| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
University of Tennessee Health Science Center (F.A.T., K.M.R., R.L., V.P., C.W., J.N., L.D.C.), Memphis, Tennessee
University of Jyväskylä (S.C.), Jyväskyla, FINLAND
Address reprint requests to: Frances A. Tylavsky, Dr.P.H., Department of Preventive Medicine, 66 N. Pauline St., Suite 633, Memphis, TN 38105. E-mail: ftylavsky{at}utmem.edu
 |
ABSTRACT |
|---|
 TOPFOOTNOTES ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
Methods: This was a longitudinal study (1.98 ± 0.07 y) of sixty-nine children (89% Caucasian, 44% male) enrolled in a calcium supplementation trial. Bone area, bone mineral content (BMC) and density (BMD) of the whole body and radius were assessed using a QDR 2000 (Hologic, Inc) dual energy x-ray absorptiometer. Serum PTH and 25(OH)D were measured using radioimmunoassays.
Results: Vitamin D stores were inversely related gain in bone area (p < 0.002), BMC (p < 0.002) BMD (p < 0.027), as well as to PTH levels (p < 0.0001). Compared to those with adequate vitamin D stores (>34 ng/ml), those who had consistently low vitamin D stores (18 ng/ml) had a 8% larger gain in bone area (p < 0.05); 11% in BMC (p < 0.05) and no differences in gain in BMD; after adjusting for baseline bone measurements, race, gender, season measured, Tanner stage, and calcium intake.
Conclusions: High normal PTH with low-normal 25(OH)D stores and moderate to high calcium intake may be beneficial to accruing larger bone size and BMC during puberty.
Key words: dual energy x-ray absorptiometry, children, bone mass, vitamin D, parathyroid hormone
|
INTRODUCTION |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
|
SUBJECTS AND METHODS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
Study Measurements
Anthropometry
Body weight was measured on a balance-beam scale to the nearest 0.1 kilogram (kg). Height was measured twice to the nearest 0.1 cm using a Harpenden stadiometer (Holtain, Wales, UK). If the two height measurements differed by greater than 0.5 cm, a third measurement was performed. The average of the two measurements within 0.5 cm was used for these analyses. Body mass index (BMI) was calculated as weight in kg divided by height in meters squared.
Dual Energy X-ray Absorptiometry (DXA): Whole-Body
A Hologic QDR bone densitometer Model 2000 was used to measure lean tissue mass (LTM in kg.), body fat in kg., bone area, in cm2, bone mineral content (BMC, in g), and bone mineral density (BMD in g/cm2) of the whole body. Whole body measurements were assessed using the array mode and analyzed using enhanced whole-body software version 5.73a. The DXA Quality Assurance manual for the study was used to standardize patient positioning and scan analysis. Two scans at the baseline visit were obtained with repositioning. The average of the two scans is reported as the values for the bone parameters.
Laboratory Assessments
Twenty-four hour urine samples were collected at the baseline visit and at 3 month intervals over a two-year period. A total of 8 samples were possible over the 24 month period. Participants were given detailed verbal as well as written instructions on how to collect a complete 24-hour urine sample. Time and volume of collections were recorded. Samples were considered to be complete if the volume exceeded 24 ml/kg of body weight and creatinine was within two standard deviations of creatinine values for height of the participant. [9]. Samples that did not meet these criteria were not included in this analysis. Urine creatinine was determined using the Jaffe reaction (picric acid). Urine calcium was measured by ocresolphthalein. The intra-assay coefficient of variation was 0.6–1.5% for creatinine and 0.9–1.2% for calcium. Urinary deoxypyridoline (Dpd) was measured by a competitive enzyme immunoassay (Pyrilinks® D, Metra Biosystems, Inc. Mountain Home, Ca). The interassay coefficient of variations for Dpd was 5%. Twenty-four-hour calcium excretion is reported in mg/kg of body weight. The value reflects the average of all 24-hour urines collected over a two-year period. The mean number of urine collections was 4.8 ± 2.3 and the median was 5 collections.
