis is dramatically altered in patients with exogenous obesity. However, most changes in the peripheral IGF system appear to be independent of the modifications in GH secretion. In addition, in contrast to current thought, not all of the observed abnormalities are reversed with a significant reduction in the BMI SD score.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The aims of the present study were 1) to determine the serum levels of GH, total IGF-I, fIGF-I, IGF-II, insulin, IGFBP-1, IGFBP-2, IGFBP-3, and GHBP in prepubertal patients with exogenous obesity at the time of diagnosis and compare these data with values from normal subjects of the same age and pubertal stage (21); 2) to investigate the changes in these parameters after reduction in the body mass index (BMI); and 3) to analyze possible correlations between these parameters.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study population included 65 prepubertal Spanish children (Tanner stage I) with exogenous obesity and 174 healthy age-matched prepubertal controls. Patients with obesity had a BMI greater than 2 SD compared to Spanish standards (22).
The mean ages of the controls and obese patients were 7.6 \ 0.2 and 8.6 \ 0.3 yr, respectively. All obese patients consulted the Division of Endocrinology of the Hospital Niño Jesús because they were overweight. All normal subjects consulted our Division of Endocrinology for presumed endocrine abnormalities and were found to be normal. The height of all normal subjects was between -1 and 1 SD according to Spanish standards (22). Spontaneous GH secretion was studied throughout a 24-h period. At least 30 min before beginning the study, a venous catheter was placed in the right arm. Between 0800–0800 h, 1 mL blood was extracted every 30 min. The blood was immediately centrifuged, and the plasma was extracted and frozen until GH analysis was performed. During hospitalization, patients were given a normal diet (breakfast, lunch, snack, and dinner) and water ad libitum and were allowed to move about normally. Lights were turned off between 2300–0700 h on the following morning. The mean level of GH per 24 h, number of GH secretory bursts per 24 h, maximum peak height of GH per 24 h, pulsatile area under the curve (PAGH), and total area under the curve were determined using the computerized mathematical algorithm Cluster (23). The integrated concentration of GH (ICGH) per 24 h was obtained by dividing the total area under the curve by 1440 (the duration of the study in minutes).
For all other parameters, the blood samples used were fasting morning samples. BMI was calculated as weight (kilograms)/height (meters)2. The BMI SD scores were based upon normative data from Spanish children (22). All normal subjects had a BMI between -1 and 1 SD. All subjects and their families were informed of the purpose of the study and gave consent as prescribed by the local human ethics committee.
BMI reduction study
All patients with exogenous obesity received an isocaloric diet with respect to their age, which was less than their habitual caloric intake. They were studied at three different points: 1) clinical diagnosis (baseline); 2) after a 25% reduction of their BMI SD scores with respect to baseline (44.6% of the patients; n = 29), which corresponded to 6 months after diagnosis; and 3) after a reduction of 50% or more of the initial BMI SD score (13.8%; n = 9), approximately 1 yr after the diagnosis.
Biochemical measurements
Serum GH measurements were performed by RIA (Nichols Laboratories, San Juan Capistrano, CA). Total IGF-I was performed by RIA (Nichols Laboratories) after acid-ethanol extraction. IGF-II, IGFBP-2, and fIGF-I were measured by RIA (Diagnostic Systems Laboratories, Webster, TX). Serum IGFBP-1 levels were determined by enzyme-linked immunosorbent assay (Medix Biochemica, Kauniainen, Finland) on nonextracted serum. IGFBP-3 was performed by RIA (Mediagnost, Tübingen, Germany). Intra- and interassay coefficients of variation were 4.2% and 7.2% for GH, 4.9% and 8.9% for IGF-I, 6.2% and 7.3% for fIGF-I, 5.2% and 8.7% for IGF-II, 4.6% and 9.8% for IGFBP-1, 5.7% and 7.2% for IGFBP-2, and 3.6% and 6.1% for IGFBP-3, respectively. Insulin was determined by RIA (Diagnostic Products Corp., Los Angeles, CA). The intra- and interassay coefficients of variations were 5.4% and 7.3%, respectively. GHBP assays were performed in duplicate using a monoclonal antibody assay (Endocrine Sciences, Calabasas Hill, CA), which includes incubating patient serum with excess radiolabeled hGH and the monoclonal antibody MoAb 263 as previously described (21). The intra- and interassay coefficients of variations were 5.6% and 9.5%, respectively.
