Effect of growth hormone therapy and puberty on bone and body composition in children with idiopathic short stature and growth hormone deficiency

Bone 37 (2005) 642 – 650 www.elsevier.com/locate/bone

Effect of growth hormone therapy and puberty on bone and body composition in children with idiopathic short stature and growth hormone deficiency Wolfgang Höglera,b , Julie Briodyc , Bin Moorea , Pei Wen Lua , Christopher T. Cowella,d,⁎ a

Institute of Endocrinology and Diabetes, The Children’s Hospital at Westmead, Sydney, Australia b Dept. of Paediatrics and Adolescent Medicine, Medical University Innsbruck, Austria c Dept. of Nuclear Medicine, The Children’s Hospital at Westmead, Sydney, Australia d Discipline of Paediatrics and Child Health, University of Sydney, Australia Received 3 March 2005; revised 6 June 2005; accepted 13 June 2005 Available online 1 September 2005

Abstract The state of bone health and the effect of growth hormone (GH) therapy on bone and body composition in children with idiopathic short stature (ISS) are largely unknown. A direct role of GH deficiency (GHD) on bone density is controversial. Using dual-energy X-ray absorptiometry, this study measured total body bone mineral content (TB BMC), body composition, and volumetric bone mineral density (vBMD) at the lumbar spine (LS) and femoral neck (FN) in 77 children (aged 3–17 years) with ISS (n = 57) and GHD (n = 20). Fifty-five children (GHD = 13) receiving GH were followed over 24 months including measurement of bone turnover. At diagnosis, size-corrected TB BMC SDS was greater (P ≤ 0.002) and LSvBMD SDS lower (P b 0.03) than zero in both prepubertal ISS and GHD subjects, but FNvBMD SDS was reduced only in the GHD group (P b 0.05). The muscle-bone relation, as assessed by the BMC/lean mass (LTM) ratio SDS was not different between groups. During GH therapy, prepubertal GHD children gained more height (1.58 [0.9] SDS) and LTM (0.87 [0.63] SDS) compared to prepubertal ISS children (0.75 [0.27] and 0.17 [0.25] SDS, respectively). Percent body fat decreased in GHD (−5.94% [4.29]) but not in ISS children. Total body BMC accrual was less than predicted in all groups accompanied by an increase in bone turnover. Puberty led to the greatest absolute, but not relative, increments in weight, LTM, BMI, bone mass, and LSvBMD. Our results show that children with ISS and GHD differ in their response to GH therapy in anthropometry, body composition, and bone measures. Despite low vBMD values at diagnosis in both prepubertal groups, size-corrected regional or TB bone data were generally within the normal range and did not increase during GH therapy in GHD or ISS children. Growth hormone had great effects on the growth plate and body composition with subsequent gains in height, LTM, bone turnover, and bone mass accrual, but no benefit for volumetric bone density over 2 years. Crown Copyright © 2005 Published by Elsevier Inc. All rights reserved. Keywords: Idiopathic short stature; Growth hormone deficiency; Osteopenia; Bone density; Children; Body composition

Introduction Idiopathic short stature (ISS) represents a heterogeneous group of short children in whom the cause of decreased childhood growth cannot be identified. Treatment with ⁎ Corresponding author. Institute of Endocrinology and Diabetes, The Children’s Hospital at Westmead, Locked Bag 4001, Westmead, NSW 2145, Sydney, Australia. Fax: +61 2 9745 3170. E-mail address: [email protected] (C.T. Cowell).

growth hormone (GH) in ISS results in a small but significant gain in final height [1–3]. Apart from one longitudinal study which reported low areal bone mineral density (BMD) at the lumbar spine in a small ISS cohort which normalized during GH therapy [4], no studies have so far addressed bone health and body composition in ISS children. In contrast, numerous studies have assessed the effect of GH on bone density and body composition in children and adults with GH deficiency (GHD). Both GH and IGF-

8756-3282/$ - see front matter. Crown Copyright © 2005 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.06.012

W. Högler et al. / Bone 37 (2005) 642–650

I have a well-recognized role in bone elongation and skeletal maturation in vitro and in vivo [5–8]. However, a direct role of GH for mineral accrual and bone density is controversial. Adults with severe, untreated isolated childhood-onset GHD [9,10] or complete GH insensitivity [11,12], as well as children with partial GH insensitivity [13] have been reported with normal size-corrected BMD and no increased number of fractures [10,14]. In addition, bone health is mainly assessed by dual-energy X-ray absorptiometry (DXA), a two-dimensional technique prone to size-dependent artifacts in children, particularly with short stature. Adjustments for bone or body size are thus essential in the interpretation of DXA data [15–17]. The increase in bone mineral content (BMC) or areal BMD z scores observed during GH treatment [18] is mostly caused by the GH-induced gain in height, but also in muscle strength and mass [19] which independently increases bone strength [20]. Lean mass [21,22], muscle size [23], and, to a lesser degree, estimated volumetric BMD (g/cm3) are low at diagnosis of GHD [21,22,24] but increase during GH therapy [13,21, 22,25]. A decreased accrual in bone mass [26,27] and lean mass [28–30] is seen at withdrawal of GH therapy during transition from adolescence to young adulthood, which can be reversed by retreatment with GH [27,31], suggesting a positive role of GH on bone health. Some previous studies are biased by inappropriate sizecorrection for DXA variables but also by not accommodating for the effect of puberty on bone mass, which may result in erroneous interpretation of DXA data if groups are inappropriately matched for pubertal stage. The present longitudinal study compares cohorts of children and adolescents with ISS and GHD with the main purposes (1) to differentiate the effects of GH therapy and puberty on bone and muscle mass, body composition, anthropometry, and bone turnover, and (2) to examine total body and regional bone scans by applying size-corrections to the areal DXA output with special consideration to the muscle–bone relation at the total body.

