| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Pediatrics (L.A., A.C., C.E., M.F.-C., P.A., D.Y.), Research Institute, Hospital Vall dHebron, Centro de Investigación Biomédica en Red de Enfermedades Raras (CIBERER), Autonomous University, 08035 Barcelona, Spain; Medical Unit (R.E.), Pfizer S.A., 28040 Madrid, Spain; Department of Clinical Biochemistry (M.L.G.), Hospital Germans Trias i Pujol, 08916 Badalona, Spain; Medical Endocrine Care (H.W.), Pfizer GmbH, 76032 Karlsruhe, Germany; and Endocrine Care Team (L.F.), Pfizer Health AB, 11251 Stockholm, Sweden
Address all correspondence and requests for reprints to: Laura Audí, Servicio de Pediatría, Unidad de Endocrinología, Hospital Maternoinfantil Vall dHebron, Paseo Vall dHebron 119, 08035 Barcelona, Spain. E-mail: laudi{at}ir.vhebron.net.
| Abstract |
|---|
|
|
|---|
Objective: Our objective was to evaluate whether the d3/fl-GHR genotypes influence insulin sensitivity in short SGA children before or after puberty onset or during GH therapy.
Design: We conducted a 2-yr prospective, controlled, randomized trial.
Setting: Thirty Spanish hospitals participated. Auxological, GH secretion, and glucose homeostasis evaluation was hospital based, whereas molecular analyses and data computation were centralized.
Patients: Patients included 219 short SGA children [body mass index SD score (SDS)
2.0]; 159 were prepubertal (group 1), and 60 had entered puberty (group 2).
Intervention: Seventy-eight patients from group 1 were treated with GH (66 µg/kg·d) for 2 yr (group 3).
Main Outcome Measures: Previous and 2-yr follow-up auxological and biochemical data were recorded, d3/fl-GHR genotypes determined, and data analyzed.
Results: In groups 1 and 2, fasting glucose, insulin, homeostasis model assessment (HOMA), and quantitative insulin sensitivity check index (QUICKI) were similar in each d3/fl-GHR genotype. Group 2 glucose, insulin, and HOMA were significantly higher and QUICKI lower than in group 1. In group 3 GH-treated patients, height SDS, growth velocity SDS, fasting glucose, insulin, and HOMA significantly increased as did body mass index SDS at the end of the second year, and QUICKI decreased during the first and second years, with no differences among the d3/fl-GHR genotypes.
Conclusion: In short SGA patients, the d3/fl-GHR genotypes do not seem to influence prepubertal or pubertal insulin sensitivity indexes or their changes over 2 yr of GH therapy (66 µg/kg·d).
| Introduction |
|---|
|
|
|---|
Although postnatal spontaneous growth parameters in short SGA and adequate-for-gestational-age (AGA) children did not seem to be influenced by the d3/fl-GHR genotypes in our studies (4, 6, 7), two data sets pointed toward an effect of these genotypes on postnatal catch-up growth: first, the decreased frequency of homozygous d3/d3-GHR genotype among the short SGA patients of the present study compared with our normal-height control population and the significantly lower target height in SGA d3/d3-GHR genotype (8), and second, the recent report of Schreiner et al. (9) showing that catch-up growth of preterm infants with birth weight under 1500 g was significantly better in children carrying at least one d3-GHR allele with respect to fl/fl-GHR homozygotes. The latter, in addition, presented significantly lower IGF-I and IGF-binding protein-3 (IGFBP-3) SD score (SDS). Whether the increased catch-up growth and IGF-I levels may subsequently affect glucose homeostasis and insulin sensitivity was not investigated in this cohort. Recently, Mericq et al. (10) reported no significant differences in baseline nocturnal GH secretion, IGF-I SDS, IGFBP-3 SDS, or insulin sensitivity among the d3/fl-GHR genotypes in prepubertal SGA children, although the number of patients was small and only four d3/d3-GHR patients were assessed. In adults, a decreased frequency of homozygous d3/d3-GHR genotype has recently been reported in a type 2 diabetic population, with patient phenotype being worse in those carrying it (11). Some of these observations would suggest that the d3/d3-GHR genotype could protect against shortness in SGA and AGA children (8, 9) and insulin resistance in the general population (11), although those with this genotype in the short SGA population (8) or type 2 diabetes population (11) could have a worse phenotype due to other genetic influences.
