| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Leuven University Center for Metabolic Bone Diseases (S.B., D.V.), and Divisions of Geriatric Medicine (S.B.) and Endocrinology (R.B., D.V.), Katholieke Universiteit Leuven, B-3000 Leuven, Belgium; VU University Medical Center (P.L.), 1007 MB, Amsterdam, The Netherlands; Divisions of Aging and Rheumatology, Immunology, and Allergy (H.A.B.-F.), The Robert B. Brigham Arthritis and Musculoskeletal Diseases Clinical Research Center, Brigham and Women’s Hospital, Boston, Massachusetts 02115; and Department of Orthopaedics and Traumatology (P.H.), Vrije Universiteit Brussel, B-1090 Brussels, Belgium
Address all correspondence and requests for reprints to: S. Boonen, M.D., Ph.D., Leuven University Center for Metabolic Bone Diseases and Division of Geriatric Medicine, University Hospital Leuven, Herestraat 49, B-3000 Leuven, Belgium. E-mail: steven.boonen{at}uz.kuleuven.ac.be.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Data Sources: MEDLINE and EMBASE.com (search terms: "vitamin D" and "hip fracture"), bibliographies of articles retrieved, and the authors’ reference files were used as data sources.
Study Selection: Selected studies were randomized controlled trials (RCTs) of oral vitamin D with or without calcium supplementation vs. placebo/no treatment in postmenopausal women and/or older men (
50 yr) specifically reporting a risk of hip fracture.
Data Extraction: Independent extraction was performed by two authors using predefined criteria, including study quality indicators.
Data Synthesis: All pooled analyses are based on random-effects models. Based on four RCTs (9083 patients), the pooled relative risk (RR) of hip fracture for vitamin D alone was 1.10 [95% confidence intervals (CI) 0.89, 1.36]. No between-trial heterogeneity was observed. For the six RCTs (45,509 patients) of vitamin D with calcium supplementation, the pooled RR for hip fracture was 0.82 (95% CI 0.71, 0.94). There was no heterogeneity between trials. In an adjusted indirect comparison of the summary RRs from the two metaanalyses, the RR for hip fracture for vitamin D with calcium vs. vitamin D alone was 0.75 (95% CI 0.58, 0.96).
Conclusions: Our analyses, designed to extend the findings of Bischoff-Ferrari et al., suggest that oral vitamin D appears to reduce the risk of hip fractures only when calcium supplementation is added.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
As negative calcium balance and the resulting secondary hyperparathyroidism, which leads to bone loss and reduced bone strength, are often caused by combined deficiencies of vitamin D and calcium (13), it is important to define the role of additional calcium in preventing hip fractures. We have performed metaanalyses to specifically address this issue. Also, our analyses update previous metaanalyses with three recent large RCTs (10, 14, 15) of vitamin D and calcium, including the Randomized Evaluation of Calcium Or vitamin D (RECORD) (10) and the Women’s Health Initiative (WHI) (15) trials.
Only RECORD has directly compared vitamin D with calcium supplementation against vitamin D alone (10). When direct evidence is considered insufficient, indirect comparisons adjusted by a common control can provide useful information about the relative efficacy of two treatments (16). We performed an adjusted indirect comparison of two pooled-risk estimates: one from a metaanalysis of vitamin D RCTs and one from a metaanalysis of RCTs of vitamin D with calcium, both against placebo/no treatment.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
We conducted a systematic review of English and non-English articles using MEDLINE (PubMed) and EMBASE.com, with the last searches undertaken in March 2006. The search terms were "vitamin D" and "hip fracture." The computerized searches were supplemented by a manual search of the bibliographies of retrieved articles and the authors’ files. Potentially relevant articles (n = 165) were assessed against prespecified eligibility and exclusion criteria (Fig. 1
). Data were independently extracted by two authors (P.H. and S.B.) and checked for accuracy in a second review. Consensus was achieved for all data.
|
We included RCTs, published in full, that assessed the relative risk (RR) of hip fracture in postmenopausal women and/or men aged 50 yr or older receiving oral vitamin D (cholecalciferol, ergocalciferol; any dose) with or without calcium vs. placebo or no treatment. Studies had to report hip fracture as a separate outcome and to detail confirmed fractures in all treatment arms. If a particular patient population was reported in more than one publication, we selected the article providing the most complete follow-up data.