Subjects and methods The study cohort comprised 77 patients (mean [SD] age 10.90 [3.12] years, 25 girls) with short stature, seen by endocrinologists at The Children's Hospital at Westmead, Sydney, Australia. The assessment of short stature involved at least one GH stimulation test. Subjects were classified as having GHD if peak GH was b10 ng/ml following 2 stimulation tests and ISS if peak GH response to stimulation was N10 ng/ml. Other causes of short stature were excluded by clinical and biochemical examination. Six individuals of the GHD group were treated with thyroxine for TSH deficiency and remained euthyroid throughout the study. Three individuals in the GHD group received Hydrocortisone replacement in a dose of 6–8 mg/m2/day. All patients had

643

a growth velocity b25th centile for age and fulfilled the Australian criteria for GH treatment. Bone densitometry, anthropometry, bone age, pubertal stages, and bone markers were assessed before the commencement of GH (baseline) and after 6, 12, and 24 months. The mean (SD) GH dose at onset was 5.23 (2.74) mg/m2/week [0.026 (0.007) mg/kg/ day]. This lower dose, in comparison to the higher doses currently recommended, was used as this longitudinal study was performed in the 1990s. The patients were seen at 3 monthly intervals for clinical assessment. Adjustment of GH dose occurred every 6 months to maintain catch-up growth. The GH dose increased significantly during the study period to 6.37 (2.78) mg/m2/week [0.027 (0.008) mg/kg/ day]. The increase was not different between groups. At commencement of GH therapy, 52 subjects (GHD, n = 20) were prepubertal and 25 pubertal (all ISS). Twenty patients (GHD, n = 5) dropped out of the study or did not attend the 2-year visit. Their 1st year growth velocity, baseline anthropometric z scores, and growth hormone dose were not different from the longitudinal study population. At the 2-year visit, two of the prepubertal GHD group and 13 of the prepubertal ISS group had entered puberty. The pubertal GHD group (n = 2) was excluded from longitudinal analysis. Therefore, complete longitudinal 2-year data sets were available for 55 patients (16 girls). These patients were then categorized into “remaining prepubertal” (ISS = 13, GHD = 13) and “pubertal”, meaning having gone into puberty or remaining pubertal (n = 29, all ISS). The Institutional Ethics Committee approved the study and informed consent was obtained from all subjects and families. Methods Anthropometry Height was measured with a Harpenden stadiometer to the nearest 1 mm and weight with an electronic scale to the nearest 20 g on the day of the DXA scan. Height and weight z scores (SD scores) were calculated according to sex and age [32]. For subjects over 18 years, height and weight z scores were calculated as for an 18-year-old. Sex-specific body mass index (BMI in kg/m2) z scores for age were calculated from data derived from Cronk and Roche [33]. Pubertal stages were assessed according to Tanner [34]. Densitometry A pencil beam DPX (Lunar Corp, Madison, WI) total body scanner with adult software (version 3.4) was used to perform DXA measurements on all subjects. Analysis was done with software version 4.7 by a single technician. The technique and measurement protocol, including qualityassurance testing, has been described previously [35,36]. In brief, the coefficients of variation for total body (TB) BMC, lean tissue mass (LTM), and percent fat mass were 0.74, 0.82, and 1.59%, respectively. Total body measurements were compared to our normative data set of healthy females

644

W. Högler et al. / Bone 37 (2005) 642–650

(n = 249) and males (n = 227) aged 3–30 years and analyzed according to a 4-step procedure recently published [16]. This procedure first involves comparison of z scores for BMC and height to account for the strong body-size relation of DXA output, and then examines the muscle– bone relation using z scores for LTM and the BMC/LTM ratio. We slightly modified the procedure in this study by exclusively reporting z scores related to age for DXA measures. Using z scores related to height in subjects with short stature, particularly after puberty, can lead to undesired comparisons with much younger, prepubertal control subjects. The adjustment for body size in the first 2 steps was done by sex-specific equations predicting TB BMC z score by height z score. Data were derived from our normative data set (girls: TB BMC z score = −0.032 + 0.738 × height z score [r2 = 0.46, P b 0.001]; boys: TB BMC z score = 0.070 + 0.639 × height z score [r2 = 0.53, P b 0.001]). Regional BMC and estimated volumetric BMD (vBMD) were derived from LS (L1–L4) and FN scans performed on the same DPX [36]. For comparison reasons, we used our normative database on healthy individuals (3–30 years) for the LS (180 females, 187 males) and the FN scans (210 females, 181 males), which is an expanded and updated version of the data set previously published [36]. Results for DXA variables are reported as age z scores and as raw data. Markers of bone metabolism Total serum calcium, inorganic phosphorus, osteocalcin (OC), serum alkaline phosphatase (ALP), and IGF-I were measured at baseline and after 6, 12, and 24 months of GH treatment. Calcium, inorganic phosphorus, and ALP were analyzed using routine laboratory methodology. Intact OC was measured by a solid-phase, two-site, chemiluminescent enzyme immunometric assay (Immulite Analyzer, Diagnostics Products Corporation, Los Angeles, USA) and IGF-I by radioimmunoassay (Bioclone Australia Pty. Ltd., Sydney, AUS). Statistics Data were analyzed using the Statistical Package for Social Sciences (SPSS Inc, Chicago, IL, USA, version 11.0). Three groups (prepubertal GHD, prepubertal, and pubertal ISS) were derived from the baseline crosssectional (n = 77) and the 2-year longitudinal (n = 55) data sets. Differences between the three groups were assessed by one-way analysis of variance (ANOVA). Post hoc comparisons among groups were assessed by the Bonferroni test. The within-group change between baseline and 2 years was assessed by paired t tests or Wilcoxon signed rank tests. The absolute change in anthropometric and densitometric variables over the 2-year period was also calculated. The difference in this change between the three groups was assessed by ANOVA and post hoc Bonferroni test. Where applicable, the difference in SDS from zero was

