点击显示 收起
1 From the Department of Clinical Nutrition (GSR and GD), the Metabolic Bone Diseases Unit (RPD-G and SI-S), and the Endocrine Laboratory (BR), Rambam Medical Center Haifa, Haifa, Israel; and the Departments of Community Medicine & Epidemiology (GR and HSR) and Pediatrics (NI-S), Carmel Medical Center, Haifa, Israel.
2 Supported by the Chief Scientist Fund of the Israeli Ministry of Health. Calcium and placebo supplements were donated by Teva Pharmaceutical Company. 3 Address reprint requests to GS Rozen, 36 Lea Street, Haifa 34403, Israel. E-mail: rgeila{at}rambam.health.gov.il.
ABSTRACT
Background: High calcium intakes during adolescence may increase bone acquisition. The magnitude of the effect of dietary
calcium supplementation and the timing of its administration to
achieve significant effects on bone health are still incompletely
defined.
Objective: The objective of this study was to assess the effect of calcium supplementation on bone mass accretion in postmenarcheal adolescent girls with low calcium intakes.
Design: A double-blind, placebo-controlled calcium supplementation study was implemented. One hundred girls with a mean (± SD) age of 14 ± 0.5 y with habitual calcium intakes < 800 mg/d completed a 12-mo protocol. The treatment group received a daily supplement containing 1000 mg elemental calcium. Bone mineral density (BMD) and bone mineral content (BMC) of the total body, lumbar spine, and femoral neck were determined at inclusion, 6 mo, and 12 mo. Also measured were serum concentrations of biochemical markers of bone turnover (osteocalcin and deoxypyridinoline), parathyroid hormone, and vitamin D.
Results: The calcium-supplemented group had greater accretion of total-body BMD and lumbar spine BMD but not BMC than did the control group. Calcium supplementation appeared selectively beneficial for girls who were 2 y postmenarcheal. Calcium supplementation significantly decreased bone turnover and decreased serum parathyroid hormone concentrations.
Conclusion: Calcium supplementation of postmenarcheal girls with low calcium intakes enhances bone mineral acquisition, especially in girls > 2 y past the onset of menarche.
Key Words: Calcium supplementation • double-blind study • adolescents • bone density • postmenarcheal girls
INTRODUCTION
As the average life span increases, osteoporosis is a growing
health concern in the Western world. Women are at higher risk
than men of developing osteoporosis as a result of naturally lower
peak bone mass and rapid bone loss after menopause. The results
of previous studies have proven that a high adult peak bone mass
is protective against late-life fragility fractures (1-4). Nutrition, in
particular adequate calcium intake, is one of the major environmental factors believed to have a positive effect on bone accretion,
within the limits of genetic boundaries (5-14). In recent years,
several studies have reported a positive effect of calcium supplementation on bone mass in children and adolescents (15-21). With
respect to puberty-treatment interactions, it is still difficult to draw
conclusions about the effect of calcium supplementation on bone
accretion during consecutive stages of pubertal transition. During
puberty, rapid bone accretion is observed as a consequence of an
increase in sex hormone concentrations (22-25). To evaluate the
effect of calcium supplementation on bone mass without the
interference of premenarcheal hormonal changes, the study group
in the present study consisted of adolescent girls who were =" BORDER="0"> 1 y
postmenarcheal (26, 27). The purpose of the present study was to
assess the effect of 1 y of calcium supplementation on bone mass
in postmenarcheal girls with low calcium intakes and to investigate the physiologic mechanism for this effect.
SUBJECTS AND METHODS
One hundred twelve girls were enrolled in a 1-y, double-blind, placebo-controlled calcium supplementation study. The
ethnic distribution of the group was 85 Jewish girls and 27
Arab girls. The hospital review board approved the study
protocol, and informed consent was obtained from the girls and
their parents.
The following inclusion criteria were applied to the intervention study: calcium intake < 800 mg/d, =" BORDER="0"> 1 y postmenarcheal, age < 15.5 y, no chronic disease, nonsmoking, and no use of contraceptives. All subjects in the present study were recruited from our earlier cross-sectional study of food habits and bone health among high school girls (28).
The girls were randomly assigned to calcium supplementation (CS) or placebo. The CS group received 1000 mg elemental calcium/d in the form of calcium carbonate chewable tablets (Tevasidan; Teva Pharmaceuticals Industry, Petah-Tiqva, Israel). The control group received identically shaped placebo tablets provided by the same manufacturer. Trained dietitians supplied the tablets monthly.
Follow-up included a monthly interview with a trained dietitian to determine dietary calcium intake. Medical history and menstrual periods were recorded. Compliance was determined by monthly pill count.
