Corticosteroid-induced osteoporosis in children: outcome after two-year follow-up, risk factors, densitometric predictive cut-off values for vertebral fractures

OBJECTIVE

To identify factors that contribute to a decreased Z score of volumetric spine bone mineral density (ZvSBMD) and the development of vertebral fractures (VF) in children receiving chronic systemic corticosteroid therapy (SCT); to describe their outcome after 2 years, and to define predictive threshold values for ZvSBMD for VF.

METHODS

Fifty-five children on SCT for >or= 6 months were prospectively followed for 2 years. In children with a ZvSBMD > -1.5, we prescribed preventive measures for osteoporosis and densitometry annually. In children with ZvSBMD <or= -1.5, we prescribed spine x-rays and those with VF received alendronate. The association between clinical and biochemical variables and the presence of VF or ZvSBMD were analyzed by logistic regression or multiple regression analysis. The threshold value of ZvSBMD for predicting VF was determined by ROC curve and the probability of having a VF was modeled by multiple logistic regressions.

RESULTS

Children who do not develop osteoporosis at first evaluation tend to maintain normal ZvSBMD after two years. Alendronate increased ZvSBM (median: at baseline: -2.69; 1 yr: -1.92; 2 years: -1.39, p < 0.001). The threshold value of ZvSBMD for predicting VF was -1.8. In this cohort, the risk of developing VF was significantly higher in children who were not ambulatory, growth retarded, treated with methotrexate for a longer time, had a family history of osteoporosis or were of non-aboriginal ancestry.

CONCLUSION

Children on SCT, who do not develop osteoporosis, tend to maintain normal BMD. Children who were not ambulatory, on methotrexate or growth retarded have higher rates of VF.

Insulin gene VNTR genotype is associated with insulin sensitivity and secretion in infancy

AIMS

We have previously demonstrated that insulin sensitivity and secretion at age 1 year was in part related to variation in weight and height gain during infancy. In order to determine whether genetic variation at the insulin gene could also influence these associations, we have studied the relationship between insulin gene variable number of tandem repeat (INS VNTR) genotypes, insulin secretion and early postnatal growth.

METHODS

We assessed fasting and dynamic insulin secretion in 99 healthy infants at age 1 year, using a short intravenous glucose tolerance test (sIVGTT). Infants were genotyped at the -23 HphI locus, as a surrogate marker for INS VNTR allele classes I and III. Anthropometric data were recorded at birth and at 1 year. Data are shown as median (interquartile range).

RESULTS

Fasting insulin levels were higher in III/III infants (n = 9) than in I/I infants [n = 55; 27.4 (17.6-75.6) pmol/l vs. 18.1 (10.3-25.2) pmol/l; P < 0.05]. Insulin secretion during the sIVGTT, as estimated by the serum insulin area under the curve, was also higher in III/III infants [2417 (891-4041) pmol min/l vs. 1208 (592-2284) pmol min/l; P < 0.05]. Fasting and postload plasma glucose levels were similar in both groups. Analysis of covariance showed that genotype differences in fasting insulin sensitivity and insulin secretion were independent of size at birth, postnatal growth velocity and current body mass index.

CONCLUSIONS

Significant associations between INS VNTR genotype and both insulin sensitivity and secretion were apparent in infancy; these might interact with childhood appetite and nutrition to impact the development of childhood obesity and insulin resistance.

Study of GH sensitivity in chilean patients with idiopathic short stature

We hypothesized that some children with idiopathic short stature in Chile might bear heterozygous mutations of the GH receptor. We selected 26 patients (3 females, 23 males) from 112 patients who consulted for idiopathic short stature at the University of Chile. Their chronological age was 8.3 +/- 1.9, and bone age was 6.1 +/- 1.0 yr. Their height was -3.0 +/- 0.7 SDS; IGF-I, -1.2 +/- 1.1 SD; IGF binding protein 3, -0.7 +/- 2.0 SDS; and GH binding protein, 0.4 +/- 0.8 SDS. Patients were admitted, and blood samples were obtained every 20 min to determine GH concentrations overnight. Coding sequences and intron-exon boundaries of exons 2-10 of GH receptor gene were amplified by PCR and subsequently analyzed through single-strand conformational analysis. Mean serum GH concentration, over 12-h, was 0.20 +/- 0.08 nM; pulse amplitude, 0.40 +/- 0.15 nM; number of peaks, 5.8 +/-1.5 peaks/12 h; peak value of GH during the 12-h sampling, 1.03 +/- 0.53 nM; and area under the curve, 151.4 +/- 56.1 nM/12 h. There were positive correlations between mean GH vs. area under the curve (P < 0.001) and GH peak (P < 0.01). The single-strand conformational analysis of the GH receptor gene showed abnormal migration for exon 6 in 9 patients and for exon 10 in 9 patients, which (by sequence analysis) corresponded to 2 polymorphisms of the GH receptor gene: an A-to-G transition in third position of codon 168 in exon 6 and a C-to-A transversion in the first position of codon 526 in exon 10. We further sequenced all coding exons and intron-exon boundaries in the most affected patients (nos. 6, 9, 11, 14, 15, 16, and 23). This analysis revealed a C-to-T transition in codon 161 of exon 6 in patient 23, which results in an amino acid change (Arg to Cys) in an heterozygous form in the patient and his father. In conclusion, the results of our study suggest that, in Chilean patients with idiopathic short stature, GH receptor gene mutations are uncommon, although we cannot exclude mutations that were missed by single-strand conformational analysis or mutations within introns or in the promoter regions of the GH receptor gene.