Statistics
All data are reported as the mean \ SEM. Analyses were performed by one-way ANOVA, followed by Scheffe’s F test. Because only a subpopulation of patients achieved the required BMI SD score reduction during the study period, statistical analyses were performed using the baseline values of each subgroup and to determine whether the baseline values differed between the subpopulations. No significant differences were found between these subpopulations in the baseline levels of any parameter, and results are represented using all data in each group. Correlations were performed using simple regression analysis. Significance was chosen as P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. The indexes
of obesity are represented in Table 2.
|
|
.
Obese subjects had significantly lower PAGH and ICGH per 24 h than
controls (Fig. 1, A and B; P < 0.001).
When obese patients decreased their BMI SD score, both
parameters increased significantly, so that after a reduction of at
least 50% of their BMI SD score their GH secretory pattern
was no longer distinguishable from that of the control population.
Furthermore, the number of GH secretory bursts (Fig. 1C) was
significantly lower (P < 0.001) in obese patients at
baseline compared to the controls, but returned to normal after a 25%
reduction of the BMI SD score.
|
|
). In contrast, fIGF-I and IGF-II levels were
significantly higher at all times (Fig. 2, B and C). All obese patients
were hyperinsulinemic at diagnosis and remained so even after
significantly reducing their BMI SD score
(P < 0.001; Fig. 2D). Serum IGFBP-1 and IGFBP-2 levels
were very low in obese children both before and after weight reduction
(P < 0.001; Fig. 3, A and B). Serum
IGFBP-3 levels were significantly higher at diagnosis and after a 25%
reduction of the BMI SD score, returning to normal only
after a 50% reduction of the BMI SD score. As shown in
Fig. 3D, serum GHBP levels were significantly elevated in obese
patients compared with controls (P < 0.001) and
returned to normal after a significant reduction of the BMI
SD score.
|
|
Table 4 shows the results of all regression
analyses performed. A significant positive correlation was found
between BMI SD score and serum GHBP (Fig. 4A), and this was true at all stages of the study (Table 4). A significant negative correlation was found between the BMI
SD score and fIGF-I (Fig. 4B), total IGF-I, IGFBP-1, and
IGFBP-2, but only after reduction of their BMI SD score by
at least 25% (Table 4). In control subjects BMI was only significantly
correlated with GHBP and IGFBP-2. No significant correlation was found
between BMI SD score and serum IGFBP-3 or insulin
levels.
|
|
) and with IGF-I and the
sum of IGF-I and IGF-II only after weight reduction. However, IGFBP-2
was significantly correlated with IGF-II only at the time of diagnosis.
Free IGF-I, IGF-II, and the sum of IGF-I and IGF-II (Fig. 4D) had
significant positive correlations with serum IGFBP-3 at all times in
obese subjects, but with total IGF-I only after a BMI SD
score reduction of 50% or more. In control subjects there was a
significant correlation between IGFBP-3 and IGF-I, IGF-II, and IGF-I
plus IGF-II, but not fIGF-I. Total IGF-I and fIGF-I had positive
correlations with serum GHBP levels at the time of diagnosis and after
a BMI SD score reduction of 50% or more. Ratio of IGFBP-2/IGF-I and IGFBP-2/fIGF-I
The ratio of IGFBP-2 to IGF-I has been proposed as a parameter to aid in the differentiation between GH deficiency and malnutrition syndromes, because in GH deficiency this ratio is increased, whereas in the later it is decreased. At diagnosis, patients with exogenous obesity had a significantly lower IGFBP-2/IGF-I ratio than controls (0.63 \ 0.24 vs. 1.73 \ 0.19, respectively; P < 0.001). This ratio remained significantly reduced even after a 25% or 50% or more reduction of the BMI SD score (0.58 \ 0.35 and 0.65 \ 0.35, respectively; by ANOVA, P < 0.001). These patients also had a significantly lower IGFBP-2/fIGF-I ratio at diagnosis (386.8 \ 47 vs. 1227.8 \ 113; P < 0.001), which also remained significantly lower than the control value after a 25% (356.1 \ 44; P < 0.001) or 50% or more reduction of the BMI SD score (397.9 \ 89; P < 0.001).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Most researchers agree that GH secretion is significantly decreased in obese patients (1, 2, 3, 4). Our data indicate not only that the pulsatile area is decreased, but that the integrated concentration during a 24-h period is also diminished. Both parameters increase after BMI reduction, so that neither is significantly different from control values after a reduction of 50% or more of the original BMI SD score. Furthermore, at diagnosis, patients with exogenous obesity had a significantly lower number of GH secretory bursts per 24 h, which was normalized after a BMI SD score reduction of only 25%. These data suggest that in exogenous obesity the reductions in PAGH and ICGH are due to modulation of both the amplitude and frequency of the GH secretory bursts, with the caveat that because of the extremely low secretion at diagnosis, low amplitude bursts may not be detected, and the pulse frequency may not actually change.