Bone status evaluation
Bone mineral density (BMD) and bone mineral content
(BMC) were measured at the total-body, lumbar spine (LS),
and femoral neck (FN) sites by dual-energy X-ray absorptiometry (Lunar DPX scanner; Lunar Corp, Madison, WI). The
precision error in vivo was 0.6%, 0.9%, and 1.5%, respectively, for the spine scans (L2-L4) at slow, medium, and fast
speeds, whereas the error was 1.2% and 1.5-2.0%, respectively, for the femur scans at slow and medium speeds. The
precision of total-body bone density was 0.5% in vitro and in
vivo (29, 30). The CV of the BMD measurement at these sites
(as determined in young, healthy adults) is between 1% and
1.6%. The scans were acquired by using the appropriate scan
mode for the patient's weight. The same technician performed
all measurements. Daily quality-control and phantom measurements were performed to ensure the stability of the equipment
during the study period.
Biochemical markers of bone turnover and
calcium-regulating hormones
Bone-specific alkaline phosphatase was assayed by immunoradiometric assay (Tandem-R-Ostase; Beckman Coulter,
Fullerton, CA). Urinary deoxypyridinoline cross-links were
evaluated in the second void collected in the morning after the
subjects had fasted overnight by use of the Pyrilinks-D enzyme-linked immunosorbent assay (Metra Biosystems, Mountain View, CA). Intact parathyroid hormone (PTH) was measured by immunoradiometric assay (Nichols Institute
Diagnostics, San Juan Capistrano, CA), and 25-hydroxyvitamin D3 was measured by 125I radioimmunoassay (DiaSorin,
Stillwater, MN). Osteocalcin was assessed in serum collected
in the morning after the subjects had fasted overnight by use of
the radioimmunoassay method (OSTK-PR kit; CIS BioInternational, Paris) (31-34).
Weight and height
The weight of the girls was measured while they were
wearing minimal clothing and no shoes. Weight was recorded
to the nearest 0.10 kg, and standing height was recorded to the
nearest 0.10 cm. Weights and heights were evaluated by the
same operator using the same equipment throughout the study
periods. All tests were performed 3 times: at enrollment and
after 6 and 12 mo.
Statistical analysis
All parameters except for femur BMC had normal distributions according to the Kolmogorov-Smirnov test of normality.
However, because of a few outliers, we preferred using nonparametric analyses. The Mann-Whitney U test was used for
comparisons of percentage increase and absolute increase in
bone measurements between the study and the control groups.
Otherwise, a two-tailed Student's t test was used for comparison of means. A two-factor repeated-measures analysis of
variance was used to assess interactions between variables over
time. A multivariate analysis using linear regression was used
to test the potential effects of different variables on the prediction of percentage gains in BMD and BMC in the different
skeletal areas. All results are given as means ± SEMs. The
level of significance for all tests was P < 0.05.
RESULTS
One hundred girls (76 Jewish girls and 24 Arab girls) completed the 12-mo intervention. The characteristics of the CS
and control groups at the time of inclusion to the intervention
study are shown in Table 1. At this time, there were no
significant differences between the CS and control groups in
age, body mass index, height, weight, time since menarche,
calcium intake, energy intake, biochemical markers of bone
turnover and calcium-regulating hormones, or BMD and BMC
of the LS (L2-L4), FN, and total body. No significant change
in anthropometric variables (weight and height) was observed
during the study period within and between the CS and placebo
groups.
View this table:
TABLE 1. . Baseline characteristics of the placebo and calcium-supplemented groups
at the time of inclusion in the intervention study1
Compliance with treatment
Twelve girls dropped out of the study. When all of the
variables shown in Table 1 were examined, there were no
significant differences between the 12 girls who did not complete the study year and the 100 who did. The reasons for
dropping out varied.
During the calcium supplementation trial, compliance with treatment was evaluated monthly by pill count. The mean compliance rate of the entire cohort was 67.33 ± 2.5%. There was no significant difference between the compliance rate of the CS group and that of the control group, but there was a significant change in compliance during the research year. Compliance dropped from 71 ± 26% during the initial 6 mo to 56 ± 34% for the remaining study period (P = 0.0001). There was no significant difference in calcium intake from diet between the CS and placebo groups, 440 ± 131 compared with 480 ± 118 mg/d, respectively. Calculated mean calcium intake was therefore 1110 ± 292 mg/d for the CS group and 480 ± 120 mg/d for the placebo group. Only 23.3% of the CS group had a calcium intake > 1300 mg/d.
Effect of calcium supplementation on bone mass
At the end of the study year, the accretion of total-body
BMD was higher in the CS group than in the control group
(3.80 ± 0.3% compared with 3.07 ± 0.29%; Table 2). The percentage accretion of BMD in the LS (L2-L4) was higher in
the CS group than in the placebo group (3.66 ± 0.35% compared with 3.00 ± 0.43%). The percentage accretion of BMD
in the FN tended to be higher in the CS group than in the
placebo group, but this difference was not significant.