Effects of oral administration of ibutamoren mesylate, a nonpeptide growth hormone secretagogue, on the growth hormone-insulin-like growth factor I axis in growth hormone-deficient children

Ibutamoren mesylate (MK-0677), an orally active nonpeptide growth hormone (GH) secretagogue, stimulates GH release through a pituitary and hypothalamic receptor that is different from the GH-releasing hormone receptor. We evaluated the safety and tolerability and the GH-insulin-like growth factor (IGF) responses to two dosages of oral ibutamoren mesylate given to children with GH deficiency for 7 to 8 days. The patients, 18 prepubertal children (15 male, 3 female) with idiopathic GH deficiency, had a chronologic age of 10.6 +/- 0.8 years (mean +/- SD), bone age of 7.4 +/- 0.7 years, growth velocity < 10th percentile for age, height < 10th percentile for age, and a maximum GH response of < or = 10 microg/L to two different GH stimulation tests. The children were assigned as follows to one of three treatment groups with ibutamoren mesylate: 0.2 mg/kg per day for 7 days (days 1-7 or 8-14) and matching placebo for the alternate 7 days (groups I and II, respectively) or 0.8 mg/kg per day for 7 days (days 8-14, group III). On day 15 all patients received an 0.8-mg/kg dose of ibutamoren mesylate. Patients in groups I and II were studied first to assess safety at the low dose before advancement to the high dose. Hormonal profiles were evaluated on day -1 (baseline) and day 15, and the results were expressed as the change from baseline within each group. After administration of ibutamoren mesylate 0.8 mg/kg for 8 days (group III), the median increases (on day 15) from baseline were as follows: 3.8 microg/L (range, 0 to 34.3) for serum GH peak concentration (P = .001), 4.3 microg x h/L (range, 1.3 to 35.6) for the GH area under the concentration-time curve from time zero to 8 hours (AUC(0-8)) (P < .001), 12 microg/L (range, -4 to 116) for serum IGF-I (P = .01), and 0.4 microg/L (range, -0.9 to 1.5) for serum IGF-binding protein 3 (IGFBP-3) (P = .01). There was no change in serum prolactin, glucose, triiodothyronine, thyroxine, thyrotropin, peak serum cortisol, and insulin concentrations or 24-hour urinary free cortisol after administration of 0.8 mg/kg per day of ibutamoren mesylate for 8 days. We conclude that short-term administration of ibutamoren mesylate can increase GH, IGF-I, and IGFBP-3 levels in some children with GH deficiency. Thus this compound is applicable for testing its effect on growth velocity.

Relationship between serum growth hormone binding protein levels and height in young men

The biochemical mediators responsible for variations in stature among normal subjects are largely unknown. To obtain some initial information about potential endocrine factors, we measured the serum concentrations of GH, IGF-1, IGFBP-3 and GHBP in healthy young men shorter than 159 cm and taller than 187 cm. We studied 14 volleyball and basketball players (tall group), and 14 jockey students from a horse racetrack (short group). A careful medical history was taken, including dietary intake, and physical examination with special attention to the possible presence of genetic stigmata was performed. Serum prealbumin was determined as an index of nutritional status. A buccal smear was performed to exclude Klinefelter's syndrome. The BMI and serum prealbumin levels were comparable in both groups of individuals. The nutritional survey, however, revealed that the tall subjects had a higher intake of calories (42.2+/-11.2 vs. 30.1+/-15.15 kcal/kg, p<0.05), and protein (1.5+/-0.6 vs. 0.8+/-0.4 mg/kg, p<0.01). Serum concentrations of GHBP did not differ in the two groups (0.95+/-0.37 nmol/l in the tall, and 0.95+/-0.53 nmol/l in the short group), and did not correlate with height, serum IGF-I levels, or BMI. We observed a significant difference in the serum concentrations of IGF-I in the two groups of individuals (42.02+/-9.37 nmol/l in the tall and 31.79+/-3.18 nmol/l in the short group, p<0.05), and this growth factor showed a positive correlation with height (r = 0.5, p<0.01). These preliminary findings suggest that final height differences in young men do not appear to be mediated by variations in GHBP concentrations.