Blood samples were drawn between 4 and 7 PM at the baseline visit and at 12 and 24 month intervals during the trial. The samples were stored at –70° C and were analyzed in batch by participant at the end of the study. Serum parathyroid hormone (PTH) was assessed by a two-site immunoradiometric assay utilizing two affinity-purified antibodies. One antibody raised against the mid and C terminal (PTH 39–84) is immobilized on plastic beads, while the other antibody raised against the N terminal PTH 1–34 is labeled with 125I (DiaSorin, Stillwater, Minnesota, USA). The coefficient of variation (CV%) for intra and inter assay is <7% with the range for normal children being 13 to 50 pg/ml. Serum 25-hydroxy-vitamin D (25(OH)D) was measured by radioimmunoassay (DiaSorin, Stillwater, Minnesota, USA). The intra- and inter-assay coefficients of variation are < 10%. The reference range for 25(OH)D was 15–60 ng/l. Osteocalcin was measured by a competitive enzyme immunoassay using a monoclonal antibody to measure both intact osteocalcin and its large N-terminal mid-fragment (Nichols, Inc. San Juan Capistrano, CA). The inter-assay coefficient of variation for osteocalcin was < 4.9%.
Vitamin D Receptor
Genomic DNA was isolated either from fresh whole blood or frozen leukocyte pellets by solid phase column extraction (Gentra Systems, Minneapolis, MN). Yield was determined by spectrophotometry, and samples were adjusted to a standard concentration. Genotyping of the Taq I polymorphism of the vitamin D receptor gene (Ile-352 C/T polymorphism; ref SNP ID rs731236) was performed by real-time PCR using an ABI Prism 7000 (Applied Biosystems, Foster City, CA). Each allele-specific probe was labeled with a different fluorochrome. Relative fluorescence of the two probes was measured in each biplex reaction, and results were displayed as scatter plots. Genotypes were determined by cluster analysis of scatter plots. Each sample was run in duplicate and checked for consistent results. Allelic designations were as follows. The uppercase "T’ allele, reported in earlier literature as absence of the TaqI recognition site, corresponded to the "T’ allele of the C/T polymorphism. The lowercase "t’ allele (presence of the TaqI site by restriction digest) corresponded to the "C’ allele of the C/T polymorphism.
Dietary Intake
Food Records.
Participants provided a 1-day food record at the baseline and 1 record every three months over the 24 month period. (N = 8). Each participant and his/her guardian were provided detailed instructions on keeping a 1-day food record using a standardized collection form. Food models and photographs of serving sizes of foods were used to estimate the quantity of all foods and beverages consumed. A research assistant was trained and certified by a Registered Dietitian to provide instructions to the participant for recording all food and beverages and to review the completed records with the participant and their parent. Sixty-eight percent of the participants provided at least 8 independent days of food records by 24 months; an additional 25% provided 5–7 days of food records during the 24 months follow-up.
Nutrient Intake.
The food records used to estimate food group consumption were analyzed for nutrient content using the NutriBase 2001 Clinical v.3.03 (CyberSoft, Inc. Phoenix, Arizona). Nutrient intake (kilocalories, protein, carbohydrate, fat, vitamin C, A and D, potassium, magnesium, phosphorus, calcium) reflects the average of all the independent food records provided by each participant. Total calcium intake, reflects the consumption from diet and compliance adjusted intake from supplements. Pill compliance was assessed based on the return of pills not consumed. Calcium intake from pills was obtained by multiplying the percent of pills returned by 1000 mg/day. Overall, calcium supplements adjusted for compliance contributed 444 mg/day for those who were randomized to 1000 mg/calcium per day. Compliance was 70–75% which is similar to other calcium supplementation studies [10,11].
Physical Activity
Subjects completed a 1-day physical activity checklist on the same days that diet records were collected (one record every three months). The physical activity log used for this study was developed and validated by Sallis and colleagues [12]. All physical activities for a 24-hour period were recorded, including activities of daily living. Energy expenditure was calculated based on the weight of the participant, the frequency and duration of each activity recorded [13,14]. Eighty one percent of the participants provided 5 or more records over the 2-year period with seventy-three percent providing 8 or more days of physical activity records.
Tanner Staging
Participants were given a standardized series of drawings to assess their own pubertal development. Girls were given line drawings of the 5 stages of breast and female pubic hair. Boys were given line drawings of boys showing the 5 stages of pubic hair development. The participant was placed in an exam room and was asked to circle the drawing that best represented their assessment of their development. This procedure was performed at the baseline, and at the 12 and 24 months follow-up visit. This procedure has been previously validated with kappa scores ranging from 0.64 to 0.66 (p < 0.001) [15].
Socioeconomic Status
Parents or legal guardians completed a social history questionnaire with regard to income, number of people residing in their household and the highest education level attained by one of the guardians/parents.
