| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
St. Joseph Hospital (D.S., R.E., E.S.P., K.M., D.V., C.P., C.K., C.J.R.), Bangor, Maine 04401; J.L. Pettis Veterans Administration Hospital (S.M.), Loma Linda, California 92357; and Boston University School of Medicine (T.C., M.F.H.), Boston, Massachusetts 02118
Address all correspondence and requests for reprints to: Clifford J. Rosen, Maine Center for Osteoporosis Research and Education, St. Joseph Hospital, 360 Broadway, Bangor, Maine 04401. E-mail: crosen{at}maine.maine.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Seasonal bone loss associated with secondary hyperparathyroidism and increased bone turnover has been described in older women living in northern latitudes (9, 10). The extent to which this contributes to the overall risk of bone loss and fractures is still not clear, although overt vitamin D deficiency as a result of reduced sunlight exposure and/or poor dietary intake can lead to rapid bone loss and eventual osteomalacia (11, 12). Previously, we described seasonal changes in BMD, PTH, and 25-OH vitamin D (25(OH)D) in healthy elderly postmenopausal women living in northern New England (13). Those women were consuming little calcium in their diets, and it was hypothesized that poor calcium intake combined with a seasonal drop in 25(OH)D worsened secondary hyperparathyroidism and thereby accelerated age-related bone loss. The purpose of the current study was to determine whether calcium supplementation, by either calcium carbonate (CaCO3), or by consuming four glasses of milk fortified with vitamin D per day (D group), could prevent bone loss and suppress markers of bone turnover, especially during winter months. In addition, we wanted to test the hypothesis that during winter, serum binding protein 4 (IGFBP-4), an inhibitory peptide that blocks IGF actions on bone, would increase, thereby limiting the ability of remodeling cells to stimulate new bone formation. Therefore we designed a 2-yr randomized, placebo-controlled trial of calcium supplementation in older postmenopausal women living in northwestern Maine.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Sixty physically active healthy Caucasian women from three rural communities in northern Maine were recruited from a cohort of over 300 potentially eligible subjects. Initial screening of all postmenopausal women over the age of 65 yr in Dover-Foxcroft, Greenville, and Skowhegan, Maine (latitude: 45.5 degrees N) occurred over 12 months before the start of the study. Medical records were examined and individual women were contacted by postcard. Inclusion criteria included: age over 65 yr, no antiosteoporosis treatments in the last 10 yr, t scores of the femoral neck (FN) that were higher than B2.5, good overall health, complete living independence, calcium intake <800 mg/day as measured by a food frequency questionnaire (FFQ), and no plans to travel south of the Mason-Dixon line (Maryland-Pennsylvania border) during two consecutive winter seasons. Also, women had to agree to be randomized to either placebo calcium, calcium carbonate, or dietary calcium supplementation and to withhold any vitamin supplementation. Women with known osteoporotic fractures, diabetes mellitus, renal insufficiency, any recent malignancy (excluding basal cell carcinoma), or congestive heart failure were excluded. Thiazide use and smoking were not exclusionary. All eligible women signed informed consent from the Institutional Review Board of St. Joseph Hospital, which approved the study and the protocol each year for three consecutive years.
The sixty women were randomized in blocks to one of the three
interventions with 20 women in each group. Sixty women started the
study and 53 completed the study after 2 yr. All records on each
subject, including adverse events, were kept in a locked room and were
coded by number rather than by name. There were 3 drop-outs in the
placebo (P) group [1 because of gastrointestinal side effects and 2
because of concurrent illnesses (i.e. stroke and myocardial
infarction)]; 1 drop-out in the milk-supplemented group (Dietary-D),
because of refusal to participate after 1 yr; and 3 drop-outs in the
CaCO3 group after 12 months (1 for myocardial infarction, 1
who moved out of state, and 1 because of a spouse illness). None
of the subjects who left the study were available for follow-up bone
density or biochemical measures at 24 months. The demographics of the
60 women who were randomized at baseline are noted in Table 1
.