View this table:
TABLE 2. . Percentage accretion of bone mineral density (BMD) at 6 and 12 mo1
The accretion of LS BMC was significantly higher in the CS
group than in the placebo group after 6 mo (3.53 ± 0.45%
compared with 2.20 ± 0.39%), but the differences were no
longer significant by the end of the intervention (Table 3).
View this table:
TABLE 3. . Percentage accretion of bone mineral content at 6 and 12 mo1
Effect of calcium supplementation on serum hormone
concentrations and markers of bone turnover
Serum PTH concentrations dropped significantly in the CS
group after 6 mo of treatment, by 4.40 pg/mL, compared with
an increase in the placebo group of 2.30 pg/mL. This difference
was no longer significant after 12 mo (decrease of 1.98 pg/mL
in the CS group compared with an increase of 2.96 pg/mL in
the placebo group).
Bone turnover as determined by serum osteocalcin concentrations was reduced significantly in the CS group after 6 mo of treatment (a decrease of 1.78 ng/mL when calculated from absolute values; P < 0.001), but did not change significantly in the placebo group (an increase of 0.19 ng/mL; Table 4).
View this table:
TABLE 4. . Comparison of plasma parathyroid hormone (PTH), osteocalcin, and bone-specific alkaline phosphatase (BAP) concentrations between the calcium-supplemented and placebo groups throughout the trial1
Concentrations of bone-specific alkaline phosphatase also
dropped significantly in the CS group after 6 mo of treatment
(by 4.27 ng/mL compared with 2.07 ng/mL in the placebo
group) and even more so by the end of the study year (by 7.10
ng/mL compared with 4.58 ng/mL in the placebo group).
Urinary concentrations of deoxypyridinoline cross-links
dropped in both groups, and serum concentrations of 25-hydroxyvitamin D3 changed according to season but were not
affected by group (data not shown).
Effect of calcium supplementation and time since
menarche on bone mass and bone turnover.
There was a significant interaction between time since menarche and treatment group on total-body mass accretion at the
end of the study period (P = 0.014). When the group was
divided by time since the onset of menarche [ie, by postmenarcheal age (PMA)], 1 y of calcium supplementation had no
benefit for girls in the 24 mo PMA group (total-body BMD
of 4.1 ± 2.1% in the CS group compared with 4.3 ± 1.5% in
the placebo group; NS) but was very beneficial for girls in the
> 24 mo PMA group (total-body BMD of 3.8 ± 1.9% in the
CS group compared with 2.1 ± 1.8% in the placebo group;P = 0.003; Figure 1). A similar, though nonsignificant, trend
was observed for LS BMC.
FIGURE 1.. Mean (± SEM) percentage gains in total-body bone mineral density during 1 y of treatment with calcium supplementation () or
placebo () by time since menarche. In the group 24 mo postmenstrual
age, n = 25 in the supplemented group and 28 in the placebo group; in the
group > 24 mo postmenstrual age, n = 28 girls in the supplemented group
and 24 girls in the placebo group. *Significantly different from the placebo
group, P = 0.003. There was a significant interaction between time since
menarche and treatment group on total-body mass accretion at the end of
the study period, P = 0.014 (Mann-Whitney U test).
Repeated-measures analysis of serum osteocalcin concentrations at inclusion and 12 mo showed a significant interaction
between time and age since menarche (P = 0.015) but no
significant interaction between group and age since menarche.
Bone turnover was significantly different as determined by
results for serum osteocalcin concentration at inclusion (14.3 ±
0.47 ng/mL compared with 12.5 ± 0.29 ng/mL in girls 24
mo PMA compared with > 24 mo PMA, respectively; P =
0.003) but were almost equal at 12 mo. This difference can be
explained by the increase in time since menarche in the 24
mo PMA group by the end of the year and was not related to
treatment. A multivariate regression model showed only time
since menarche and treatment group to be significant for the
BMD and BMC changes found.
Regression model
A multivariate analysis using linear regression was conducted to determine the most important parameters affecting
change in BMD during the intervention year. The model included change in weight and height during the intervention
period, group (CS or placebo), change in bone turnover as
determined by osteocalcin and deoxypyridinoline cross-links,
change in serum PTH concentrations, PMA, ethnic group, and
other demographic variables. The most dominant variants positively influencing change in BMD were change in weight
during the research year and BMD at the time of inclusion in
the study. Calcium supplementation was significant in most
models, as was PMA. Markers of bone turnover and calcium-regulating hormones were not significant predictors of change
in BMD during the intervention period.