Statistical Analysis
Data were analyzed using JMP software [16]. Means and standard errors were calculated for all continuous measures. To evaluate those who had consistently low or high 25(OH)D levels, individuals were grouped based on the consistency of their 25(OH)D levels: We placed them into one of three categories: (1) Low was defined as serum 25(OH)D < 18 ng/l at all measurement points; (2) The high group was defined as serum 25(OH)D levels w > 34 ng/l at each measurement time; and (3) the middle group was composed of the remaining participants (
18 ng/l but 
34 ng/l). Analysis of variance was used to compare differences between the 25(OH)D groups adjusted for age, height, weight, BMI, percent body fat, household size, parent's education and nutrient intake. Chi square test was used to determine if there were differences between the 25(OH)D groups for levels of family income and percent that consumed the Estimated Average Requirement (EAR) or Adequate Intake (AI) for the respective nutrient.
Multiple regression analyses was used to describe the continuous relationship between serum 25(OH)D and annualized percent gain in bone area, BMC and BMD while controlling for Tanner stage of sexual maturity at the 24 month visit, season measured for the annual bone assessments, gender, race and treatment. The covariates included for adjustment were evaluated by adding and removing variables from the regression equation and observing the change in AIC (Aikake's information criterion). A p-value of less than 0.05 was considered to be a significant relationship between serum 25(OH)D and annualized change in bone parameters. The effect of variation in growth speed or VDR taq allele combinations were tested by incorporating change in height or VDR taq allele combinations after the final model was selected.
Comparisons of bone parameters and percent change in bone parameters, PTH and 25(OH)D, markers of bone turnover, urinary sodium and calcium and change in height and weight among the three 25(OH)D groups were evaluated using analysis of covariance. When annualized percent change in bone parameters were evaluated, the baseline bone measurement that corresponded to the dependent variable was added to the above stated covariates. A P value of less than 0.05 was considered statistically significant when comparing differences across the 25(OH)D groups. If the P value was less than 0.05, point comparisons were made between the groups using Tukey's adjustment for multiple comparisons.
|
RESULTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
|
|
Annualized gain in bone area, BMC and BMD was determined over the length of follow-up. Twenty-five hydroxy vitamin D was negatively related to the gain in total body bone area (p = 0.003), BMC (p = 0.005) and BMD (p = 0.15) These associations remained after adjusting for gender, race, final tanner stage, season in which the person had their annual measurements made, treatment assignment and baseline bone measurements (Fig. 1). However, when taking into account variation in the gain in height as a measure of growth velocity, the significance disappeared. Table 3 shows that individuals with consistently low serum 25(OH)D levels (<18 ng/l) had an 8% higher gain in bone area, 11% higher gain in BMC and no difference in BMD compared to those with consistently adequate serum 25(OH)D levels (>34 ng/l) over the follow-up period. Those with low serum 25(OH)D levels had higher PTH (p = 0.002), but similar 24-hour urinary calcium excretion compared to those with the highest 25(OH)D levels. We did not find any association between 25(OH)D levels and markers of bone formation (osteocalcin) or resorption (DpD) at the baseline or 24 months follow-up or between treatment groups (Data not shown). The lowest 25(OH)D group had a higher change in height and weight compared to the highest quartile of 25(OH)D group (p < 0.05, Tukey's adjustment for multiple comparisons).
|
|
|
DISCUSSION |
|---|
The concentration of 25(OH)D and PTH can fluctuate according to dietary intake and exposure of sunlight [17]. This study found seasonal effects on 25(OH)D status as have others [18–21] and controlled for it in our analyses. In humans, however, the sun exposure and skin synthesis of vitamin D is a variable that is difficult at best to quantify and was not addressed in this study.
During adolescence one compensatory mechanism to meet the needs of skeletal growth is through the increased intestinal absorption of calcium via the vitamin D and PTH axis [22–23]. It is well established that serum 25(OH)D levels are negatively related to PTH levels in adults [24] and children [4,5,21,25]. In adults the current dogma is to supplement those with sub-optimal levels of 25(OH)D with vitamin D to suppress PTH in favor of preserving bone mass. In our study, all children had PTH levels within normal limits, but those with lower 25(OH)D had systematically high-normal PTH compared to those with higher 25(OH)D stores. There are two possible mechanisms which may lead to lower 25(OH)D levels. The first is that lower intake of vitamin D or less sunlight exposure leads to lower stores. The second is that higher levels of PTH associated with inadequate calcium intake secondary to growth demands leads to conversion of 25(OH)D to 1,25OHD. Lee et al. [26] using double labeled stable Ca isotopes demonstrated that in adolescents with low calcium intake, fractional absorption of calcium was negatively related to 25(OH)D stores and positively related to PTH. The authors hypothesized that the low 25(OH)D stores could be due to the increased conversion of 25(OH)D to 1,25(OH)2 D3. Animal studies provide support that increases in 1,25(OH)2 D3 results in lower 25(OH)D stores [27]. Whether the same phenomenon occurs in adolescents and is related to rapid growth has not been tested.