Near final height in pubertal growth hormone (GH)-deficient patients treated with GH alone or in combination with luteinizing hormone-releasing hormone analog: results of a prospective, randomized trial

To study the effects of delaying puberty in GH-deficient (GHD) children, we studied 21 GHD (9 boys, 14 girls), treatment-naive, pubertal patients in a prospective, randomized trial. Their chronological age was 14.3 +/- 1.6 yr, and their bone age was 11.3 +/- 1.1 yr (mean +/- SD) at the beginning of the study. Four patients who developed hypogonadotropic hypogonadism were subsequently excluded from the study. Patients were randomly assigned to receive GH + LH-releasing hormone analog (LHRH-A) (n = 7), or GH alone (n = 10). GH and LHRH-A treatment started simultaneously in each patient. GH (Nutropin) was administered at a dose of 0.1 U/kg x day sc, until patients reached a bone age (BA) of 14 yr in girls and 16 yr in boys, and LHRH-A (Lupron depot) was administered at a dose of 300 microg/ kg every 28 days in during 3 yr. We defined GH deficiency as patients with a growth velocity less than 4 cm/yr, BA delay more than 1 yr in relationship to chronological age, GH response to two stimulation tests less than 7 microg/L, associated with low serum insulin-like growth factor I and insulin-like growth factor binding protein 3 levels. Statistical analysis was performed by ANOVA or Kruskall Wallis when variances were not homogeneous. We observed a significant decrease in the rate of BA maturation in the group treated with GH+LHRH-A (1.5 +/- 0.2 yr) compared with the group treated with GH alone (4.2 +/-0.5 yr) during the 3 years of LHRH-A therapy (P < 0.05). This delay in BA maturation produced a significant gain in final height in the group treated with GH+LHRH-A, which reached - 1.3 +/- 0.5 SD score compared with -2.7 +/- 0.3 SD score (P < 0.05) in the group treated with GH alone. These results indicate that delaying puberty with LHRH-A in GHD children during treatment with GH increases final height.

Autosomal dominant hypoparathyroidism associated with short stature and premature osteoarthritis

Familial hypoparathyroidism is an unusual and genetically heterogeneous group of disorders that may be isolated or may be associated with congenital or acquired abnormalities in other organs or glands. We have evaluated a family with a novel syndrome of autosomal dominant hypoparathyroidism, short stature, and premature osteoarthritis. A 74-yr-old female (generation I) presented with hypoparathyroidism, a movement disorder secondary to ectopic calcification of the cerebellum and basal ganglia, and a history of knee and hip replacements for osteoarthritis. Two members of generation II and one member of generation III were also documented with hypoparathyroidism, short stature, and premature osteoarthritis evident as early as 11 yr. Because of the known association between autosomal dominant hypoparathyroidism and activating mutations of the calcium-sensing receptor (CaR) gene, further studies were performed. Sequencing of PCR-amplified genomic DNA revealed a leucine to valine substitution at position 616 in the first transmembrane domain of the CaR, which cosegregated with the disorder. However, this amino acid sequence change did not affect the total accumulation of inositol phosphates as a function of extracellular calcium concentrations in transfected HEK-293 cells. In conclusion, a sequence alteration in the coding region of the CaR gene was identified, but is not conclusively involved in the etiology of this novel syndrome. The cosegregation of hypoparathyroidism, short stature, and osteoarthritis in this kindred does suggest a genetic abnormality involving a common molecular mechanism in parathyroid, bone, and cartilage.

High fluid intake increases urine free cortisol excretion in normal subjects

To test the hypothesis that increased fluid intake increases the urine free cortisol, we prescribed 5 liters of fluid intake per day or a normal fluid intake according to a randomized cross-over design in six normal volunteers. Each period lasted 5 days, with a 2-day washout period of normal fluid intake between the two periods. Urine free cortisol, 17-hydroxycorticosteroids, and creatinine were measured daily during each study period, and the average value over each 5-day period was calculated for each subject. High fluid intake caused a significant increase in the mean urine free cortisol [126 +/- 33 (SD) vs. 77 +/- 18 micrograms/day, P < 0.005]. The frequency of urine free cortisol results that exceeded the upper normal limit of 95 micrograms/day was also much higher during high fluid intake (23/30 vs. 6/30, P < 0.005). By contrast, urine 17-hydroxycorticosteroids (high fluid vs. normal fluid: 5.3 +/- 1.5 vs. 5.0 +/- 1.7 mg/day, respectively, P = not significant) and urine creatinine (1.51 +/- 0.48 vs. 1.45 +/- 0.37 g/day, P = not significant) did not differ between the two study periods. We conclude that high fluid intake (5 liters/day) increases free cortisol excretion without an increase in urine 17-hydroxycorticosteroids. Thus, mild to moderate increases in urine cortisol excretion may not indicate hypercortisolism in individuals who have a high fluid intake and urine volume.