OBJECTIVE: To evaluate longitudinal growth, pubertal development and final height in patients with congenital hypothyroidism (CH) detected by a neonatal screening programme, and to identify factors potentially affecting growth outcome. PATIENTS: Fifty-five patients (41 females) detected by neonatal screening and followed longitudinally from the time of diagnosis and treatment (25+/-5 days) up to the age of 17+/-0.5 years were evaluated retrospectively. RESULTS: Pubertal development began and progressed normally in both males and females. In boys, a testicular volume of 4 ml was reached at 11.3+/-1.0 years. In girls breast enlargement (B2) occurred at a mean age of 10.3+/-1.2 years and the mean age of menarche was 12.5+/-1.2 years. The onset and the progression of puberty were independent of the aetiology, the severity of CH and the timing of the beginning of treatment. Girls treated with an initial amount of L-thyroxine (L-T4) greater than 8 microg/kg per day showed an earlier onset of puberty (B2 9.4+/-0.9 years; menarche 11.5+/-0.8 years) compared with girls treated with a lower initial dose of L-T4 (B2 10.5+/-1.2 years; menarche 12.6+/-1.2 years; P<0.02). However, both groups attained a similar final height (-0.1+/-1.0 SDS and 0.4+/-1.0 SDS, respectively), which in both cases was above the target height (P=0.03). All the patients in the study attained a mean final height (0.1+/-1.1 SDS) within the normal range for the reference population and above the target height (-0.9+/-0.9 SDS, P<0.0001). No significant relationship was found between final height and severity of CH at diagnosis, initial L-T4 dosage or aetiology of the defect. Patients with ectopic gland, thyroid aplasia or in situ gland attained a similar mean final height (0.1+/-1.1 SDS, 0.5+/-1.0 SDS and -0.5+/-1.0 SDS, respectively), which was in all cases greater than target height (-1.0+/-0.9, -0.6+/-0.8, -0.9+/-0.8 respectively; P<0.05). CONCLUSIONS: Our results suggest that conventional management of children with CH detected by neonatal screening leads to normal sexual development and normal adult height, and that the major factor determining height in these children is familial genetic growth potential.
M Salerno, M Micillo, S Di Maio, D Capalbo, P Ferri, T Lettiero, and A Tenore
M Cools, P Hoebeke, K P Wolffenbuttel, H Stoop, R Hersmus, M Barbaro, A Wedell, H Brüggenwirth, L H J Looijenga, and S L S Drop
Most patients with NR5A1 (SF-1) mutations and poor virilization at birth are sex-assigned female and receive early gonadectomy. Although studies in pituitary-specific Sf-1 knockout mice suggest hypogonadotropic hypogonadism, little is known about endocrine function at puberty and on germ cell tumor risk in patients with SF-1 mutations. This study reports on the natural course during puberty and on gonadal histology in two adolescents with SF-1 mutations and predominantly female phenotype at birth.
Design and methods
Clinical and hormonal data and histopathological studies are reported in one male and one female adolescent with, respectively, a nonsense mutation (c.9T>A, p.Tyr3X) and a deletion of the first two coding exons (NCBI36/hg18 Chr9:g.(126306276-126307705)_(126303229-126302828)del) of NR5A1, both predicted to fully disrupt gene function.
LH and testosterone concentrations were in the normal male range, virilization was disproportionate to the neonatal phenotype. In the girl, gonadectomy at 13 years revealed incomplete spermatogenesis and bilateral precursor lesions of testicular carcinoma in situ. In the boy, at the age of 12, numerous germ cells without signs of malignancy were present in bilateral testicular biopsy specimen.
In SF-1 mutations, the neonatal phenotype poorly predicts virilization at puberty. Even in poorly virilized cases at birth, male gender assignment may allow spontaneous puberty without signs of hypogonadotropic hypogonadism, and possibly fertility. Patients with SF-1 mutations are at increased risk for malignant germ cell tumors. In case of preserved gonads, early orchidopexy and germ cell tumor screening is warranted. The finding of premalignant and/or malignant changes should prompt gonadectomy or possibly irradiation.