Lower 25(OH)D levels accompanied with higher normal PTH levels were associated with larger accrual of bone area, BMC and BMD secondary to higher gain in height. Physiologically, PTH has been shown to have anabolic actions on trabecular bone, and catabolic actions on cortical bone and is affected by weight bearing capacity in adults [28–29]. PTH increased periosteal bone formation, leading to wider bones and increased endosteal resorption resulting in less dense bones in animal models [30]. In growing rats with low dietary calcium intake, supplementation with vitamin D blunted the PTH levels and periosteal bone formation, increased 25(OH)D stores, but was unable to affect the endosteal resorption of cortical bone [30]. Histomorphometric data suggested that vitamin D supplementation increased maturation-related cancellous bone volume/total volume only in those with adequate calcium intake, with no affect in those with a low calcium diet [30]. Collectively, these data imply that elevated PTH is a compensatory mechanism due to low calcium intake and low 25-OH D stores may be reflective of this process. In contrast, vitamin D supplementation with adequate calcium intake, resulted in larger bone volume, density, thicker trabeculae, higher 25(OH)D and 1,25 OH D [30]. These data imply the availability of the calcium for deposition, at least in rat models, may be the determining factor for whether PTH is a net benefit or liability and vitamin D may modulate the outcome. A study in adolescents supports the importance of PTH not 25(OH)D in modulating BMD when calcium intake is low or inadequate [31]. Our current study data supports the role of PTH in modulating BMD, but implicates 25(OH)D in the process. Those with higher levels of PTH had lower 25(OH)D levels and it can be hypothesized that the PTH increased periosteal expansion as indicated by bone area but BMC also increased resulting no difference in BMD. The concomitant increase in BMC may be reflective of the adaptive capability exhibited by increased PTH levels in the presence of moderate calcium intake during growth.
Our previous findings showed that detrimental effects of low vitamin D and higher normal PTH levels are not reflected in BMD by DXA [5] but were evidenced by increasing porosity in cortical indices of the tibia and larger cross sectional area of the distal radius by pQCT [5]. Therefore, there remains concern that our data are a result of an insensitive measurement technique to detect differentiating effects on cortical and trabecular bone. However, it cannot be ruled out that differences in BMD are size dependent artifacts of DXA measurements.
Genetic factors have been suggested to interact with the vitamin D metabolism [7,8,32]. Our attempt to evaluate this effect was through the VDR Taq polymorphisms. In this cohort, VDR Taq alleles did not have an effect modification on 25 (OH)D's relationship with the accrual of bone area, BMC or BMD after controlling for gender, ethnicity, season measured, average total calcium intake and Tanner stage of sexual development at the 24-month exam. These results imply that genetic regulation by VDR taq alleles either regulates growth as suggested by Tao et al [7] or are confounded by gain in weight and height.
Study Limitations
While our study does provide support that lower 25(OH)D levels and high normal PTH levels in the presence of moderate to high calcium intake may be beneficial to accrual of bone mass, it does have limitations. The sample was small and limited to 69 children from affluent families. The relatively low number of males and blacks make it difficult generalized to all ethnic groups or socioeconomic groups. This study enrolled children at Tanner stage II and may not be generalizable to younger and older children. Many other hormones and genetic factors [33] can affect bone accrual which were not addressed by this paper, thus their influence can not be ruled out. Since this is a secondary analysis, these results will need to be verified in other populations either through secondary analyses or through a prospective controlled trial
Summary
In summary, higher normal PTH level accompanied by lower 25(OH)D levels with moderate to high calcium intake appear to have beneficial effects on the accrual of bone area and BMC secondary to growth for the whole body in pubertal children after controlling for gender, race, season measured, Tanner stage of pubertal development and use of calcium supplements. However, cumulative effects on peak bone mass and on cortical versus trabecular bone remain to be determined.
|
Contributors |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
None of the authors have financial or personal affiliations with the sponsors of this research effort.
|
ACKNOWLEDGMENTS |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
Received March 4, 2005. Accepted February 6, 2007.
|
REFERENCES |
|---|
|
TOPFOOTNOTESABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSContributorsACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. M. Weaver, L. D. McCabe, G. P. McCabe, M. Braun, B. R. Martin, L. A. DiMeglio, and M. Peacock Vitamin D Status and Calcium Metabolism in Adolescent Black and White Girls on a Range of Controlled Calcium Intakes J. Clin. Endocrinol. Metab., October 1, 2008; 93(10): 3907 - 3914. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