|
Women were randomized to one of three treatment arms: 1) The placebo group (P), which used a CaCO3-matching placebo supplement (Rhone-Polenc-Rorer, Collegeville, PA) taken twice per day with meals; 2) The calcium-supplemented group (CaCO3), which included a 500-mg tablet of CaCO3 (Smith-Kline Beecham, PA) taken twice per day with meals; or 3) the milk-supplemented arm (Dietary-D), a group that supplemented their intake with dietary calcium, principally in the form of four 8-ounce glasses of milk per day. The pill groups were double-blinded, whereas the milk-supplemented group was open-label but blinded to the principle investigator, the dual energy x-ray absorptiometry technician, and the data collectors. Stipends were provided for buying milk at the local grocery. Three brands of milk were consumed by the diet-supplemented group, and the most popular brand was tested for vitamin D content. Receipts that included a listing of milk products were submitted quarterly to the investigators. Milk labels were also collected. Every 3 months women were educated in good dietary practices and informed about recent information concerning osteoporosis. Women were told to avoid any vitamin supplementation.
Dietary records and screening
Eligible women were initially screened with a FFQ developed and validated in elderly women by our group (14). At entrance into the study, 4-day food records were obtained on each subject; this was repeated at 6-month intervals throughout the study along with the FFQ. No attempt was made to modify diets once subjects were randomized to the placebo- or CaCO3-supplemented groups; in the dietary group, women received special instructions about their diets at 0, 6, 12, and 18 months plus frequent phone calls about methods of increasing dietary calcium intake. All women met with registered dietitians every 3 months for educational support. Four-day records were coded by a graduate student and analyzed using the Nutritionist III software program (Silverton, OR). FFQs were tabulated by a dietitian. Calcium intake reported in this study was derived from the 4-day records, even though there was a strong correlation between FFQ and the 4-day food records for Ca intake (r = 0.65, P < 0.001). A validated exercise questionnaire was also administered to all women at baseline and every 6 months. Results were reported in minutes per week of exercise and intensity of physical activity.
BMD
Bone density of the spine (L2-L4) and femur [greater trochanter (GT) and FN] were done at 6-month intervals beginning in August of 1993 until August of 1995. Scanning was performed in August, February, August, February, August on a Lunar DPX-L (Lunar, Madison, WI) by one experienced technician at St. Joseph Hospital. She remained blinded to treatment groups. The subjects were not notified of their BMD results until 6 months after the close of the study. Quality assurance was maintained by daily calibration and use of two phantoms, one employed in an ongoing randomized trial of a bisphosphonate. The scans were performed in the morning. Short-term in vivo precision for the spine and hip (GT and FN) was 1.5%, 2.5%, and 2.7% respectively. Long-term precision (1 month) for individual subjects in the study for the spine was 2.0% and 2.5% for the GT. In this paper, results of the lumbar spine (L2-L4), FN, and GT of the hip are presented. The total femur BMD program was not available at the time this study was conducted.
Laboratory studies and biochemical markers
Screening. Screening of serum samples (SMA-20) using a (Beckman Inc., Brea, CA) multichannel instrument was employed for serum calcium, protein, and albumin measurements. Urinary calcium/creatinine ratios were measured on 24-h samples delivered on the day of BMD measurements. Urinary calcium was measured by an automated colorimetric system and creatinine by the Kodak instrument. Complete blood counts were done at the start and completion of the study.