DISCUSSION
To the best of our knowledge, the present study is the first
randomized controlled supplementation trial in adolescent
postmenarcheal girls consuming a diet naturally low in calcium. The results of this study show that calcium supplementation significantly enhanced bone acquisition at the LS and the
total body after 6 mo; after 12 mo, significant increases were
observed for BMD but not for BMC. The percentage gain in
bone mass was less pronounced in our study than in previously
reported calcium supplementation trials in younger age groups
(15-22). A possible explanation of this difference is that the
subjects of the present study were adolescent, postmenarcheal
girls, whereas in most other studies the subjects were prepubertal or were undergoing puberty, a time of a natural growth
spurt and therefore more pronounced bone gain. The fact that
BMC measurements were no longer significant after 12 mo
may be related to the significant decline in compliance with
treatment. Another possible reason for this result is the habitually low calcium intakes of this group; even with the supplementation, some girls in the CS group were not receiving
adequate amounts of calcium to accommodate their growth
potential during rapid skeletal modeling.
During the first 6 mo of the intervention study, there was a significant reduction in serum PTH, bone-specific alkaline phosphatase, and serum osteocalcin concentrations in the CS group. Previous reports showed that a positive effect on bone mass was achieved when bone turnover declined (35-37). In an intervention study of calcium supplementation in subjects with low calcium intake, a decline in bone turnover was discussed as a possible mechanism of bone gain (18). In our study, however, we believe that the bone gain was not due to a reduction in bone turnover for 2 reasons: 1) the significant reduction in PTH and serum osteocalcin concentrations was cancelled out by the end of the study year and 2) a multivariate model ruled out the change in PTH and serum osteocalcin concentrations as variables affecting change in bone mass.
An unexpected result of our research was that calcium supplementation was more beneficial to girls with longer a PMA (> 24 mo postmenarcheal) than for girls with a shorter PMA ( 24 mo postmenarcheal). Even though girls with a longer PMA had a natural deceleration in bone turnover (because they were further from the growth spurt of puberty), calcium supplementation elevated bone gain to a level equal to that of the group with a shorter PMA. Calcium supplementation acted to decrease the deceleration of bone accretion that occurs with increase in age and to create a difference between the CS and control groups that was similar to the natural difference occurring with age, as seen in the control group when comparing girls by PMA. This difference existed even though the girls who were > 24 mo postmenarcheal were significantly older (mean age of 15.2 ± 0.59 y in girls > 24 mo postmenarcheal compared with 14.6 ± 0.64 y in girls 24 mo postmenarcheal; P 0.0001) and had less chance of bone accretion as the result of the natural decrease in bone turnover.
We believe that the effect of the total calcium dose consumed by the girls with initially low calcium intakes was more beneficial for the girls with lower bone turnover and consequently lower calcium requirements (with a higher PMA) than for girls with an earlier PMA. These results are encouraging clinically and indicate that there is an extended window of opportunity for bone acquisition than previously reported or questioned (23, 26, 38-41). This finding, although in contrast with the results of other studies, supports observations from a longitudinal study that showed a positive effect of calcium on bone gain during the third decade of life (42). This may have significant clinical implications. For example, calcium supplementation in cases such as late recovery from adolescent anorexia nervosa may prove beneficial to bone health (43, 44).
In a multivariate regression model, calcium supplementation was a significant factor influencing bone acquisition. According to our calculations, the positive effect of calcium supplementation on bone accretion was achieved at an average calcium intake of 1200 mg/d: 500 mg from the low-calcium diet plus 70% compliance with the supplement, equal to 700 mg Ca. This finding supports the new recommendations for calcium intake in adolescent diets (45) and is supported by other publications (6, 23).
In conclusion, the association between calcium intake and bone health at all ages is well established, yet a positive effect on bone accretion is believed to be limited to the early stages of sexual maturation, especially in girls. A unique, not yet fully explained finding in our study was the benefit of calcium supplementation on bone mass accretion in girls =" BORDER="0"> 2 y past the onset of menarche.
ACKNOWLEDGMENTS
SI-S designed the study, obtained the research grant support from the
Chief Scientist Fund of the Israeli Ministry of Health, and supervised
GSR's work and the writing of the manuscript. GSR did field research
work for the project and writing as part of the requirements for a doctoral
thesis. GR participated in the supervision of the research and statistical
analysis. HSR performed the statistical analysis. The follow up for the
Arab speaking group was performed by GD. All laboratory work was
performed by BR. RPD-G contributed to the fieldwork in the study, and
NI-S performed the pediatric follow-up for the study group. None of the
authors had any financial or personal interest in this research. All work was
purely based on academic interest.
REFERENCES