assessed by a one-sample t test. For bone marker analysis, the within-group difference in concentration from baseline at each time point was assessed by Wilcoxon signed rank test. Data are presented as means (SD) and P values b 0.05 were considered significant.

Results Characteristics of the cross-sectional population at baseline Table 1 shows the characteristics of the study population at baseline. There were no significant differences in SD scores for height, weight, TB BMC, and LTM between the three groups. However, the prepubertal GHD group had greater percent body fat (P = 0.002) and BMI SDS (P = 0.028) than the prepubertal ISS group. As a function of body size, BMC and LTM SDS were below zero in all groups (P b 0.001), but significantly greater than the corresponding height SDS (P b 0.05, independent sample t test). The observed TB BMC SDS was greater than predicted (zero) in both prepubertal groups (P ≤ 0.002) with greatest values in the GHD group (P = 0.01). No subject had a sizecorrected TB BMC b −2 SD. The muscle–bone relation, as assessed by the BMC/LTM ratio SDS was not different between groups and below zero (P b 0.05) in all groups. Lumbar spine vBMD SDS was below zero in both prepubertal groups (P b 0.03) and FN vBMD SDS in the GHD group (P b 0.05). A LS vBMD of b −2 SD was found in 3/18 (17%), 3/30 (10%), and 1/24 (4%) in the GHD, prepubertal, and pubertal ISS groups, respectively. Respective numbers (percentages) for FN vBMD were 1/19 (5%), 2/ 29 (7%), and 0/25 (0%). Changes in anthropometry and body composition during 2 years of GH treatment A significant increase in height SDS and weight SDS was observed in each group without changes in BMI SDS (Table 2). The GHD group had the greatest gain in height SDS (P ≤ 0.003) compared with both ISS groups and a greater gain in weight SDS than the pubertal ISS group (P = 0.031). Percent body fat decreased in the GHD group (P b 0.001), which differed significantly from both ISS groups (P ≤ 0.009) where no change from baseline was seen. Similar to the changes in height SDS, LTM SDS increased in all groups (P b 0.03) with the greater gain being observed in the GHD group compared to the prepubertal ISS group (P = 0.003) (Table 2). Changes in TB bone mass and the muscle–bone relation during GH treatment The increase in TB BMC of the pubertal ISS group (496.70 [186.56] g) was nearly twice that of the prepubertal GHD (290.32 [68.24] g) and ISS groups (251.51 [90.99] g). Total

W. Högler et al. / Bone 37 (2005) 642–650

645

Table 1 Characteristics of the cross-sectional study population at baseline (n = 77) with differences between groups (ANOVA)

documented by a decreasing BMC/LTM ratio SDS in all groups (P b 0.05) (Table 2).

Baseline

GHD prepubertal, n = 20

ISS prepubertal, n = 32

ISS pubertal, n = 25

Difference between groups, P value

Changes in regional bone mass and density during GH treatment

Age (years)

9.42 (3.73)

9.78 (2.30)

13.50 (1.52)

b0.001 a

119.53 (11.26) −2.77 (0.60) b 22.96 (5.65) −1.92 (0.56) b 15.82 (1.74) −0.88 (0.84) b

138.68 (6.54) −2.80 (0.73) b 35.60 (8.08) −1.68 (1.01) b 18.40 (3.20) −0.74 (1.13) e

b0.001 a

Similar to the TB results, the increase in LS BMC in the pubertal ISS group by far exceeded the increase in both other groups. Lumbar BMC SDS increased significantly from baseline in the GHD group

0.182 b0.001 c 0.342 0.001 d 0.028 f

Table 2 The change from baseline in anthropometric and densitometric variables after 2 years of GH therapy with differences between groups (ANOVA)

812.60 (245.70) −1.51 (0.72) b 0.32 (0.53) e

1319.82 b0.001 (310.35) −1.78 (1.03) b 0.257 0.09 (0.68) 0.01 g

a

18.08 (4.38) 27.12 (6.02) b0.001 −1.89 (0.73) b −2.24 (1.03) b 0.322 −0.93 (0.89) b −1.09 (0.93) b 0.188

a

Anthropometry Height (cm) 116.32 (19.88) Height SDS −3.18 (1.18) b Weight (kg) 26.67 (13.92) Weight SDS −1.53 (1.29) b 18.36 (3.94) BMI (kg/m2) BMI SDS −0.01 (1.16) Total body (TB) BMC (g) 826.14 (433.06) BMC SDS −1.38 (0.70) b Obs-pred 0.63 (0.50) b BMC SDS ⁎ LTM (kg) 17.67 (7.18) LTM SDS −2.06 (0.85) b BMC/LTM −0.56 (1.17) h SDS Percent fat 23.77 (11.72) (%) Lumbar spine (LS) BMC (g) 12.76 (5.77) BMC SDS −1.80 (0.98) b vBMD 0.26 (0.05) (g/cm3) vBMD SDS −0.72 (1.24) h