D. B. GRANT, D. B. DUNGER, and E. C. BURNS
This paper reviews the outcome in 12 children with hyperinsulinaemic hypoglycaemia who first developed symptoms between the ages of 2 and 8 months and who were treated with diazoxide (5 - 20 mg/kg/day) for 2-13 years. Two cases required subtotal pancreatectomy at the ages of 5 and 10 years because of recurrent hypoglycaemia and one girl with severe retardation died at the age of 6 years while still on diazoxide therapy. Two patients aged 3.5 and 9 years are still on treatment and in 7 cases diazoxide was discontinued between the ages of 2.5 and 14 years, indicating that spontaneous remission can be expected in a high proportion of children with post-neonatal hyperinsulinaemic hypoglycaemia. Of the 9 children who started diazoxide within 3 months of the onset of symptoms, 5 are of normal intelligence and 4 are moderately retarded (IQs 63-71). In 3 children diazoxide was started 8 months to 3 years after the onset of symptoms; two are retarded (IQs 60-70) and the third was severely retarded and died aged 6 years.
H. Krawczynska, M. Zachmann, and A. Prader
Urinary testosterone glucuronide and sulphate was determined separately by gas chromatography in 39 newborns and young infants (34 males and 5 females). In all cases, testosterone sulphate was higher than glucuronide. Boys excreted more of both conjugates (sulphate 6.7, glucuronide 2.2 μg/24 h) than girls (1.1 and 0.7 μg/24 h, respectively). Boys older than 3 weeks had higher values than boys younger than 2 weeks. The levels correlated positively with chronological age, negatively with the gestational age and not at all with the bilirubin levels. It is concluded that testosterone is excreted preferentially as the sulphate in the newborn period and that the high sulphokinase activity in foetal and neonatal testes is more likely responsible for this phenomenon than an impaired glucuronizing capacity of the liver.
M Peter, K Bunger, SL Drop, and WG Sippell
We performed a molecular genetic study in two patients with congenital hypoaldosteronism. An original study of these patients was published in this Journal in 1982. Both index cases, a girl (patient 1) and a boy (patient 2). presented with salt-wasting and failure to thrive in the neonatal period. Parents of patient 1 were not related, whereas the parents of patient 2 were cousins. Endocrine studies had shown a defect in 18-oxidation of 18-OH-corticosterone in patient 1 and a defect in the 18-hydroxylation of corticosterone in patient 2. Plasma aldosterone was decreased in both patients, whereas 18-OH-corticosterone was elevated in patient 1 and decreased in patient 2. Plasma corticosterone and 11-deoxycorticosterone were elevated in both patients, whereas cortisol and its precursors were in the normal range. According to the nomenclature proposed by Ulick, the defects are termed corticosterone methyl oxidase (CMO) deficiency type II in patient 1, and type I in patient 2 respectively. Genetic defects in the gene CYP11B2 encoding aldosterone synthase have been described in a few cases. In patient 1, we identified only one heterozygous amino acid substitution (V386A) in exon 7, which has no deleterious effect on the enzyme activity. In patient 2 and his older brother, we identified a homozygous single base exchange (G to T) in codon 255 (GAG), causing a premature stop codon E255X (TAG). The mutant enzyme has lost the five terminal exons containing the haem binding site, and is thus a loss of function enzyme. This is only the second report of a patient with CMO deficiency type II without a mutation in the exons and exon-intron boundaries, whereas the biochemical phenotype of the two brothers with CMO deficiency type I can be explained by the patient's genotype.