In this study we confirm previous data showing markedly increased GHBP levels in patients with exogenous obesity (19, 20). This increase in GHBP may reflect an up-regulation of GH receptor concentrations by the state of overnutrition. Our data are in agreement with those of Ramussen et al. (20), who showed that serum GHBP levels returned to normal after a calorie-restricted diet. In our study, after only a 25% reduction in the BMI SD score, GHBP levels returned to normal, as did GH secretion. There was a positive correlation between GHBP levels and BMI at all times in the obese patients and the controls, similar to that reported previously in normal subjects (21). If changes in circulating GHBP levels reflect changes in GH receptor levels in target tissue, these results would suggest a greater sensitivity of these tissues to GH in obese subjects. This could partially explain normal IGF-I and increased IGFBP-3 levels in the face of significantly decreased GH secretion.
Studies analyzing GH secretion and the peripheral GH-IGF system at the time of diagnosis and after weight reduction in the same patients are scant at best (6). We postulated that the major changes occurring in the peripheral IGF axis are controlled largely by signals other than GH, including nutritional factors. Indeed, many of the parameters reported here to be abnormal do not return to control values even after normalization of GH secretion. In addition, serum IGF-I levels are normal when GH secretion is decreased and do not change as GH levels return to control values, further supporting an important role for factors other than GH in the regulation of this system in obese subjects. A similar discordance between GH secretion and the peripheral IGF axis is seen in malnutrition due to anorexia nervosa. In these patients, GH secretion may be normal, increased, or decreased, but total IGF-I levels are decreased dramatically regardless of GH secretory profiles. In anorexia, serum IGFBP-1 and IGFBP-2 levels are both increased significantly, whereas IGFBP-3 is decreased, and similar to what is seen in obesity, these parameters do not return to control values even after normalization of GH secretion (see accompanying manuscript). Together, these observations suggest that in both overnutrition and extreme undernutrition, the peripheral GH-IGF system depends little on the GH secretory pattern.
Serum concentrations of total IGF-I in patients with exogenous obesity have been reported to be low (5, 6, 7), high (8, 9, 10, 11), and normal (12, 13, 14). Our data confirm previous studies by Slowinska-Srzednicka et al. (12), Ghigo et al. (13), and Frystyk et al. (14) showing that total IGF-I levels in these subjects are in the normal range. Furthermore, after BMI SD score reduction, we did not see a significant change in IGF-I levels. This is in contrast to other nutritional disorders, such as diabetes mellitus (24), coeliac disease (25), and anorexia nervosa (see accompanying manuscript), in which total IGF-I levels are significantly modified. Furthermore, in anorexic patients, IGF-I levels did not return to normal after 1 yr of treatment, and at least 2 yr were needed to normalize IGF-I concentrations in coeliac patients, suggesting that an extended period of nutritional therapy may be necessary to normalize the GH-IGF system.
In obese children, fIGF-I levels were significantly elevated at the time of diagnosis, remaining so even after BMI SD score reduction. This observation could explain the normal to increased growth of these subjects despite their low GH levels. If fIGF-I is the biologically active fraction, these children, although having normal total IGF-I levels, would have more biologically active IGF-I to promote normal growth and to feedback to inhibit GH secretion. The presence of high fIGF-I levels in children with exogenous obesity could be related to their hyperinsulinemia, which most likely is involved in the reduction of IGFBP-1 and IGFBP-2 levels. Although IGFBP-3 is the main carrier of IGF-I in serum, it is normally saturated. Hence, a significant decrease in another serum binding protein, in this case both IGFBP-1 and IGFBP-2, could increase the portion of IGF in the free fraction. In addition, we found serum IGF-II levels to be significantly elevated in obese children, which is in agreement with the findings of Frystyk et al. (14), and this factor would also compete for binding to the IGFBPs. Therefore, fIGF-I appears to be a better predictor than IGF-I of a disruption or imbalance in the GH-IGF axis, at least in obesity.