13.75 (6.77)

16.94 (11.04) 0.002 f

13.25 (3.83) 20.79 (4.71) b0.001 −1.48 (0.65) b −2.06 (0.70) b 0.026 i 0.27 (0.04) 0.31 (0.03) 0.001 c −0.62 (1.09) e −0.16 (1.00)

Femoral neck (FN) BMC (g) 1.84 (0.95) 1.58 (0.53) 0.50 (0.15) 0.49 (0.18) vBMD (g/cm3) vBMD SDS −0.49 (0.91) h −0.25 (0.98)

a

0.196

2.60 (0.88) 0.66 (0.16)

b0.001 c b0.001 c

0.21 (0.90)

0.046

P b 0.001, pubertal ISS group different from both other groups. P b 0.001 compared to zero. c P b 0.01, pubertal ISS group different from both other groups. d P b 0.01, prepubertal ISS group different from both other groups. e P b 0.01 compared to zero. f P b 0.03, GHD group different from the prepubertal ISS group. g P b 0.01, GHD group different from the pubertal ISS group. h P b 0.05 compared to zero. i P b 0.05, prepubertal ISS group different from pubertal ISS group. ⁎ The observed total body BMC SDS minus that predicted by height SDS is given (for equations, see Methods). a

b

Anthropometry Height (cm) Height SDS Weight (kg) Weight SDS BMI (kg/m2) BMI SDS

ISS prepubertal, n = 13

ISS pubertal, n = 29

Difference between groups, P value

18.48 (4.07) 1.58 (0.90) b 7.76 (3.07) 0.85 (0.76) e 0.13 (1.29) −0.22 (0.49)

13.61 (1.53) 0.75 (0.27) b 7.22 (2.11) 0.56 (0.35) b 1.15 (1.05) −0.02 (0.34)

16.42 (3.07) 0.71 (0.54) b 11.1 (4.17) 0.39 (0.44) b 1.28 (1.28) 0.15 (0.53)

0.001 a b0.001 c 0.002 d 0.036 f 0.022 f 0.142

251.51 (90.99) 0.10 (0.23) −0.39 (0.17) b

496.70 b0.001 g (186.56) 0.11 (0.39) 0.241 −0.37 (0.41) b 0.043 f

Total body DXA BMC (g) 290.32 (68.24) BMC SDS 0.32 (0.49) h Diff obs-pred −0.69 (0.47) b BMC SDS ⁎ LTM (kg) 7.16 (2.02) LTM SDS 0.87 (0.63) b BMC/LTM −0.76 (0.57) b SDS Percent fat −5.94 (4.29) b (%) Lumbar spine (LS) BMC (g) 5.31 (1.96) BMC SDS 0.79 (0.52) e vBMD 0.01 (0.03) (g/cm3) vBMD SDS 0.11 (0.72) Femoral neck (FN) BMC (g) 0.41 (0.27) vBMD 0.02 (0.04) (g/cm3) vBMD SDS 0.19 (0.45)

5.44 (0.98) 9.67 (3.61) b0.001 g h b 0.17 (0.25) 0.59 (0.54) 0.004 i −0.49 (0.78) h −0.29 (0.63) h 0.106 1.33 (5.20)

−1.01 (4.67)

b0.001 c

3.59 (1.53) 0.02 (0.28) 0.01 (0.02)

11.08 (4.02) 0.25 (0.40) e 0.03 (0.02)

b0.001 g 0.001 c 0.007 d

0.06 (0.71)

0.14 (0.61)

0.943

0.38 (0.39) 0.02 (0.10)

0.67 (0.47) 0.00 (0.08)

0.134 0.664

−0.31 (0.51)

−0.04 (0.85)

0.345

P b 0.01, prepubertal ISS groups different from both other groups. P b 0.001 compared to zero. c P b 0.01, GHD group different from both other groups. d P b 0.05, pubertal ISS group different from both other groups. e P b 0.01 compared to zero. f P b 0.05, GHD group different from pubertal ISS group. g P b 0.001, pubertal ISS group different from both other groups. h P b 0.05 compared to zero. i P b 0.01, GHD group different from the prepubertal ISS group. ⁎ The difference in the size-corrected total body BMC SDS is given (for equations, see Methods). a

b

body BMC SDS increased in the GHD group (P = 0.037) and the pubertal (P = 0.05), but not the prepubertal (P = 0.116) ISS group. The relative gain (expressed as SDS) in TB BMC was lower than the relative gain in height (P b 0.03) and LTM (P b 0.05, t test). Thus, the difference between the observed and predicted TB BMC SDS decreased from baseline during the 2-year GH treatment in all groups (P b 0.001, Fig. 1A). The greater relative gain in LTM compared to BMC was also

GHD prepubertal, n = 13

646

W. Högler et al. / Bone 37 (2005) 642–650

months of GH therapy in any of the groups (Figs. 1B, C; Table 2). Change in bone marker concentrations There was no change from baseline in calcium concentrations in any of the three groups at any time point. However, inorganic phosphorus, OC, and ALP increased in all groups during the first 12 months (P b 0.05). At 24 months, the increase in OC and ALP persisted in all groups (P b 0.02) but inorganic phosphorus remained greater than baseline only in the GHD group. A significant increase in IGF-I (P b 0.05) from baseline was observed which persisted at all subsequent time points in all groups (Fig. 2).