In SGA patients, insulin resistance has been reported to be more prevalent in those with spontaneous catch-up growth (12, 13) and also in the short group when treated with GH (14, 15, 16, 17).
GH influences glucose homeostasis through increased lipolysis and insulin antagonism, resulting in glucose intolerance and insulin resistance (18, 19). Decreased insulin sensitivity has been described as physiological in children at puberty (20, 21) while developing in parallel with the increase in GH secretion; this phenomenon is also observed in growth-retarded children treated with GH (22, 23).
To ascertain whether the d3/fl-GHR genotypes influence glucose homeostasis in short SGA children, we analyzed data on insulin sensitivity in 219 short SGA patients from our prospective controlled trial (4, 8), 159 prepubertal, 60 pubertal, and 78 of the prepubertal patients treated with GH (66 µg/kg·d) for 2 yr, and data were analyzed according to the d3/fl-GHR genotypes.
| Patients and Methods |
|---|
|
|
|---|
A total of 222 Spanish short SGA Caucasian patients (111 boys, 9.5 ± 3.3 yr; 111 girls, 8.7 ± 3.0 yr) were included in a 2-yr prospective, controlled trial. Thirty hospitals participated, and patients were recruited from September 2001 to December 2002. Patients were assigned to treatment or not in a randomized and double-blind manner at each hospital. Inclusion criteria were gestational age more than 35 wk; birth weight and/or birth length less than –2 SD (24); age over 3 yr; height less than –2 SD (25); never having been treated with GH or other anabolic agents; normal thyroid, kidney, gastrointestinal, lung, and liver functions; GH response peak higher than 10 ng/ml; and normal karyotype in girls. Exclusion criteria were neonatal brain injury, chromosomopathies, malformation syndromes, chronic diseases, and steroid therapy.
Height and weight were measured 12 and 6 months before study inclusion, at inclusion, and every 4 months thereafter. Growth velocities before inclusion and for the two consecutive 1-yr periods were calculated. Parental heights were measured in 166 patients (81 boys and 85 girls). Height and weight were transformed into SDS according to age-, sex-, and pubertal stage-matched control values recently reported in Spanish cross-sectional and longitudinal studies (25, 26, 27).
Of the 222 patients, 219 had BMI SDS below +2.0 and were included in the present study (three boys with BMI SDS over +2.1 were excluded). Of the 219 patients, 159 (81 boys, 8.1 ± 2.7 yr; 78 girls, 7.2 ± 2.4 yr) were prepubertal at inclusion and remained so during the following year (group 1) and 60 (27 boys, 13.9 ± 0.9 yr; 33 girls, 12.1 ± 1.1 yr) had started puberty at inclusion or entered puberty during the following year (testicular volume > 4 ml in boys and appearance of breast buds in girls) (group 2).
Seventy-eight patients from group 1 (46 boys, 8.3 ± 2.7 yr; 32 girls, 7.3 ± 2.9 yr) were treated with GH (66 µg/kg body weight/day) for 2 yr and remained prepubertal during the following 24 months (group 3), except 15 of them who entered puberty at the end of the second year. Height SDS, yearly height SDS increment (first- or second-year height SDS minus inclusion height SDS), body mass index (BMI) SDS, and growth velocity (GV) SDS were calculated.
Blood fasting glucose and insulin were measured at inclusion in all patients and at the end of the first and the second years in group 3 GH-treated patients.
d3/fl-GHR genotyping
The d3/fl-GHR genotypes were evaluated at our laboratory according to Pantel et al. (1) with modifications (4, 8). When a homozygous d3/d3-GHR genotype was detected (a single band corresponding to 532 bp) and/or when a band potentially corresponding to the 935-bp product was mildly amplified, a second PCR using only G1 and G3 primers was carried out, followed by electrophoresis.
Hormone measurements
Serum glucose, insulin, and GH were measured in all patients at each hospital laboratory by commercial assays. Insulin resistance and sensitivity were assessed calculating the homeostasis model assessment (HOMA) index (28) and the quantitative insulin sensitivity check index (QUICKI) (29), respectively. HOMA and QUICKI values were transformed into SDS according to age- and pubertal stage-matched control values (30).