Markers of bone turnover. All assays were performed in batches at the laboratory of the Maine Center for Osteoporosis Research and Education. Osteocalcin (OC) was determined on freshly thawed samples using an RIA kit from Diagnostics Systems Laboratory (Webster TX). The intra- and interassay coefficients of variation (CVs) were 6.5% and 8.8%, respectively. Urinary N-telopeptide (Ntx) was measured by an enzyme-linked immunoassay (ELISA) from Ostex (Seattle, WA) on 24-h urine collections. Results are reported as bone collagen equivalents (BCE) per millimole of creatinine. Intra- and interassay CVs were 8.0% and 10.0%, respectively. PTH was determined by an immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). Intra- and interassay CVs were 2.4% and 5.0%, respectively. Serum 25(OH)D, was measured by competitive protein binding in the laboratory of Dr. Michael Holick. The intra- and interassay CVs were 4% and 10%, respectively. IGF-I was measured by a RIA after acid-ethanol cryoprecipitation using a kit from Nichols Institute Diagnostics. The inter- and intraassey CVs have been reported previously for this cohort and were approximately 5.6% and 3.0%, respectively (15). IGFBP-4 was determined by a RIA developed and validated in the laboratory of Dr. S. Mohan. Inter- and intraassay CVs have been reported previously and were approximately 8.0% and 5.0%, respectively (16).
Statistical analyses
The three treatment groups were analyzed according to their original randomized assignment with intention to treat analysis conducted at the end of the study. All results in the text and graphs are expressed as the mean ± SEM. Changes in BMD are represented as percent change from baseline in the figures as well as the absolute change in BMD in grams per centimeter squared in the tables. Statistical significance by treatment group was assessed for percent change and absolute change by site using one-way ANOVA. Paired Student’s t test was employed to assess differences from baseline. Change in biochemical markers by group were also analyzed by one-way ANOVA using a PC software program (Instat II, San Diego, CA). Ninety five percent confidence intervals are reported for seasonal changes in L-S spine, femoral BMD, GT BMD, PTH, and 25(OH)D. Univariate and multivariate regression analyses were performed using SAS software (Cary, NC) at the Public Health Research Institute in Portland, ME. The independent variables were calcium intake, 25(OH)D, Ntx, PTH, OC, IGF-I, IGFBP-4, age, body mass index (BMI), thiazide use, smoking history, and baseline BMD. Dependent variables were percent change in BMD and final BMD of the spine, FN, and GT. Stepwise regression was performed starting with univariate analysis of calcium intake vs. change in BMD. Other variables were subsequently added to the model. Statistical significance was assigned at a P < 0.05 for all analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The baseline characteristics of the subjects who entered the study
are noted in Table 1
. There were no statistical differences in any
parameter noted in Table 1 among the various treatment groups. Smoking
history was present in 7 women, although none currently. Current
thiazide use was noted in 19 women. BMD at the spine, FN, and GT did
not differ among the groups at baseline when expressed as z
scores (data not shown) or in grams per centimeter squared as noted in
Table 1.
Intervention end points
Dietary intake. In each group dietary intakes of calcium as
measured by 4-day food records over the 2 yr of the study were
relatively consistent, except during the 2nd yr when the
milk-supplemented group (D group) consumed less calcium than in year 1
(Fig. 1A). As projected a priori, total calcium intake
differed in a stepwise fashion: placebo, 699 ± 49 mg/day
vs. dietary, 1052 ± 118 mg/day vs.
CaCO3, 1678 ± 57 mg/day; P < 0.001
(Fig. 1A). Those intakes were consistent with mean 24-hr urinary
calcium/creatinine excretions (Fig. 1B), which showed an increase in
the CaCO3 group (P < 0.02 vs.
baseline) and a decrease in the P-treated women (P <
0.02 vs. baseline and P < 0.01
vs. CaCO3). Based on activity recalls from the
questionnaire, there were no differences in exercise duration or
intensity among the three groups at any point during the study.