These patients were hyperinsulinemic throughout the entire study period, which may help to explain their low serum IGFBP-1 and IGFBP-2 levels even after significantly reducing their BMI SD score. Serum IGFBP-1 and IGFBP-2, generally thought to be non-GH dependent, were 244% and 260% lower than control values, respectively, and both parameters correlated inversely with insulin levels, as demonstrated previously (26). IGFBP-I had a significant inverse correlation with both total and free IGF-I, although these correlations were stronger after BMI reduction. IGFBP-2 correlated significantly with fIGF-I at all time points, but with total IGF-I only after BMI reduction. These data are in agreement with those reported by Baxter, who suggested that IGFBP-1 acts as a regulator of IGF-I bioavailability (27), and we suggest that IGFBP-2 may also be involved.
Obese children had significantly elevated serum IGFBP-3 levels, similar to what has been described previously (12, 14, 28). This suggests that GH hyposecretion induced by obesity has different effects on the IGF system than those seen in primary GH deficiency, as IGFBP-3 is low in these GH-deficient patients. This further supports the hypothesis that factors other than GH are intervening in the control of this axis in obese patients. This GH-dependent binding protein, IGFBP-3, was found to correlate with IGF-I and the sum of IGF-I plus IGF-II, as seen in normal patients (21), although the correlation with IGF-I was not significant at the time of diagnosis. In addition, we have shown that IGFBP-3 levels correlate with both fIGF-I and IGF-II serum levels.
Many of the correlations between components of the GH-IGF system reported in normal subjects cease to exist in obese subjects before weight loss, whereas other correlations become highly significant (e.g. IGFBP-2 with IGF-II). The physiological significance of these correlations is not always clear. Although some correlations, such as that between IGF-I and IGFBP-3, two GH-regulated proteins, are more obvious, others are not. What can be deduced from the results reported here in obese subjects as well as those in anorexic patients (see accompanying manuscript) is that there is an imbalance in this system during nutritional disorders and that the correlations seen in normal subjects no longer exist during the pathological state. Furthermore, as the system starts to normalize, many of these correlations return.
In summary, this study demonstrates that the GH-IGF axis is dramatically altered in prepubertal children with exogenous obesity and that the peripheral changes are largely GH independent. These children secrete very low amounts of GH, but total IGF-I is in the normal range. Obese patients are hyperinsulinemic and have significantly elevated serum fIGF-I, IGF-II, IGFBP-3, and GHBP levels, although IGFBP-1 and IGFBP-2 are significantly decreased. The cascade of effects is difficult to determine because the obesity occurs over an extended period of time, and the reported changes most likely also occur slowly; however, the hyperinsulinemia, a direct result of overnutrition, may underlie the decreased production of IGFBP-1 and IGFBP-2. This, in turn, could cause more IGF-I to be in the free, biologically active form. More biologically active IGF-I could feedback to reduce GH release while maintaining normal to increased systemic growth. A reduction in the BMI SD score results in significant changes in some of these parameters, but the children remain hyperinsulinimic with elevated fIGF-I and IGF-II and very low IGFBP-1 and IGFBP-2 levels. Normalization of the entire axis may require a more extended period of time.
|
Footnotes |
|---|
Received October 28, 1996.
Revised December 11, 1996.
Revised January 24, 1997.