Serum IGF-I, IGFBP-3, and leptin were measured in 157 patients (101 included in group 1) at a central laboratory (M.L.G.) by commercial assays: Nichols Institute (San Juan Capistrano, CA) for IGF-I and IGFBP-3 and Linco Research (St. Louis, MO) for leptin. Reference values for IGF-I and IGFBP-3 SDS calculations were generated in the central laboratory (31).
Ethics
This work was approved by the ethics committees of each participating hospital. Written informed consent was obtained for each subject over 12 yr of age, and informed parental consent was also obtained for all patients regardless of age.
Statistical analysis
Results are expressed as mean ± SD or as percentiles. Differences for the variables evaluated between groups 1 and 2 and between baseline and first and second years for group 3 and among the d3/fl-GHR genotypes in each group were calculated using parametric (Bonferroni/Dunn corrected ANOVA) or nonparametric (Mann-Whitney U and Kruskal-Wallis) tests; Kolmogorov-Smirnov test was applied to test for normality distribution. A stepwise regression analysis was performed between baseline ln-transformed HOMA index values and other biochemical and anthropometric measurements in 101 prepubertal patients included in group 1. The Statview 4.5 program (Abacus Concepts, CA) was used.
| Results |
|---|
|
|
|---|
At inclusion, chronological age (CA) was significantly higher (P < 0.0001) in patients having started puberty (group 2) and did not differ among the d3/fl-GHR genotypes in each prepubertal (group 1) and pubertal (group 2) group (Table 1
). Birth weight SDS, birth length SDS, paternal height SDS, maternal height SDS, target height SDS, height SDS, and BMI SDS did not differ between groups 1 and 2 or among the three d3/fl-GHR genotypes in each group, except for BMI SDS in group 2, which was significantly lower in the homozygous fl/fl-GHR genotype (Table 1
).
|
In 101 group 1 patients (43 girls and 58 boys), no statistically significant differences were observed for anthropometric or biochemical parameters between sexes, except for leptin serum levels, which were significantly higher in girls (4.4 ± 3.1 vs. 3.3 ± 2.0 np/ml; P = 0.0018). Stepwise regression analysis between ln-transformed HOMA index values and other biochemical and anthropometric parameters failed to reveal any significant association.
Group 3: patients prepubertal at inclusion and during 2 yr of GH therapy
Seventy-eight patients from group 1 were treated with GH (66 µg/kg·d) for 2 yr and remained strictly prepubertal during the follow-up (group 3) except 15 of them who presented first signs of puberty at the end of the second year of GH therapy. At inclusion, and similar to the entire group 1, CA, height SDS, BMI SDS, GV SDS, fasting glucose and insulin, HOMA, HOMA SDS, QUICKI, and QUICKI SDS were similar in each d3/fl-GHR genotype (Table 2
). HOMA and QUICKI differed significantly from values in prepubertal controls, with HOMA being lower and QUICKI higher (Fig. 1
). Only five of 78 (6.4%) patients presented a HOMA index higher than 3; three of the five had a heterozygous d3/fl-GHR genotype, and two were homozygous, one d3/d3-GHR, and the other fl/fl-GHR.
|
|
height SDS) at 1 and 2 yr among the d3/fl-GHR genotypes (Table 2
BMI SDS had increased significantly at the end of yr 2 (P < 0.0001), and this increase was mainly due to the d3/fl-GHR and fl/fl-GHR genotype values. A tendency toward lower BMI SDS values was observed in the d3/d3-GHR genotype vs. the other two genotypes at yr 1 and 2, although they did not reach statistical significance; the d3/d3-GHR genotype was the only one to show no BMI SDS increase at the end of the second year (Table 2
).