Discussion

Fig. 1. Observed minus predicted TB BMC SDS (A) and volumetric BMD (g/cm3) at the LS (B) and the FN (C) before (gray boxes) and after 24 months of GH therapy (white boxes). Significant differences between baseline and 24 months are given. Boxes show median and interquartile range, whiskers 10th and 90th centiles and dots 5th and 95th centiles.

(P = 0.003) and the pubertal (P = 0.007), but not the prepubertal ISS group (P = 0.854). The LS BMC SDS increase in the GHD group was significantly greater than in the prepubertal and pubertal ISS groups (P b 0.01). Volumetric BMD at the LS increased only in the pubertal ISS group (P b 0.001). However, there was no significant change in LS and FN vBMD SDS during 24

Our results demonstrate that children with ISS and GHD differ in their response to GH treatment in anthropometry, body composition, and bone measures. Prepubertal GHD children gained more height and LTM (in SDS) compared to ISS children and decreased their percent fat mass during GH therapy. Puberty led to the expected greatest absolute, but not relative, increments in weight, LTM, BMI, bone mass, and LS vBMD. Volumetric BMD SDS values were lower than zero at baseline at the spine in both prepubertal groups. However, size-corrected regional or total body bone data were generally within the normal range and neither increased during GH therapy in GHD or ISS children over the 2-year period. Growth hormone had great effects on the growth plate and body composition with subsequent gains in height, LTM, bone markers, and bone mass accrual, but no benefit for volumetric bone density. Our longitudinal cohort represents the largest group of ISS children examined for bone health and body composition so far. Bone measures were corrected for size to avoid misinterpretation of the two-dimensional DXA output. Additionally, accounting for the impact of puberty on bone accrual, as demonstrated earlier [37], is essential in DXA data interpretation. Three-dimensional bone measures cannot be directly measured by DXA; however, our results are supported by a study showing normal cortical BMD in GHD children using pQCT [23]. The lack of a pubertal GHD group in the comparison of GHD and ISS children is a potential limitation of the study. In a small cohort of prepubertal ISS children reduced LS areal BMD was found before commencing GH therapy compared to an approximately age-, bone age-, and heightmatched control group [4]. The low–normal LS volumetric BMD results in our prepubertal ISS group confirm these findings despite the differing modes of size-corrections used. The etiology for the prepubertal reduction in LS volumetric BMD is unclear but we found that the absolute number with values b −2 SD is low. The increase in absolute areal BMD [4] and bone markers during GH therapy is similar to our

W. Högler et al. / Bone 37 (2005) 642–650

647

Fig. 2. Serum concentrations of IGF-I, inorganic phosphorus, alkaline phosphatase (ALP), and osteocalcin (OC) at baseline and at 6, 12, and 24 months of GH therapy. Differences between baseline and 24 months are given (Wilcoxon signed rank test). Boxes show median and interquartile range, whiskers 10th and 90th centiles and dots 5th and 95th centiles.

results. Our results indicate that bone mass and density in ISS are normal, at least after puberty. In agreement with previous studies [21,22,24], our GHD cohort had reduced LS vBMD at diagnosis, but also reduced FN vBMD compared to zero. However, most values were within the low–normal range. Two previous longitudinal studies have examined the effect of GH on size-corrected bone data in GHD children. Using pQCT, Schweizer et al. [23] measured low cortical area and thickness but normal forearm cortical BMD at diagnosis. Cortical BMD decreased during 1 year of GH therapy suggesting an increase in remodeling space. Similarly, the predominantly cortical TB BMC accrual was less than predicted in our study, as the relative increase in height was greater than in BMC, similar to a previous study using DXA [21,22]. This Dutch study reported an increase in LS vBMD SDS during GH therapy. However, this increase was not fully independent of height and disappeared once corrected for bone age [21,22]. Our results show no significant increase in vBMD SDS during GH treatment which contrasts previous studies in GHD or GH-resistant patients under GH or IGF-1 treatment [13,21,22]. Longer observation periods may be required to observe such an effect, although long-term data from the largest study so far are not convincing [22].

Body composition changes in childhood-onset GHD children and adults during GH therapy through increasing LTM and decreasing fat mass [21–23,25,30,31,38]. This process is reversed at the time of withdrawal of GH therapy [28,29,39]. We are the first to describe this GHinduced effect is much less in non-GHD children with ISS, particularly in regards to fat mass. The increase in LTM SDS reflects the anabolic action of GH in healthy individuals. Puberty resulted in the greatest absolute, but not relative gains in weight, LTM, BMI, and BMC, revealing the combined action of GH and sex hormones on bone accrual and body composition. Similar to our previous study in Turner Syndrome [37], puberty had a greater effect on LTM than on bone mass accrual. The “indirect” contribution of LTM, as the greatest predictor of BMC [16,40], to BMC accrual should not be neglected when assessing the effect of GH, or puberty, on bone. The most important question in the ongoing debate about the role of GH on pediatric bone health is the clinical relevance. An increased incidence of fractures has only been reported in adult-onset GHD, a group of patients largely suffering from multiple pituitary hormone deficiencies (MPHD) [41–43]. In a large data set of mixed adult/ childhood-onset GHD and MPHD, GHD seemed to be