|
), with the
group consuming the greatest amount of calcium (CaCO3)
showing a significant increase in lumbar BMD (+3.7 ± 0.9% over
baseline; P < 0.01). There were no significant changes
after 2 yr in L-S BMD from baseline for women in the placebo
(P = 0.45) and dietary (P = 0.20)
groups. A similar trend was noted for GT BMD, with the greatest
decrease in bone density noted in the placebo group (-3.0 ±
1.9% from baseline; P < 0.05) and a stepwise increase
in GT BMD in those women consuming the most calcium (CaCO3)
(Fig. 2B and Table 2). The changes in GT
BMD in CaCO3- and dietary-supplemented women were not
statistically different from each other (P > 0.20) or
from baseline values (P > 0.30 for CaCO3
vs. baseline; P > 0.20 for dietary
vs. baseline). Like lumbar BMD, FN BMD increased in the
CaCO3-treated group significantly (+3.0%;
P = 0.05 vs. baseline) and there was a trend
for a significant difference between CaCO3 and the other
two groups (CaCO3, +3.0%; placebo -0.3%; dietary,
-1.8%; P = 0.10 by ANOVA) (Table 2). By comparing the
change in FN-BMD over 2 yr in those women receiving CaCO3
supplementation (>1500 mg/day of calcium) vs. those
receiving either placebo or milk (D group) (<1000 mg/day), there was a
statistical difference for the change in FN-BMD (P <
0.03).
|
|
. GT bone loss
occurred almost exclusively during the winter months. FN BMD also
varied by season, with the most loss during winter months. During the
winter months, CaCO3-treated women did not lose bone
(+0.019 ± 0.007 g/cm2), whereas the placebo- and
dietary-treated women exhibited slight bone loss (placebo, -0.007
± 0.006 g/cm2; dietary, -0.003 + 0.008
g/cm2; P < 0.03 by ANOVA).
|
). During summer, serum 25(OH)D levels returned to
baseline (Table 3 and Fig. 4). Serum PTH levels
increased 45% (P < 0.001) over the 24 months in all
women (Table 3) but the greatest increase was noted in the placebo
group (+49 ± 9%; P = 0.01 vs baseline). All of
the increases in PTH occurred during the winter months and there was a
strong inverse relationship between change in 25(OH)D and change in PTH
among women in the placebo group during the winter months (r =
-0.56, P < 0.02). But in contrast to 25(OH)D, PTH
levels did not return to baseline during the summers (Table 3). Indeed, for most women, serum PTH did
not change from winter to summer. However, during the 2nd yr, from
winter to summer PTH levels were suppressed by 10% in the
CaCO3-supplemented group (41.2 ± 2.5 to 37.5 ±
2.2 pg/mL; P = 0.05).
|
|
). All of the changes in OC occurred during the winter
months (P < 0.001), whereas during the summer, OC
levels remained stable. Similarly, changes in urinary NTx for 24 months
were dependent on group assignment. P-treated women showed a modest
38 ± 18% increase (P < 0.05 vs.
baseline) in urine NTx, whereas there was no change in the
dietary group and a slight suppression of NTx in the
CaCO3-treated women (34 ± 4 BCE/nm creatinine
(CR) to 28 ± 5 BCE/nmCR, P < 0.01
vs. baseline) (Fig. 6). When changes in urine
NTx were examined as the mean of two winters and two summers, there was
a 45 ± 15% (P < 0.009) increase from summer to
winter in women treated with placebo, whereas during the summers,
significant changes in NTx were not noted in any group.
|
|
In sum, major changes in BMD and markers of bone turnover occurred in
women with the lowest calcium intake (P group). The changes in
biochemical markers among women consuming the least amount of calcium
are best summarized in Figure 7 where individual
responses in the placebo group are noted. Consistently, these changes
occurred only during the winter months.
|
The other biochemical marker that correlated with hip bone density was urinary NTx. After the first time point, urine NTx values showed a progressively stronger relationship to GT BMD time point 1 (winter 1) (P = 0.10), time point 2 (P = 0.01), time point 3 (winter 2) (P = 0.003) and time point 4 (P = 0.003). However, for percent change in GT BMD, NTx was predictive (P = 0.05) only during the 1st yr of the trial. Putting serum 25(OH)D levels into the model did not change the significant effects of NTx on GT BMD.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This study differed from those trials in part because we focused on seasonal bone loss and the effects of calcium supplementation on bone turnover in relation to those seasonal changes. In this study, we did not select fractures as an end point nor did we examine whether women had prevalent fractures at randomization. But like those other studies, we also were able to show significant femoral bone loss in older women consuming <800 mg of calcium/day. Furthermore our trial demonstrated that adequate calcium supplementation alone could prevent seasonal bone loss despite rather significant declines in serum 25(OH)D. Finally, it is apparent from this study that calcium supplementation can prevent seasonal increases in bone resorption and changes in serum IGFBP-4. However, we cannot extrapolate from these data that healthy older women supplemented with 1000 mg/day of CaCO3 will have less osteoporotic fractures; but because high turnover is a strong risk factor for bone loss and may also be an independent predictor of future fractures in the elderly, it is conceivable that this intervention will be shown to prevent fractures (4, 5, 25).