J Kratzsch, A Deimel, A Galler, T Kapellen, A Klinghammer, and W Kiess
OBJECTIVE: We investigated whether or not serum levels of the soluble leptin receptor (sOB-R) and leptin are related to anthropometric and metabolic changes during pubertal development of children and adolescents with type 1 diabetes mellitus. DESIGN AND METHODS: Blood levels of sOB-R, leptin and HbA1C, as well as body-mass index (BMI), diabetes duration and daily insulin doses, were determined in 212 (97 girls; 115 boys) children with type 1 diabetes mellitus and compared with the sOB-R serum levels in 526 healthy children and adolescents. RESULTS: OB-R serum levels and parallel values of the molar ratio between sOB-R and leptin were significantly higher in children with diabetes than in normal children (P<0.05) in almost all investigated Tanner stages. Furthermore, in the entire group of patients, we demonstrated statistically significant correlations (P<0.02) between sOB-R and the duration of diabetes (r=0.30), HbA1c levels (r=0.32) and the insulin dose (r=0.18). Multiple-regression analysis revealed that HbA1c (12.4%), height (7.9%) and duration of diabetes (8.7%) contributed to 29% variance of sOB-R in diabetic children. CONCLUSIONS: Our data suggest that poor glycemic control in diabetes may lead to increased serum levels of sOB-R. This regulation of sOB-R appears to be independent of leptin, but may have an impact on leptin action. The consequently developing molar excess of sOB-R related to leptin could reduce leptin sensitivity and may, therefore, influence leptin-related anthropometric and metabolic abnormalities.
W Kiess, M Anil, WF Blum, P Englaro, A Juul, A Attanasio, J Dotsch, and W Rascher
The ob protein, termed leptin, is produced by adipocytes and is thought to act as an afferent satiety signal regulating weight through suppressing appetite and stimulating energy expenditure in humans and/or rodents. Insulin has been found to be a potent stimulator of leptin expression in rodents. It is unclear at present whether this insulin action is a direct or an indirect effect. To investigate whether leptin concentrations in children and adolescents with type 1 diabetes (IDDM) were related to metabolic status, body weight, body mass index and insulin treatment, we have measured leptin concentrations in serum from 13 newly diagnosed IDDM patients before the beginning of insulin treatment (8 girls, 5 boys, aged 4.7-17.5 years) and in 134 patients with IDDM during treatment (64 girls, 70 boys, aged 2.6-20.1 years) using a specific radioimmunoassay. The data from patients with diabetes were compared with normative data that were derived from a large cohort of healthy children and adolescents. Serum from children with newly diagnosed diabetes had significantly lower levels of leptin (mean 1.28+/-1.60 ng/ml, range 0.14-6.13 ng/ml) compared with healthy children (n=710) (mean 2.2 ng/ml, range 0.26-14.4ng/ml) and compared with insulin-treated children and adolescents (mean 5.18+/-5.48 ng/ml, range 0.26-29.77 ng/ml) (P<0.0001) even after adjustment for gender and body mass index (BMI). Serum leptin levels in patients with IDDM were significantly correlated with BMI (r=0.42, P<0.0001). Multiple regression analysis showed that age and BMI were significantly correlated with leptin levels, while duration of diabetes, mean HbA1c levels, insulin dose and plasma glucose, triglyceride and cholesterol levels were not. Females had higher serum leptin concentrations than males even when adjusted for BMI (P<0.0001). Surprisingly and most importantly, leptin levels in insulin-treated young adult (Tanner stage 5) patients were significantly higher than values found in the healthy nondiabetic reference population when adjusted for sex, Tanner stage and BMI. These findings suggest that leptin levels in IDDM patients show a similar dependency on adipose tissue and age as in healthy, normal children. The data provide evidence that insulin may be of importance as a regulator of serum leptin levels in vivo not only in rodents but also in humans. It is hypothesized that the elevated BMI-adjusted leptin levels in adolescents with IDDM could indicate either that these patients may be oversubstituted by the intensified insulin therapy that they are receiving or that their body composition and body fat content may differ from that of healthy adolescents in the sense that they have a relative increase in fat mass.
Kaspar Sørensen and Anders Juul
Early pubertal timing is consistently associated with increased BMI percentile-for-age in pubertal girls, while data in boys are more ambiguous. However, higher BMI percentile-for-age may be a result of the earlier puberty per se rather than vice versa. The aim was to evaluate markers of adiposity in relation to pubertal timing and reproductive hormone levels in healthy pubertal boys and girls.
Population-based cross-sectional study (The Copenhagen Puberty Study). Eight-hundred and two healthy Caucasian children and adolescents (486 girls) aged 8.5–16.5 years participated. BMI and bioelectric impedance analyses (BIA) were used to estimate adiposity. Clinical pubertal markers (Tanner stages and testicular volume) were evaluated. LH, FSH, estradiol, testosterone, SHBG and IGF1 levels were determined by immunoassays.