Fasting glucose and insulin significantly increased at the end of the first and second years of GH therapy as did HOMA and HOMA SDS (Table 2
and Fig. 1
), whereas QUICKI and QUICKI SDS significantly decreased (Table 2
and Fig. 1
) and values were similar in each d3/fl-GHR genotype (Table 2
). At 1 and 2 yr of GH therapy, HOMA and QUICKI differed significantly from the normal control values (Fig. 1
); HOMA was significantly higher and QUICKI significantly lower. HOMA and QUICKI values did not differ between 1 and 2 yr of GH therapy. HOMA and QUICKI at 1 and 2 yr were significantly correlated with values at inclusion (r2 = 0.322, P < 0.0001, and r2 = 0.424, P < 0.0001, respectively). During GH therapy, 22 additional SGA patients presented a HOMA index over 3; 15 of the 22 (68%) presented a fl/fl-GHR genotype, six (27%) were heterozygous d3/fl-GHR, and one (5%) was homozygous d3/d3-GHR.
| Discussion |
|---|
|
|
|---|
To ascertain the effect of puberty and of GH therapy in our cohort of short SGA children with BMI SDS below +2.0, insulin resistance and sensitivity were estimated using HOMA and QUICKI calculations at baseline in prepubertal and pubertal groups and in a prepubertal subgroup treated with GH (66 µg/kg·d) for 2 yr, and these data were analyzed according to d3/fl-GHR genotypes.
Our results demonstrated that insulin resistance increased and insulin sensitivity decreased significantly in short SGA children both at puberty (significant differences between prepubertal group 1 and pubertal group 2) and when on GH therapy (at the end of yr 1 and 2 in group 3), which concurs with previously shown data (14, 15, 16, 17, 32). In each group (baseline prepubertal and pubertal and prepubertal after 1 and 2 yr of GH therapy), fasting glucose, insulin, HOMA, and QUICKI did not differ among the three d3/fl-GHR genotypes, indicating a lack of influence of these genotypes in short SGA children on glucose homeostasis at prepuberty and puberty and when patients were on high-dose GH therapy. Although the hyperinsulinemic-euglycemic clamp is considered the gold standard for assessing insulin sensitivity, it has recently been shown that simpler estimates of insulin resistance and sensitivity in nondiabetic children and adolescents, such as fasting glucose-to-insulin ratio, HOMA, and QUICKI are closely related to the hyperinsulinemic-euglycemic clamp (33, 34, 35). Despite restricting the analysis to nonobese children (only three were excluded), we obtained similar results when we chose to repeat the analyses with a narrower BMI SDS range (from –1.75 to +1.75), as representative of a strictly normal range (data not shown).
Insulin resistance has been reported in SGA children even at early ages (32, 36), although it was more prevalent in prepubertal children with early rapid weight gain and catch-up growth (12, 13, 36); however, a recent study reported normal glucose tolerance in a case-control study of SGA and AGA children (37). Our data in short SGA children with a BMI below +2.0 SDS show that insulin resistance, measured by a HOMA index higher than 3 (38), was present at prepuberty in 6.4%; among these, the d3/fl-GHR genotype distribution did not differ from that of the entire group 1 (data not shown), and the percentages remained similar at baseline in group 3 subsequently treated with GH. Baseline prepubertal values compared with those in a prepubertal normal height and weight control group showed significantly decreased HOMA and increased QUICKI. This again indicated the heterogeneity of glucose homeostasis in short, nonobese prepubertal SGA patients. During GH therapy, 22 of 73 patients (30%) who had a normal baseline values became insulin resistant, and this yielded values significantly different from those in the control group. Distribution of d3/fl-GHR genotypes again showed no association with the appearance of insulin insensitivity. This increase in insulin resistance under GH therapy has been described in short SGA patients (14, 15, 16, 17), although it has also been shown that insulin sensitivity returned to baseline values when GH therapy was withdrawn (39, 40, 41) and that long-term GH therapy does not increase the risk for diabetes type 2 and metabolic syndrome in adulthood (42). If GH therapy worsened insulin sensitivity in a certain percentage of SGA patients (30% in our study), this was not related to the d3/fl-GHR genotypes.
Although 15 of the 78 GH-treated patients entered puberty at the end of the second year, we showed previously that growth response did not differ between the entire group and the strictly prepubertal one (4), nor did glucose homeostasis parameters, as shown in the present study in which no difference was observed between the first- and second-year values. Moreover, results were similar when group 3 was restricted to the 63 strictly prepubertal patients (data not shown).