Although this study reinforces the importance of calcium
supplementation in elderly women, there are several new findings that
require more careful examination. First, we recruited healthy elderly
women without overt osteoporosis or a history of osteoporotic
fractures and with normal indices of bone turnover (Table 1).
Yet, it is clear that even in this group of older postmenopausal women,
low calcium intakes combined with seasonal changes in 25(OH)D can
result in accelerated bone turnover and bone loss. Second, in contrast
to other studies, we focused on seasonal changes in bone turnover and
how these indices affected the rate of bone loss and the pattern of
changes in other biochemical indices. Our data suggest that seasonal
effects on bone mass may be even greater than previously appreciated.
For example, serum 25(OH)D declined by 20–25% during winter, although
in all groups it returned to baseline during the summers. On the other
hand, serum PTH rose 20% each winter but did not return to baseline
values during the summer months. This stepwise pattern of winter
increases also held for OC, which went up 70% in the placebo-treated
women, and for urinary Ntx excretion, which rose nearly 40% in women
consuming calcium <800 mg/day. Thus, bone turnover in elderly
women consuming minimal amounts of calcium may be reset at a
significantly higher level during winter. Furthermore, calcium
supplementation may be most appropriately administered to older women
during the winter when 25(OH)D levels are at their nadir, PTH is
maximally stimulated and markers of bone turnover are increased.
Third, this is the first study to report longitudinal changes in the serum IGF regulatory system in older postmenopausal women in respect to bone density and biochemical markers of bone turnover. Although there was only a marginal decline in serum IGF-I in the placebo-treated women, seasonal changes in serum IGFBP-4 were significant and could be important in understanding how rapid bone loss may occur during the winter. IGFBP-4 is a 24-kilodalton IGF binding protein that circulates in relatively large concentrations (7). It is synthesized in numerous tissues but a major site of production is the skeleton (7). IGFBP-4 binds to IGF-I and inhibits IGF skeletal activity, especially with respect to the recruitment and differentiation of osteoblasts (7, 16). In vitro, PTH and 1,25 dihydroxyvitamin D induce IGFBP-4 synthesis and secretion (26). Two studies (16, 27) have reported an age-associated rise in serum IGFBP-4, and in one preliminary study (27), serum IGFBP-4 levels were very high in hip fracture patients. This has led investigators to propose that calcium deficiency stimulates PTH release, which increases bone resorption but at the same time induces IGFBP-4 synthesis, thereby limiting bone formation (27). This scenario could also apply to elderly women living in northern latitudes who have low calcium intake. In our study, women with calcium intakes below 1 g/day during the winter showed a marked increase in serum levels of IGFBP-4 (possibly as a result of PTH stimulation), which could limit activation of bone formation. When reduced bone formation is combined with high rates of bone resorption, rapid bone loss could result. Further studies will be required to assess the significance of changes in serum IGFBP-4 in elderly individuals and the relationship of serum IGFBP-4 to tissue levels of this peptide. Currently, studies are ongoing in larger cohorts to assess the predictive value of IGFBP-4 for fracture risk.