Ineligible studies
Uncontrolled trials, observational studies, and RCTs in which hip fracture was not reported in all treatment arms were excluded. Because placebo/no treatment was the intermediary in the indirect comparison, we also excluded RCTs with an active control, including calcium. RCTs in patients with unstable health or receiving steroids were excluded, because these patients were at higher risk for falls and fractures, potentially confounding the analyses.
Definitions
The primary outcome measure was the RR of hip fracture in patients receiving vitamin D with or without calcium compared with patients receiving placebo/no treatment. Data on all nonvertebral fractures were considered supportive.
Quality assessment
The methodological quality of each study was assessed with respect to the method of randomization, random allocation concealment, blinding of treatment allocation, and study withdrawals.
Studies identified
All studies were identified through our MEDLINE search. Four RCTs of vitamin D alone vs. placebo and six of vitamin D with additional calcium vs. placebo/no treatment (10, 14, 15, 17, 18, 19) met our inclusion criteria. One RCT (10) randomized patients to more than one active treatment arm (vitamin D with calcium or vitamin D alone).
Statistical analyses
Two metaanalyses were conducted; one of the RCTs compared vitamin D vs. placebo/no treatment, the other RCT compared vitamin D and calcium vs. placebo/no treatment. Where appropriate, we also explored the effect of the vitamin D dose. Outcomes were analyzed on an intention-to-treat (ITT) basis using the same standardized end-point definitions as in the primary studies. Because differences between the individual RCTs were expected, the analyses were performed using a DerSimonian and Laird random-effects model (20, 21); an inverse variance fixed-effects model was used in sensitivity analyses (20, 21). However, because the resultant point estimates were essentially similar with random-effects and fixed-effects models, we present only the random-effects analysis that incorporates both between- and within-study variation (providing the more conservative estimate). Pooled-effect estimates are presented as the RR with the 95% confidence interval (CI); all P values are two-sided. Analysis using odds ratios gave very similar results and only the RR data are presented. Also, the number needed to treat (NNT) to prevent one fracture was determined.
The results were examined for heterogeneity by visually examining forest plots, comparing the summary RR using random- and fixed-effects models, and using formal statistical tests for heterogeneity and trial inconsistency. Between-study heterogeneity was assessed using the 
2 distributed Cochran’s Q test (P 
0.10 indicating significance) (20, 21). Because the power of the 2 test to detect heterogeneity is low when there are few trials, as in this review, we quantified heterogeneity by calculating the I2 statistic: values less than 25%, 25–50%, and more than 50% indicate low, moderate, and high heterogeneity, respectively (22).
To assess the RR for hip fracture for vitamin D with calcium vs. vitamin D without calcium, we used the methods of Bucher et al. (23) and Song et al. (16). For this comparison, we calculated the heterogeneity statistic 2 indirect (23) and the corresponding I2 statistic.
Sensitivity analyses, defined a priori, were also undertaken using a weighted least-squares linear regression model (21). The independent variables considered were: sample size, mean age, proportion of women, publication year, trial duration, baseline hip fracture risk, compliance, and withdrawal rates. This was a univariate analysis, because the number of trials provided insufficient power for multivariate testing. To evaluate the impact of individual studies on the overall results, we performed a one-way sensitivity analysis by omitting one study at a time and recomputing the pooled RR. Potential publication bias was explored by drawing a funnel plot (21, 24). Publication bias was formally analyzed using the Begg and Mazumdar (25) and Egger et al. (24) tests. These showed no evidence of publication bias (results available from the authors).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
). All RCTs used cholecalciferol; the dose was 700–800 IU/d or equivalent in six trials (10, 14, 17, 18, 19, 26) and 400 IU/d in the other three trials (15, 27, 28). Calcium (500–1200 mg/d) was given with vitamin D in six RCTs (10, 14, 15), including one arm of RECORD (14, 18, 19). Planned mean treatment duration was 24–84 months.