648

W. Högler et al. / Bone 37 (2005) 642–650

associated with fractures [41], but fracture prevalence at the radius was significantly lower in childhood-onset compared to adult onset GHD. Notably, low bone mass is an infrequent finding in adult-onset GHD [44] despite typically altered body composition [45]. A recent meta-analysis found that GH treatment in adult-onset GHD patients had only minimal effects on BMD, concluding that there is insufficient evidence to regard osteopenia as an indication for GH therapy [46]. Fracture prevalence in isolated childhood-onset GHD is normal when MPHD is excluded [10,14]. More importantly, severe untreated GHD or GH insensitivity does not lead to osteoporosis [9–13] and there is no increased fracture risk in these rare GH deplete or resistant groups despite low LTM and high fat mass [12]. Interestingly, however, GH therapy per se seems to protect against fractures [14,46]. So far, the clinical evidence shows no short- or long-term consequence of isolated childhood-onset GHD on fracture risk. The evidence from in vitro and animal studies seems to contrast this clinical evidence, indicating abnormal collagen morphology [47,48] and altered bone geometry [5] in GHD rats. The reduced periosteal diameter, cortical thickness, and area in GHD rats [5,8] were also measured in GHD children [23]. Short stature and thus narrow bones have less strength from a biomechanical point of view and might predispose to fractures independent of GHD. Similar to studies in humans, rat volumetric BMD was normal using pQCT [8] but low using DXA [49], which questions the accuracy of current imaging techniques and the size-corrections applied. In addition, there are well-known metabolic actions of GH on osteoblasts, kidney, bone formation, and growth plate activity [7]. These actions are reflected by the rapid increase in IGF-I, phosphorus, and markers of bone formation during GH therapy in our study and those of others [6,21,22,31,50– 54]. In summary, despite altered bone geometry and low LS or FN vBMD in this and previous [14,21,22] studies in children with isolated GHD and known metabolic GH actions, there is poor evidence for an increased fracture risk for the peripheral or axial skeleton. We conclude that ISS children respond less to GH therapy in body composition, bone mass accrual, and also in growth velocity compared to GHD children. Bone density and body composition are within normal limits in ISS, particularly after puberty. The greater gain in height and LTM compared to BMC and the constant vBMD in both groups implies a delay in BMC accrual relative to bone elongation and muscle acquisition, a phenomenon which occurs physiologically during growth [55]. Growth hormone may primarily act on other functional determinants of bone strength like muscle strength, bone size, geometry, and turnover rather than on the density of the bone. Although GH therapy seems to protect against fractures, severe isolated childhood-onset GHD has not been associated with an increased fracture prevalence, which implies GH therapy is not indicated from the standpoint of bone health. The predominant effect of puberty on anthropometric and densitometric variables needs

to be considered in the planning of longitudinal studies in children.

References [1] Leschek EW, Rose SR, Yanovski JA, Troendle JF, Quigley CA, Chipman JJ, et al. Effect of growth hormone treatment on adult height in peripubertal children with idiopathic short stature: a randomized, double-blind, placebo-controlled trial. J Clin Endocrinol Metab 2004;89:3140–8. [2] Hintz RL, Attie KM, Baptista J, Roche A, Genentech Collaborative Group. Effect of growth hormone treatment on adult height of children with idiopathic short stature. N Engl J Med 1999;340:502–7. [3] Wit JM, Rekers-Mombarg LT, Cutler GB, Crowe B, Beck TJ, Roberts K, et al. Growth hormone (GH) treatment to final height in children with idiopathic short stature: evidence for a dose effect. J Pediatr 2005;146:45–53. [4] Lanes R, Gunczler P, Esaa S, Weisinger JR. The effect of short- and long-term growth hormone treatment on bone mineral density and bone metabolism of prepubertal children with idiopathic short stature: a 3-year study. Clin Endocrinol (Oxf) 2002;57:725–30. [5] Kim BT, Mosekilde L, Duan Y, Zhang XZ, Tornvig L, Thomsen JS, et al. The structural and hormonal basis of sex differences in peak appendicular bone strength in rats. J Bone Miner Res 2003;18:150–5. [6] Inzucchi SE, Robbins RJ. Clinical review 61: effects of growth hormone on human bone biology. J Clin Endocrinol Metab 1994;79: 691–4. [7] Ohlsson C, Bengtsson BA, Isaksson OG, Andreassen TT, Slootweg MC. Growth hormone and bone. Endocr Rev 1998;19:55–79. [8] Sjogren K, Bohlooly YM, Olsson B, Coschigano K, Tornell J, Mohan S, et al. Disproportional skeletal growth and markedly decreased bone mineral content in growth hormone receptor −/− mice. Biochem Biophys Res Commun 2000;267:603–8. [9] Maheshwari HG, Bouillon R, Nijs J, Oganov VS, Bakulin AV, Baumann G. The impact of congenital, severe, untreated growth hormone (GH) deficiency on bone size and density in young adults: insights from genetic GH-releasing hormone receptor deficiency. J Clin Endocrinol Metab 2003;88:2614–8. [10] Bouillon R, Koledova E, Bezlepkina O, Nijs J, Shavrikhova E, Nagaeva E, et al. Bone status and fracture prevalence in Russian adults with childhood-onset growth hormone deficiency. J Clin Endocrinol Metab 2004;89:4993–8. [11] Benbassat CA, Eshed V, Kamjin M, Laron Z. Are adult patients with Laron syndrome osteopenic? A comparison between dual-energy Xray absorptiometry and volumetric bone densities. J Clin Endocrinol Metab 2003;88:4586–9. [12] Bachrach LK, Marcus R, Ott SM, Rosenbloom AL, Vasconez O, Martinez V, et al. Bone mineral, histomorphometry, and body composition in adults with growth hormone receptor deficiency. J Bone Miner Res 1998;13:415–21. [13] Shaw NJ, Fraser NC, Rose S, Crabtree NJ, Boivin CM. Bone density and body composition in children with growth hormone insensitivity syndrome receiving recombinant IGF-I. Clin Endocrinol (Oxf) 2003;59:487–91. [14] Baroncelli GI, Bertelloni S, Sodini F, Saggese G. Lumbar bone mineral density at final height and prevalence of fractures in treated children with GH deficiency. J Clin Endocrinol Metab 2002;87:3624–31. [15] Prentice A, Parsons TJ, Cole TJ. Uncritical use of bone mineral density in absorptiometry may lead to size-related artifacts in the identification of bone mineral determinants. Am J Clin Nutr 1994;60:837–42. [16] Högler W, Briody J, Woodhead HJ, Chan A, Cowell CT. Importance of lean mass in the interpretation of total body densitometry in children and adolescents. J Pediatr 2003;143:81–8. [17] Carter DR, Bouxsein ML, Marcus R. New approaches for interpreting projected bone densitometry data. J Bone Miner Res 1992;7:137–45.

W. Högler et al. / Bone 37 (2005) 642–650 [18] Saggese G, Baroncelli GI, Bertelloni S, Barsanti S. The effect of long-term growth hormone (GH) treatment on bone mineral density in children with GH deficiency. Role of GH in the attainment of peak bone mass. J Clin Endocrinol Metab 1996;81: 3077–83. [19] Hulthen L, Bengtsson BA, Sunnerhagen KS, Hallberg L, Grimby G, Johannsson G. GH is needed for the maturation of muscle mass and strength in adolescents. J Clin Endocrinol Metab 2001;86: 4765–70. [20] Schonau E, Werhahn E, Schiedermaier U, Mokow E, Schiessl H, Scheidhauer K, et al. Influence of muscle strength on bone strength during childhood and adolescence. Horm Res 1996;45(Suppl 1): 63–6. [21] Boot AM, Engels MA, Boerma GJ, Krenning EP, de Muinck KeizerSchrama SM. Changes in bone mineral density, body composition, and lipid metabolism during growth hormone (GH) treatment in children with GH deficiency. J Clin Endocrinol Metab 1997;82: 2423–8. [22] van der Sluis I, Boot AM, Hop WC, De Rijke YB, Krenning EP, de Muinck Keizer-Schrama SM. Long-term effects of growth hormone therapy on bone mineral density, body composition, and serum lipid levels in growth hormone deficient children: a 6-year follow-up study. Horm Res 2002;58:207–14. [23] Schweizer R, Martin DD, Schwarze CP, Binder G, Georgiadou A, Ihle J, et al. Cortical bone density is normal in prepubertal children with growth hormone (GH) deficiency, but initially decreases during GH replacement due to early bone remodeling. J Clin Endocrinol Metab 2003;88:5266–72. [24] Baroncelli GI, Bertelloni S, Ceccarelli C, Saggese G. Measurement of volumetric bone mineral density accurately determines degree of lumbar undermineralization in children with growth hormone deficiency. J Clin Endocrinol Metab 1998;83:3150–4. [25] Leger J, Carel C, Legrand I, Paulsen A, Hassan M, Czernichow P. Magnetic resonance imaging evaluation of adipose tissue and muscle tissue mass in children with growth hormone (GH) deficiency, Turner's syndrome, and intrauterine growth retardation during the first year of treatment with GH. J Clin Endocrinol Metab 1994;78: 904–9. [26] Drake WM, Carroll PV, Maher KT, Metcalfe KA, Camacho-Hubner C, Shaw NJ, et al. The effect of cessation of growth hormone (GH) therapy on bone mineral accretion in GH-deficient adolescents at the completion of linear growth. J Clin Endocrinol Metab 2003;88: 1658–63. [27] Shalet SM, Shavrikova E, Cromer M, Child CJ, Keller E, Zapletalova J, et al. Effect of growth hormone (GH) treatment on bone in postpubertal GH-deficient patients: a 2-year randomized, controlled, dose-ranging study. J Clin Endocrinol Metab 2003;88:4124–9. [28] Tauber M, Jouret B, Cartault A, Lounis N, Gayrard M, Marcouyeux C, et al. Adolescents with partial growth hormone (GH) deficiency develop alterations of body composition after GH discontinuation and require follow-up. J Clin Endocrinol Metab 2003;88:5101–6. [29] Ogle GD, Moore B, Lu PW, Craighead A, Briody JN, Cowell CT. Changes in body composition and bone density after discontinuation of growth hormone therapy in adolescence: an interim report. Acta Paediatr, Suppl 1994;399:3–7. [30] Attanasio AF, Shavrikova E, Blum WF, Cromer M, Child CJ, Paskova M, et al. Continued growth hormone (GH) treatment after final height is necessary to complete somatic development in childhood-onset GHdeficient patients. J Clin Endocrinol Metab 2004;89:4857–62. [31] Underwood LE, Attie KM, Baptista J. Growth hormone (GH) doseresponse in young adults with childhood-onset GH deficiency: a twoyear, multicenter, multiple-dose, placebo-controlled study. J Clin Endocrinol Metab 2003;88:5273–80. [32] Hamill PV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM. Physical growth: national center for health statistics percentiles. Am J Clin Nutr 1979;32:607–29. [33] Cronk CE, Roche AF. Race- and sex-specific reference data for triceps

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

649

and subscapular skinfolds and weight/stature. Am J Clin Nutr 1982;35: 347–54. Tanner JM. Physical growth and development. In: Forfar JO, Arnell CC, editors. Textbook of pediatrics. Edinburgh: Churchill Livingstone; 1978. p. 249–303. Ogle GD, Allen JR, Humphries IR, Lu PW, Briody JN, Morley K, et al. Body-composition assessment by dual-energy x-ray absorptiometry in subjects aged 4–26 y. Am J Clin Nutr 1995;61:746–53. Lu PW, Cowell CT, LLoyd-Jones SA, Briody JN, Howman-Giles R. Volumetric bone mineral density in normal subjects, aged 5–27 years. J Clin Endocrinol Metab 1996;81:1586–90. Högler W, Briody J, Moore B, Garnett S, Lu PW, Cowell CT. Importance of estrogen on bone health in turner syndrome: a crosssectional and longitudinal study using dual-energy x-ray absorptiometry. J Clin Endocrinol Metab 2004;89:193–9. Ogle GD, Rosenberg AR, Calligeros D, Kainer G. Effects of growth hormone treatment for short stature on calcium homeostasis, bone mineralisation, and body composition. Horm Res 1994;41:16–20. Colle M, Auzerie J. Discontinuation of growth hormone therapy in growth-hormone-deficient patients: assessment of body fat mass using bioelectrical impedance. Horm Res 1993;39:192–6. Crabtree NJ, Kibirige MS, Fordham JN, Banks LM, Muntoni F, Chinn D, et al. The relationship between lean body mass and bone mineral content in paediatric health and disease. Bone 2004;35:965–72. Wuster C, Abs R, Bengtsson BA, Bennmarker H, Feldt-Rasmussen U, Hernberg-Stahl E, et al. The influence of growth hormone deficiency, growth hormone replacement therapy, and other aspects of hypopituitarism on fracture rate and bone mineral density. J Bone Miner Res 2001;16:398–405. Rosen T, Wilhelmsen L, Landin-Wilhelmsen K, Lappas G, Bengtsson BA. Increased fracture frequency in adult patients with hypopituitarism and GH deficiency. Eur J Endocrinol 1997;137:240–5. Vestergaard P, Jorgensen JO, Hagen C, Hoeck HC, Laurberg P, Rejnmark L, et al. Fracture risk is increased in patients with GH deficiency or untreated prolactinomas—A case-control study. Clin Endocrinol (Oxf) 2002;56:159–67. Murray RD, Columb B, Adams JE, Shalet SM. Low bone mass is an infrequent feature of the adult growth hormone deficiency syndrome in middle-age adults and the elderly. J Clin Endocrinol Metab 2004;89: 1124–30. Murray RD, Adams JE, Shalet SM. Adults with partial growth hormone deficiency have an adverse body composition. J Clin Endocrinol Metab 2004;89:1586–91. Davidson P, Milne R, Chase D, Cooper C. Growth hormone replacement in adults and bone mineral density: a systematic review and meta-analysis. Clin Endocrinol (Oxf) 2004;60:92–8. Lange M, Qvortrup K, Lander SO, Flyvbjerg A, Nowak J, Petersen MM, et al. Abnormal bone collagen morphology and decreased bone strength in growth hormone-deficient rats. Bone 2004;35:178–85. Evans BA, Warner JT, Elford C, Evans SL, Laib A, Bains RK, et al. Morphological determinants of femoral strength in growth hormonedeficient transgenic growth-retarded (Tgr) rats. J Bone Miner Res 2003;18:1308–16. Zhang XZ, Kalu DN, Erbas B, Hopper JL, Seeman E. The effects of gonadectomy on bone size, mass, and volumetric density in growing rats are gender-, site-, and growth hormone-specific. J Bone Miner Res 1999;14:802–9. Cowell CT, Woodhead HJ, Brody J. Bone markers and bone mineral density during growth hormone treatment in children with growth hormone deficiency. Horm Res 2000;54(Suppl 1):44–51. Baroncelli GI, Bertelloni S, Ceccarelli C, Cupelli D, Saggese G. Dynamics of bone turnover in children with GH deficiency treated with GH until final height. Eur J Endocrinol 2000;142:549–56. Fujimoto S, Kubo T, Tanaka H, Miura M, Seino Y. Urinary pyridinoline and deoxypyridinoline in healthy children and in children with growth hormone deficiency. J Clin Endocrinol Metab 1995; 80:1922–8.

650

W. Högler et al. / Bone 37 (2005) 642–650

[53] Vihervuori E, Turpeinen M, Siimes MA, Koistinen H, Sorva R. Collagen formation and degradation increase during growth hormone therapy in children. Bone 1997;20:133–8. [54] Saggese G, Baroncelli GI, Bertelloni S, Cinquanta L, Di Nero G. Effects of long-term treatment with growth hormone on bone and mineral

metabolism in children with growth hormone deficiency. J Pediatr 1993;122:37–45. [55] Rauch F, Bailey DA, Baxter-Jones A, Mirwald R, Faulkner R. The ‘muscle–bone unit’ during the pubertal growth spurt. Bone 2004;34: 771–5.