In all age groups, higher BMI (all 1 year age-groups, P≤0.041) was found with early compared with late maturation, despite similar BIA–estimated body fat percentage (BIA–BF%). Neither BMI nor BIA–BF% differed for a given stage of maturation. BMI percentile-for-age and prevalence of overweight/obesity were higher in the early compared with late matured pubertal children (all P≤0.038), despite similar BIA–BF%. Pubertal girls with BIA–BF >29% had significantly lower LH and FSH levels compared with normal-weight girls (P≤0.041).
Early maturational timing was not associated with higher adiposity for a given stage of puberty. Using BMI percentile-for-age overestimated the degree of adiposity in early pubertal compared with late pubertal children.
Ponce Cedric Fouejeu Wamba, Jie Mi, Xiao-Yuan Zhao, Mei-Xian Zhang, Yu Wen, Hong Cheng, Dong-Qing Hou, and Katherine Cianflone
Childhood obesity is increasing worldwide and is increasingly associated with metabolic syndrome (MetS). Our aim was to examine acylation stimulating protein (ASP) and its precursor complement C3, in normal, overweight, and obese Chinese children and adolescents, and the relationships with body size, blood parameters, pubertal development, family environment, and MetS.
Children and adolescents (n=1603) from 6 to 18 years, boys (n=873) and girls (n=730), including normal weight (n=603), overweight (n=291) and obese (n=709) were assessed for body size parameters, pubertal development, blood lipids, glucose, insulin, ASP, and C3.
ASP levels were increased in overweight and obese versus normal weight (P<0.001), while C3 showed little variation. This effect of overweight/obesity remained throughout early stages when boys and girls were separated by pubertal development or age, although age and pubertal status itself had no effect. Separation based on ASP quintiles demonstrated significant associations with blood cholesterol, triglyceride, low-density lipoprotein cholesterol (LDL-Chol), glucose, insulin, and homeostatic model assessment of insulin resistance in boys, and LDL-Chol, high-density lipoprotein cholesterol, and glucose in girls. A positive correlation with mother's body mass index in boys and girls (P=0.002 and P=0.014 respectively) as well as birth weight (P<0.001) was noted. MetS was strongly associated with increased ASP, the presence of a single MetS factor (especially hypertension, central obesity, or hyperglycemia) was associated with increased ASP.
Changes in the plasma adipokine ASP in early obesity are associated with blood lipid and glucose modifications, family environment, and distinct MetS risk factors.
Celia Aradillas-García, Martha Rodríguez-Morán, María Eugenia Garay-Sevilla, Juan Manuel Malacara, Ramón Alberto Rascon-Pacheco, and Fernando Guerrero-Romero
Several cutoff points of the homeostasis model assessment of insulin resistance (HOMA-IR; varying from 2.5 to 4.0) have been suggested for diagnosing IR in youth. In this study, we determined the distribution of the HOMA-IR in Mexican children and adolescents.
Design and methods
A total of 6132 children and adolescents from San Luis Potosi, León, Queretaro, and Durango, which are cities in central and northern Mexico, were enrolled in a population-based cross-sectional study. Eligible participants were apparently healthy children and adolescents aged 6–18 years. Pregnancy and the presence of chronic illnesses were exclusion criteria.
A total of 3701 (60.3%) girls and 2431 (39.7%) boys were included in this study. In the overall population, the mean body mass index, insulin levels, and fasting glucose levels were 21.8±1.3 kg/m2, 7.1±3.2 μU/ml, and 86.2±10.0 mg/dl respectively. The concentrations of insulin and fasting glucose gradually increased from 6 to 12 years of age, whereas the concentrations tended to plateau in the 13- to 18-year-old population. The absolute mean of the HOMA-IR was 2.89±0.7. The HOMA-IR gradually increased with age and reached a plateau at 13 years of age.
Because the insulin concentrations, glucose levels, and HOMA-IR exhibited a gradual increase with age that was not related to obesity, our results suggested that the evaluation of IR in children should be based on percentiles of the HOMA-IR rather than a dichotomous value derived from a single cutoff point.