|
Metaanalysis of RCTs of vitamin D supplementation alone
There were 9083 patients from four RCTs (10, 26, 27, 28) comparing vitamin D alone with placebo (Table 2). No significant heterogeneity was demonstrated between the studies (Q test P = 0.83; I2 = 0%). The pooled RR was 1.10 (95% CI 0.89, 1.36; P = 0.38; Fig. 2 and Table 2). The pooled RR for the two RCTs (27, 28) using 400 IU/d was 1.14 (95% CI 0.87, 1.49) and was 1.04 (95% CI 0.75, 1.46) for the two RCTs using 700–800 IU/d (10, 26).
|
|
Metaanalysis of RCTs of combined vitamin D and calcium supplementation
There were 45,509 patients from six RCTs (10, 14, 15, 17, 18, 19) comparing combined vitamin D and calcium with placebo/no treatment (Table 2
). The vitamin D dose was 700–800 IU/d, with the exception of one trial which used 400 IU/d (15). There was no evidence of statistical heterogeneity (Q test P = 0.39; I2 = 5%). The pooled RR was 0.82 (95% CI 0.71, 0.94; P = 0.0005), suggesting that combined vitamin D and calcium supplementation reduces the risk of hip fracture by 18% compared with no supplementation (Table 2 and Fig. 3A). The pooled-risk difference was 0.4% (95% CI 0.2%, 0.9%; P = 0.0004), giving an NNT to prevent one hip fracture of 276 (95% CI 165, 843) over 24–84 months.
|
) revealed a slightly greater reduction in the hip fracture RR compared with the main analysis (21% vs. 18%). All the RCTs also provided supportive data on all nonvertebral fractures. Statistical heterogeneity was moderate (I2 = 50%; P = 0.08). The pooled RR was 0.88 (95% CI 0.78, 0.99; P = 0.036), suggesting that vitamin D plus calcium supplementation reduces the risk of all nonvertebral fractures by 12%. The pooled-risk difference was 1.4% (95% CI 0.0, 2.84; P = 0.045); therefore the NNT to prevent one nonvertebral fracture was 72 (95% CI 35, 834) over 24–84 months.
Indirect comparison of combined vitamin D and calcium vs. vitamin D alone
The indirect comparison of the pooled estimates for hip fracture from the two metaanalyses gave an adjusted RR of 0.75 (95% CI 0.58, 0.96; P = 0.021) in favor of vitamin D with additional calcium, suggesting that the combination reduces the risk of hip fracture by 25% compared with vitamin D alone. The test for heterogeneity was not statistically significant (Q test P = 0.89; I2 = 0%).
Sensitivity analysis
In the univariate meta-regression sensitivity analysis, no a priori variable was significant: sample size (P = 0.81), mean age (P = 0.88), proportion of women (P = 0.23), publication year (P = 0.32), trial duration (P = 0.51), baseline hip fracture risk (P = 0.13), compliance (P = 0.45), or drop-out rates (P = 0.41). The one-way sensitivity analysis showed no evidence of statistical heterogeneity between the studies in either metaanalysis. In both analyses, the magnitude of resulting RRs was similar to the original pooled RR, with only one 95% CI including unity [when the Chapuy trial (17) was omitted; RR 0.88, 95% CI 0.74, 1.04, P = 0.12].