Accepted March 25, 1997.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. Iwakura, T. Akamizu, H. Ariyasu, T. Irako, K. Hosoda, K. Nakao, and K. Kangawa Effects of ghrelin administration on decreased growth hormone status in obese animals Am J Physiol Endocrinol Metab, September 1, 2007; 293(3): E819 - E825. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Eliakim, D. Nemet, F. Zaldivar, R. G. McMurray, F. L. Culler, P. Galassetti, and D. M. Cooper Reduced exercise-associated response of the GH-IGF-I axis and catecholamines in obese children and adolescents J Appl Physiol, May 1, 2006; 100(5): 1630 - 1637. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Afghani, M. L. Cruz, and M. I. Goran Impaired Glucose Tolerance and Bone Mineral Content in Overweight Latino Children With a Family History of Type 2 Diabetes Diabetes Care, February 1, 2005; 28(2): 372 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Buijs, J. A. Romijn, J. Burggraaf, M. L. de Kam, M. Frolich, M. T. Ackermans, H. P. Sauerwein, A. F. Cohen, A. E. Meinders, and H. Pijl Glucose homeostasis in abdominal obesity: hepatic hyperresponsiveness to growth hormone action Am J Physiol Endocrinol Metab, July 1, 2004; 287(1): E63 - E68. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. De Marinis, A. Bianchi, A. Mancini, R. Gentilella, M. Perrelli, A. Giampietro, T. Porcelli, L. Tilaro, A. Fusco, D. Valle, et al. Growth Hormone Secretion and Leptin in Morbid Obesity before and after Biliopancreatic Diversion: Relationships with Insulin and Body Composition J. Clin. Endocrinol. Metab., January 1, 2004; 89(1): 174 - 180. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W F Paterson and M D C Donaldson Growth hormone therapy in the Prader-Willi syndrome Arch. Dis. Child., April 1, 2003; 88(4): 283 - 285. [Full Text] [PDF]
|
|
|
|
|
|
M. Ozata, C. Dieguez, and F. F. Casanueva The Inhibition of Growth Hormone Secretion Presented in Obesity Is Not Mediated by the High Leptin Levels: A Study in Human Leptin Deficiency Patients J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 312 - 316. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Sandhu, D. B. Dunger, and E. L. Giovannucci Insulin, Insulin-Like Growth Factor-I (IGF-I), IGF Binding Proteins, Their Biologic Interactions, and Colorectal Cancer J Natl Cancer Inst, July 3, 2002; 94(13): 972 - 980. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Z. Kasa-Vubu, A. Barkan, P. Olton, T. Meckmongkol, N. E. Carlson, and C. M. Foster Incomplete Modified Fast in Obese Early Pubertal Girls Leads to an Increase in 24-Hour Growth Hormone Concentration and a Lessening of the Circadian Pattern in Leptin J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1885 - 1893. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. T. Chlebowski, E. Aiello, and A. McTiernan Weight Loss in Breast Cancer Patient Management J. Clin. Oncol., February 15, 2002; 20(4): 1128 - 1143. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Silfen, A. M. Manibo, M. Ferin, D. J. McMahon, L. S. Levine, and S. E. Oberfield Elevated Free IGF-I Levels in Prepubertal Hispanic Girls with Premature Adrenarche: Relationship with Hyperandrogenism and Insulin Sensitivity J. Clin. Endocrinol. Metab., January 1, 2002; 87(1): 398 - 403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Coutant, F. Boux de Casson, S. Rouleau, O. Douay, E. Mathieu, M. Audran, and J. M. Limal Body Composition, Fasting Leptin, and Sex Steroid Administration Determine GH Sensitivity in Peripubertal Short Children J. Clin. Endocrinol. Metab., December 1, 2001; 86(12): 5805 - 5812. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Pijl, J. G. Langendonk, J. Burggraaf, M. Frolich, A. F. Cohen, J. D. Veldhuis, and A. E. Meinders Altered Neuroregulation of GH Secretion in Viscerally Obese Premenopausal Women J. Clin. Endocrinol. Metab., November 1, 2001; 86(11): 5509 - 5515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. KAMEL, S. NORGREN, A. ELIMAM, P. DANIELSSON, and C. MARCUS Effects of Growth Hormone Treatment in Obese Prepubertal Boys J. Clin. Endocrinol. Metab., April 1, 2000; 85(4): 1412 - 1419. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. H. Slyper Childhood Obesity, Adipose Tissue Distribution, and the Pediatric Practitioner Pediatrics, July 1, 1998; 102(1): 4e - 4. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, children with exogenous obesity after ponderal reduction
equal to 25% of their original BMI SD scores (n =
29); 
, children with exogenous obesity after a ponderal reduction of
50% or more of their BMI SD scores (n = 9).