GH therapy in prepubertal short SGA children has been shown to improve body composition and significantly increase BMI SDS from baseline negative values to normal mean values (43) as also shown in our study. BMI SDS tended to be lower in the homozygous d3/d3-GHR genotype and at the end of the second year did not increase in this genotype, whereas it was significantly higher in the fl/fl-GHR genotype, and this change was not accompanied by significant differences in growth response or in glucose homeostasis. Whether this represents a real metabolic difference in response to GH therapy or a fortuitous finding remains to be confirmed by subsequent studies. A similar finding was recently reported by Mericq et al. (10) in a group of prepubertal SGA patients (at baseline examination, the four SGA children with a d3/d3-GHR genotype showed the lowest BMI SDS compared with the other two genotypes, although it did not reach statistical significance owing to the limited number of individuals) and by Binder et al. (44) in Turner syndrome patients, both at start and end of GH therapy.
In summary, the results of our 2-yr controlled clinical trial show that in short non-GH-deficient SGA children, the spontaneous growth rate and glucose homeostasis, at prepuberty and puberty, were not influenced by the d3/fl-GHR genotypes, nor was patient responsiveness to 66 µg/kg·d GH. In view of the low d3/d3-GHR genotype frequency, a bias in our results cannot be completely ruled out, and additional controlled prospective studies, including a similar number of each d3/fl-GHR genotype, will be required to confirm our results.
| Acknowledgments |
|---|
Members of the Spanish SGA Study Group include J. Ajram, A. Aragones, A. Arroyos, A. Balaguer, M. J. Ballester, J. Bel, M. V. Borrás, J. Bosch, N. Cabrinety, M. Caimarí, M. Camprubí, P. Cantero, R. Cañete, G. Cao, G. Carreras, R. Corripio, A. De Ureta, C. J. Del Valle, R. Espigares, A. Feliu, C. Fernández, J. Ferragut, A. Ferrández, A. Ferrer, M.E. Gallego, A. Gómez, J. P. González, R. Gracia, G. Grau, M. Gussinyé, P. Gutierrez, M.T. Herráez, L. Ibáñez, J. I. Labarta, J. L. Lechuga, G. Lledó, A. Llusà, R. López, J. P. López-Siguero, L. Lorenzo, C. Luzuriaga, A. Mainou, M. V. Marcos, M. J. Martínez-Aedo, P. Martul, E. Mayayo, A. Montesdeoca, R. Nosas, A. Oliver, M. J. Pisonero, N. Pons, J. M. Rial, I. Rica, S. Rite, J. Rodrigo, F. Rodríguez Hierro, A. Rodríguez, I. Rodríguez, A. Romo, J. Sánchez del Pozo, E. Sastre, B. Sinués, B. Sobradillo, J. Tacons, A. Vela, and E. Vicens-Calvet.
| Footnotes |
|---|
Disclosure Summary: L.A., A.C., C.E., M.F.-C., P.A., D.Y., R.E., M.L.G., H.W., and L.F. have nothing to declare.
First Published Online April 29, 2008
1 * L.A. and A.C. contributed equally to this work. ![]()
Abbreviations: AGA, Adequate for gestational age; BMI, body mass index; CA, chronological age; GHR, GH receptor; GV, growth velocity; HOMA, homeostasis model assessment; IGFBP-3, IGF-binding protein-3; QUICKI, quantitative insulin sensitivity check index; SDS, SD score; SGA, short for gestational age.
Received January 22, 2008.
Accepted April 23, 2008.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. J. E. Wassenaar, O. M. Dekkers, A. M. Pereira, J. M. Wit, J. W. Smit, N. R. Biermasz, and J. A. Romijn Impact of the Exon 3-Deleted Growth Hormone (GH) Receptor Polymorphism on Baseline Height and the Growth Response to Recombinant Human GH Therapy in GH-Deficient (GHD) and Non-GHD Children with Short Stature: A Systematic Review and Meta-Analysis J. Clin. Endocrinol. Metab., October 1, 2009; 94(10): 3721 - 3730. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Sorensen, L. Aksglaede, T. Munch-Andersen, N. J. Aachmann-Andersen, H. Leffers, J. W. Helge, L. Hilsted, and A. Juul Impact of the Growth Hormone Receptor Exon 3 Deletion Gene Polymorphism on Glucose Metabolism, Lipids, and Insulin-Like Growth Factor-I Levels during Puberty J. Clin. Endocrinol. Metab., August 1, 2009; 94(8): 2966 - 2969. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |