| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Menzies Research Institute (S.P., G.J.), Hobart, Tasmania 7000, Australia; and Miami Institute for Human Genomics (D.M.), Miami, Florida 33136
Address all correspondence and requests for reprints to: Graeme Jones, Menzies Research Institute, Private Bag 23, Hobart, Tasmania 7000, Australia. E-mail: g.jones{at}utas.edu.au.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Methods: A convenience sample of 640 male and female children aged 7–17 yr was studied. All were assessed for body composition (dual-energy x-ray absorptiometry), diet, strength, dexterity, habitual physical activity, sunlight exposure, smoking, and medication use. Skeletal age was assigned using the Tanner-Whitehouse-2 method.
Results: Subjects with a SAD greater than the 75th percentile had significantly higher height, weight, and Tanner stage compared with all other subjects. Bone-free lean mass, fat mass, and grip strength were positively associated with SAD. In multivariate analysis, ever smoking and use of inhaled corticosteroids were negatively associated with SAD, whereas milk drinking was positively associated with SAD. There was no significant association between sunlight exposure, television watching, light, or strenuous exercise and SAD.
Conclusions: The results of this study should be regarded as hypothesis generating but are biologically plausible and suggest that body composition, strength, diet, ever smoking, and inhaled corticosteroid use may be determinants of bone maturity relative to age and thus affect fracture risk in children. However, more studies are necessary to explore other determinants of SAD such as genetic and perinatal factors and whether SAD influences peak bone mass.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Selection of cases and controls
From May, 1998, to January, 2002, subjects aged 7–17 yr old who sustained a single-site upper limb fracture were invited to take part from a preexisting fracture registry (6). In brief, the fracture registry received reports from all radiology providers containing the word fracture from Southern Tasmania on a monthly basis. This region is larger than and fully encompasses the study area. Potential subjects were then screened from the registry by the authors and invited to take part by a letter of invitation with a follow-up telephone call for subject and parental consent. Fracture subjects were excluded if they had diseases that would prevent them completing the full protocol such as cerebral palsy and arthrogryposis, had moved out of the study area, were not enrolled in school, or had a previous upper limb fracture between the ages 9 and 16 yr. Nonresponders who were more than 3 months past the date of fracture were also excluded.
Controls were randomly selected from the same school class as the cases in the ratio of one control for every case (using available school lists and random numbers). They were also individually matched with cases by gender. Potential controls that had experienced an upper limb fracture between the ages of 9 and 16 yr were excluded. The response rate was 56% in both cases and controls.
There were no significant differences in mean age, male percentage, proportion of fracture type, and age distribution between responders and nonresponders for cases (all P > 0.05), although there was a trend toward higher nonresponse among males (11).
A total of 32 subjects were excluded in the whole group due to prior fracture.
Measurements
Exact age was calculated. Weight was measured with light indoor clothing without shoes to the nearest 0.1 kg using a single set of calibrated scales (Heine, Dover, NH). Height was measured without shoes to the nearest 0.1 cm by stadiometer (The Leicester height measure; Invicta Plastics Ltd., Oadby, UK). Body mass index was calculated (weight/height2). Grip strength was measured twice in the left hand using a bulb dynamometer (North Coast, Morgan Hill, CA) and the mean measure expressed in kilograms. Tanner stage was assessed by a validated self-administered instrument using drawings made from Tanner’s photographs, which illustrate the five stages of genital development for boys and breast development for girls (12).
Physical activity, sunlight exposure, milk intake (drinks per week), cola intake (drinks per week), smoking history, and inhaled corticosteroid use were assessed by questionnaire. The questionnaire had items on days of either strenuous activity or light activity for greater than 20 min in the last 2 wk (1, none; 2, 1–2 d; 3, 3–5 d; 4, 6–8 d; 5, 
9 d); daily television, computer and video viewing in the last week (1, 
1 h; 2, 2–3 h; 3, 4–5 h; 4, 6 h); average length of time per day spent outside (sunlight exposure) in summer and winter on weekdays, weekends, and school holidays (1, <2 h/d; 2, 2–3 h/d; 3, 3–4 h/d; 4, >4 h/d). Inhaled corticosteroid use for beclomethasone, budesonide, or fluticasone was recorded (1, no; 2, yes) if the child had used preventative asthma inhalants for a period of 3 months or longer in the preceding 12 months. Oral corticosteroid use was recorded (1, no; 2, yes) if the child had regularly taken oral prednisolone for a period of 3 months or longer in the preceding 12 months. Current smoking (1, no; 2, yes) and having ever smoked a cigarette (1, no; 2, yes) were also assessed. Milk, cola, and carbonated drink consumption was added to the questionnaire from 2000 onwards and assessed each drink type intake per week (1, none; 2, 1–2 drinks/wk; 3, 3–5 drinks/wk; 4, 6–7 drinks/wk; 5, >7 drinks/wk).
Bone mineral density and bone mineral content at the total body, hip, and spine using a Hologic QDR2000 densitometer were measured as previously described and reported (11). Total body lean mass and fat mass were also available from these scans. Bone-free lean mass was calculated by subtracting total body bone mineral content from total body lean mass (as measured by dual-energy x-ray absorptiometry).
Radiographs of the left hand and wrist were made using a double-wrapped Osray T-4 (Agfa-Gevaert) film, with an exposure time varying from 0.3–0.5 sec. Skeletal age was evaluated by one examiner (D.M.) according to the Tanner-Whitehouse-2 (TW2) method (13). The TW2 method involves the examination of 20 bones of the hand and wrist and the assignment of a letter grade to each bone dependent on the attainment of clearly described bone-specific maturity indicators. The letter grade is then converted to a numeric score, and the scores are summed for each individual to give a maturity score on a scale of 0–1000. In this study, the radius-ulna-short bones option of the TW2 method, using only the radius, ulna, and 11 short bones of the hand, was used to determine skeletal age. Maturity scores were converted to skeletal age for each individual using gender-specific tables for the radius-ulna-short bones (TW2) method. A maturity score of 1000 equates to a bone age of 16 yr in girls and 18 yr in boys in this method (because girls on average complete their skeletal growth/maturation 2 yr earlier than boys.) The outcome variable in all analyses was SAD, which was defined as skeletal age minus chronological age. In girls aged over 16 yr and boys over 18 yr with a maturity score of 1000, the SAD score was automatically regarded as zero. There were seven girls in the study with a maturity score of 1000 who were over 16.0 yr old (and none over 17 yr old). This small group of girls was included in the analyses, but excluding them did not change any results. None of the boys in the study had reached the age of 18 yr.
A test of reproducibility was carried out on 31 radiographs 1 month apart with an intraobserver coefficient of variation of 1.1%.
Statistics
ANOVA testing was used to compare mean differences in study factors between subjects with a SAD less than 25th percentile (SAD < –0.288), 25–75th percentile, and more than the 75th percentile (SAD > +1.262). Given SAD was normally distributed, univariate and multivariate correlation analyses were appropriately used for examining the association between SAD and body composition and environmental factors. SAD models were adjusted for chronological age, fracture status (case or control), Tanner stage, height, and weight. Additional models were adjusted for chronological age, fracture status (case or control), Tanner stage, height, and lean body mass. In standard regression model diagnostic checks, we found that age was best modeled as a linear term in the multivariate models. Tanner stage was included in the models as a ranked ordinal variable; however, if Tanner stage was treated as a categorical variable, then all associations remained statistically unchanged.
All statistical analyses were performed on SPSS version 14.0 for Windows (Cary, NC) or STATA version 10.0 for Windows (College Station, TX).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
documents the physical characteristics of the study population comparing subjects with a SAD less than the 25th percentile, SAD 25–75th percentile, and SAD more than the 75th percentile. There was a significant difference in age between the subjects with a SAD less than the 25th percentile and SAD more than the 75th percentile. Subjects with a SAD more than the 75th percentile had a significantly higher height, weight, and Tanner stage compared with all other subjects. Bone-free lean mass, fat mass, and grip strength were also significantly higher in those with a SAD more than the 75th percentile. Height, weight, Tanner stage, bone-free lean mass, fat mass, and grip strength were significantly different when comparing all three groups.
|
Table 2 details the multivariate associations between SAD and body composition. Bone-free lean mass, adjusted for age and case control status, was significantly associated with SAD in the whole group as well as in males and females separately where the magnitude of the association was similar. In comparison, fat mass was significantly but less strongly associated with SAD when examining the whole group. Fat mass in males was significantly positively associated with SAD. Fat mass in females had a similar coefficient; however, the association with SAD did not reach statistical significance. Further adjustment for Tanner stage did not change the statistical significance of these results, but most associations became smaller in magnitude.
|
and Fig. 1 document the association between left grip strength and SAD. Grip strength remained positively associated with SAD when adjusted for chronological age, case control status, Tanner stage, height, and weight but not when lean body mass was substituted for weight (results not shown). The mean average left grip strength of the 73 left-handed children was 21.60 kg and the mean average left grip of 530 right-handed children was 22.90 kg. (34 children did not have a handedness side allocated on the questionnaire.) There was no statistically significant difference in age, SAD, or average left grip strength between the left- and right-handed children (results not shown).
|
details the univariate and multivariate environmental associations with SAD. In univariate analysis, ever smoking and current smoking were not significantly associated with SAD. However, after adjustment for chronological age, case control status, Tanner stage, height, and lean body mass (but not weight), ever smoking was significantly negatively associated with SAD. The youngest subject to report ever smoking was aged 8.48 yr. In multivariate analysis as described above for smoking, excluding children aged less than 8.48 yr, and children aged less than 10.0 yr, SAD remained significantly negatively associated with ever smoking (standardized β-coefficients –0.080 and –0.082, respectively, P < 0.02). Milk drinking was significantly positively associated with SAD. When adjusted for age, case control status, Tanner stage, and weight, the association remained. Further adjustment for height led to a nonsignificant result. Inhaled corticosteroid use was associated with SAD. When adjusted for age, case control status, Tanner stage, and weight, the association remained. Adjustment, in addition, for height led to a nonsignificant result.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Lean body mass had the strongest association with SAD, which may be mediated through a variety of pathways including physical (mechanical loading), nutritional, hormonal, and genetic pathways. Common factors may drive both increases in lean body mass and skeletal maturity seen at the time of puberty. Candidate factors include GH, IGF-I, estradiol (only limited data that estradiol increases lean body mass), and testosterone, all of which have anabolic effects on bone accretion and linear bone growth and stimulate increases in lean body mass at the time of puberty (14). The known association between greater muscle mass and a higher bone mineral density is likely to be explained by genes regulating size (15). The association between whether a child has a positive or negative SAD and their total lean body mass also may be explained by genetic factors regulating both muscle size and the rate of development and maturation of the growth plates. It should also be noted that exercising a dominant upper arm can lead to improvements in bone mineral density not wholly accounted for by improvements in muscle mass (16). Potentially, the same is true for SAD, and an exercised muscle (for example, as determined by grip strength) may have more trophic influence on the rate of development of the underlying skeleton. Left grip strength was also associated with SAD, which became negative after adjustment for lean body mass; however, lean body mass remained significant, suggesting the effect is mediated through lean body mass or on the causal pathway.
This study also notes that fat mass had a weaker but still positive association with SAD. This finding is in keeping with the report by Clark et al. (17) that adipose tissue stimulates bone growth in prepubertal children. Obese children are younger and taller and have more advanced bone age to chronological age ratio than nonobese children of a similar pubertal stage (18), suggesting a role for adipose tissue in skeletal maturation. Possible mechanisms include the stresses from mechanical loading and the metabolic effects (direct or indirect) of hormones such as leptin secreted and regulated by the adipocytes (19). Obesity may indirectly hasten the time to skeletal maturity by affecting the timing of puberty, with affected children entering puberty earlier due to higher estradiol levels or leptin levels (18). Endocrine factors that affect bone age may also be determinants of fat mass. In prepubertal children, fat mass is related to both serum IGF-I levels and estradiol levels (19). Around the time of puberty, possible confounding interactions between SAD and fat mass are complex. Conclusions are limited in a cross-sectional study, but all of the constitutional associations we report persist after adjustment for pubertal stage; thus, confounding does not seem likely.
Milk drinking had a positive significant association with SAD in univariate analysis, which was largely mediated through larger body size. There was also a significant positive difference in milk intake per week between the children in each of the three categories in Table 1. This most likely reflects greater nutrition from milk drinking mediated by protein, growth factors, and/or calcium. Calcium supplements in children have a small effect on bone mass (20); however, they may improve bone size (21). Cadogan et al. (22) reported that increased milk consumption enhanced bone mineral acquisition in adolescent girls, and this may have been mediated through serum IGF-I, which was higher in the intervention group. In addition, high intakes of milk and not meat have been shown to increase serum IGF-I levels in 8-yr-old boys (23). IGF-I is bone anabolic, and this may explain some of the benefits of milk drinking with regard to SAD. Because IGF-I stimulates the growth plate chondrocytes and osteoblast differentiation (24), a child with higher IGF-I levels [due to milk consumption as shown by the paper from Cadogan et al. (22)] might reasonably be expected to be ahead in the process of skeletal growth/maturation when compared with an age-matched child with lower IGF-I levels.
Interestingly, cola drinking and television viewing had no apparent effect on SAD despite findings that cola drinking is associated with increased wrist and forearm fracture risk in children (25) and television viewing has a dose-dependent association with wrist and forearm fractures in children (26).
Ever smoking was negatively associated with SAD on multivariate analysis after adjustment for confounders and was associated with a 3-month delay in bone age. This is potentially biological but is more likely explained by socioeconomic factors consistent with the report of Cole and Cole (27), who reported an inverse association between bone age and social deprivation that was thought possibly to be a result of social deprivation retarding skeletal maturation during a critical period in early life.
It has been reported that high-dose inhaled corticosteroids reduce the acquisition of bone mineral in prepubertal children (28, 29). It has long been known that oral corticosteroids have an effect on bone maturity. This study found that inhaled corticosteroid use for a period of 3 months out of the last 12 months was associated with a 5-month delay in bone age. This is consistent with data that inhaled corticosteroid use is associated with increased fracture risk (30). Again, this was largely mediated by smaller body size. Conclusions on the total dose and duration are limited because data were not collected on these factors, but oral prednisolone use was not associated with SAD.
In clinical terms, for example, a child who was a high milk drinker (more than seven drinks milk per week, compared with one to two drinks per week), did not take inhaled corticosteroids, did not ever smoke, and had a high grip strength above the mean (22.9 kg) would have a skeletal age 2.07 yr ahead. Based on the association between SAD and fracture that we have previously reported (10), this equates to a 35% reduction in upper limb fracture risk and a 64% reduction in hand fracture risk.
Limitations of this study include the cross-sectional convenience design, with the original study having used a matched design. This was broken for the current study; thus, it was necessary to adjust for case control status. According to Miettinen, results can be generalizable if three key criteria are met: clear study entry criteria, wide distribution of study variables, and adequate sample size (31). Our study meets all these criteria. It would be difficult, however, to design longitudinal or interventional studies to assess determinants of SAD because at the completion of bone maturation in a longitudinal study, all SADs will trend to zero.
In conclusion, the results of this study should be regarded as hypothesis generating but are biologically plausible and suggest that body composition, strength, diet, ever smoking, and inhaled corticosteroid use may be determinants of bone maturity relative to age and thus affect fracture risk in children. However, additional studies are necessary to explore other determinants of SAD such as genetic and perinatal factors and whether SAD influences peak bone mass.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Information: All authors have nothing to declare.
First Published Online December 4, 2007
Abbreviations: SAD, Skeletal age deviation; TW2, Tanner-Whitehouse-2.
Received May 23, 2007.
Accepted November 26, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