Finally, changes in PTH during this study are instructive and may clarify the complex regulation of this calciotropic peptide in elderly women. Both Riggs et al. (22) and Dawson-Hughes et al. (21) reported a decline in PTH of 5–15% during calcium supplementation with or without vitamin D. In our study, PTH levels rose in all three groups by 20–25% during winter irrespective of treatment arm, but did not return to their starting values during the summer, despite improvement in serum 25(OH)D. The reasons for this are not entirely clear, but several possibilities exist. First, there may be a threshold dose of calcium that is required to suppress PTH during states of relative vitamin D deficiency. In the Riggs et al. (22) trial, total calcium intake in the treated group exceeded 2400 mg/day, and that group noted a 5% reduction in serum PTH. In the trial conducted by Dawson-Hughes et al. (21), total calcium intake was only 1300 mg/day, and yet PTH levels were suppressed by nearly 20% in men. But those subjects were also supplemented with 700 IU of vitamin D per day. Therefore, conclusions about a threshold dose for calcium supplementation in this age group are still not clear. Second, it is conceivable that changes in serum 25(OH)D drive the increase in PTH, rather than overall daily calcium intake. Several lines of evidence support this possibility. For example, it has been reported that in primary hyperparathyroidism, higher PTH levels are noted in those individuals who are 25(OH)D deficient (28). Also, cross-sectional studies have noted a strong inverse relationship between PTH and serum 25(OH)D (29). In addition, Ooms et al. (30) recently reported that vitamin D supplementation without calcium suppressed PTH levels by 10%. In our study, we noted a strong inverse relationship between changes in 25(OH)D and changes in PTH in women consuming <800 mg/day of calcium. This would suggest that PTH may be very responsive to endogenous changes in vitamin D status. Clearly, further studies will be needed to investigate how PTH is regulated in older postmenopausal women.
There were several limitations to this study. First, the number of
subjects was relatively small compared with other randomized trials.
Despite that, changes in spinal and FN BMD in the
CaCO3-supplemented group were significant, and the overall
trend by group followed that noted for the GT. Still, there can be
significant measurement variance over time at other skeletal sites.
Therefore, effects of calcium supplementation on seasonal changes in
bone mass at primarily cortical sites in elderly women residing in
northern latitudes will require a larger study and will need to use
sites such as the wrist and total body. The small number of subjects in
this study is also relevant for another reason, i.e.
compliance. Although we believe overall participation in this trial was
better than projected, and the number of drop-outs after 2 yr was lower
than in other intervention trials (14%), the variability in dietary
intake makes conclusions about dairy supplementation with milk somewhat
problematic. But it would appear from our data that by increasing
daily milk consumption, women can prevent bone loss from the GT.
However, the number of subjects was small and there were significant
changes in dietary behavior after 1 yr that resulted in a lower mean
calcium intake in the dietary group than predicted a priori
(i.e. women in the dietary group consumed 
1 g/day of
calcium during the 2nd yr but 1200 mg/day during the 1st yr; Fig. 1).
In fact, three women in the dietary group had daily calcium intakes
<800 mg/day. Thus a much larger study of milk supplementation will be
required to assess the efficacy of this form of intervention in respect
to bone density. Third, as noted above, we did not examine the effects
of calcium supplementation on fracture rates of the spine or other
nonvertebral sites. Therefore firm conclusions about the effects of
seasonal variation on fracture risk require further studies
Finally, seasonal declines in serum 25(OH)D in the dietary-supplemented group were not prevented despite an increase in vitamin dietary-fortified milk consumption. To ascertain the reason for this finding, we measured vitamin D in a quart of skimmed milk consumed by the dietary group and found an approximate concentration of 325 IU of vitamin D/quart. Assuming adequate bioavailability from milk, this should have led to a less dramatic decline in serum 25(OH)D during winter in the D group. However, we only tested 1 quart of skimmed milk, and it is very likely that not all of the milk contained exactly the same amount of vitamin D per quart (31). Moreover, many women did not end up drinking four glasses of milk per day but rather only two or three glasses per day. Furthermore, based on recent studies, it is now clear that most forms of dietary supplementation will not be able to provide adequate vitamin D, even if women are able to consume four glasses of milk per day (29). For example, Dawson-Hughes et al. (32) showed that 700 IU of vitamin D (but not 200 IU) was sufficient to reduce (although not eliminate) femoral bone loss as well as to prevent seasonal changes in 25(OH)D and PTH. Recently, the Institute of Medicine modified their recommendation for vitamin D and suggested that 600 IU/day was the minimum requirement for men and women over age 70 (33). This would require older women living in northern latitudes to drink six glasses of milk per day to get adequate vitamin D from their diets. Based on our previous cross-sectional study of a cohort almost identical to this one, there is virtually no sunlight-induced vitamin D production in skin during the winter months in northern Maine (13). Hence, there is a need for dietary vitamin D supplementation to maintain serum 25(OH)D at summertime levels. However, the optimal vitamin D dose will have to be determined by larger dose-response studies.
In summary, we report that calcium supplementation prevents bone loss in elderly postmenopausal women primarily by slowing bone resorption during the winter months. From this study, it appears that 1000 mg/day of calcium supplementation is sufficient, even without added vitamin D, to prevent winter bone loss. Moreover, increased bone resorption, possibly coupled to changes in bone formation mediated through the IGF regulatory system, is the most likely mechanism responsible for seasonal changes in femoral bone mass among elderly women living in northern latitudes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received June 2, 1998.
Revised August 5, 1998.
Accepted August 12, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Y. Manios, G. Moschonis, G. Trovas, and G. P Lyritis Changes in biochemical indexes of bone metabolism and bone mineral density after a 12-mo dietary intervention program: the Postmenopausal Health Study Am. J. Clinical Nutrition, September 1, 2007; 86(3): 781 - 789. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Diorio, S. Berube, C. Byrne, B. Masse, N. Hebert-Croteau, M. Yaffe, G. Cote, M. Pollak, and J. Brisson Influence of Insulin-like Growth Factors on the Strength of the Relation of Vitamin D and Calcium Intakes to Mammographic Breast Density Cancer Res., January 1, 2006; 66(1): 588 - 597. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M Weaver and J. C Fleet Vitamin D requirements: current and future Am. J. Clinical Nutrition, December 1, 2004; 80(6): 1735S - 1739S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.-H. Lin, F. Ginty, L. J. Appel, M. Aickin, A. Bohannon, P. Garnero, D. Barclay, and L. P. Svetkey The DASH Diet and Sodium Reduction Improve Markers of Bone Turnover and Calcium Metabolism in Adults J. Nutr., October 1, 2003; 133(10): 3130 - 3136. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Feskanich, W. C Willett, and G. A Colditz Calcium, vitamin D, milk consumption, and hip fractures: a prospective study among postmenopausal women Am. J. Clinical Nutrition, February 1, 2003; 77(2): 504 - 511. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. I. Barr Increased Dairy Product or Calcium Intake: Is Body Weight or Composition Affected in Humans? J. Nutr., January 1, 2003; 133(1): 245S - 248. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Rucker, J. A. Allan, G. H. Fick, and D. A. Hanley Vitamin D insufficiency in a population of healthy western Canadians Can. Med. Assoc. J., June 1, 2002; 166(12): 1517 - 1524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Rapuri, H. K. Kinyamu, J. C. Gallagher, and V. Haynatzka Seasonal Changes in Calciotropic Hormones, Bone Markers, and Bone Mineral Density in Elderly Women J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 2024 - 2032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. P. Heaney Calcium, Dairy Products and Osteoporosis J. Am. Coll. Nutr., April 1, 2000; 19(90002): 83S - 99. [Abstract] [Full Text]
|
|
|
|
|
|
F. L Suarez, J. Adshead, J. K Furne, and M. D Levitt Reply to WB Grant Am. J. Clinical Nutrition, August 1, 1999; 70(2): 302 - 303. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||
, Calcium; •, dietary; 
,
placebo.
represents an individual subject. 
,
Represents mean ± SEM for percent change from
baseline.