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Calcium is essential for bone health and slows the loss of bone that occurs in both sexes from the fourth decade (30, 31). However, although our analyses show that calcium plays a critical role when combined with vitamin D, calcium alone appears to be insufficient for fracture prevention. A metaanalysis of studies undertaken by 1998 (9) was unable to meaningfully address the impact on hip fractures, because only two relevant studies had been published, with a total sample size of 222. Since then, RECORD (10) reported no difference between calcium alone and placebo, whereas another well-conducted recent study (11) concluded that calcium was ineffective in preventing clinical fractures in ambulatory patients; hip fracture was not an individual study endpoint. Similarly, the metaanalysis presented here suggests that vitamin D alone does not reduce hip fracture risk.
There is a close physiological relationship between calcium and vitamin D in maintaining bone homeostasis. Vitamin D modulates the intestinal absorption of dietary calcium, with low levels impairing calcium absorption, resulting in a compensatory increase in parathyroid hormone, causing net bone resorption (13, 32, 33). Moreover, with age, the intestinal absorption of calcium in response to vitamin D declines (34). The elderly are at risk of vitamin D insufficiency due to their reduced mobility and consequent reduced exposure to UV light. The skin’s capacity to synthesize vitamin D also declines with age (35). Because vitamin D is needed to maintain calcium homeostasis (8), calcium and vitamin D insufficiency commonly coexist in older adults (18, 36). Intervention studies tend to show that combined supplementation has a greater effect on serum parathyroid hormone than vitamin D alone (37). In parallel with the reversal of secondary hyperparathyroidism, combined vitamin D and calcium increases bone mineral density in the lumbar spine, femoral neck, and hip by 4–6% (38, 39). Falls are a further risk factor for fractures (40). Vitamin D insufficiency is implicated in declining muscular strength, a potential factor in falls in the elderly (37). The benefit of vitamin D in reducing falls in the elderly has been confirmed by metaanalysis (41), with the authors unable to exclude the need for additional calcium.
Therefore, our finding that combined vitamin D and calcium supplementation prevents hip (and other nonvertebral) fracture is supported by a strong biological rationale. It follows from this rationale that trials with vitamin D and calcium may have a negative outcome if the patients are not vitamin D deficient to a degree where a large improvement in secondary hyperparathyroidism can be expected. In addition, poor compliance will impact on outcomes. This may explain why the trials with the most positive outcomes have been performed in nursing homes and in apartments for elderly people (10). Institutionalized elderly people tend to be more vitamin D deficient than those living in the community, and because administration is generally supervised, compliance should be high.
Although the physiological rationale for using combination therapy for fracture prevention is persuasive, the only direct head-to-head RCT vs. vitamin D alone, the RECORD trial (10), found no evidence of any incremental benefit with combined supplementation vs. vitamin D or indeed placebo. One explanation for the apparent contradiction between RECORD and our analyses may be differences in statistical power. Our indirect comparison had a large evidence base: 53,260 patients with 1,186 hip fractures. In contrast, the direct comparison in RECORD involved 2649 patients, with only 93 hip fractures (10). Assuming an RR of 0.75 with combined supplementation vs. vitamin D alone as shown in our analysis, RECORD had only 23% power to detect a similar effect at a significance level of 0.05.
The findings of RECORD may also be partly attributable to the study population: patients were younger than the nursing home residents studied in earlier positive trials (mean 77 yr vs. 84 yr) and living in the community, and therefore, at lower risk for fractures (10, 17, 18). Furthermore, compliance with trial medication was less than 60% (10) and potentially important for a treatment with only modest, although clinically significant, benefits in terms of fracture reduction in higher-risk patients in whom medication administration was supervised.
The WHI trial also targeted postmenopausal women living in the community, 82% of whom were aged less than 70 yr (15). The average baseline calcium intake of participants exceeded 1000 mg/d, close to current recommendations, and 42% had a daily vitamin D intake above 400 IU. More than half were receiving hormone-replacement therapy on study entry, often as part of the active hormone-replacement therapy program of the WHI. The observed hip fracture rate was less than half that projected (16 vs. 34 per 10,000), suggesting that a significant proportion of the women were not vitamin D or calcium deficient. The findings also highlighted the importance of compliance. In an analysis restricted to those who took at least 80% of their assigned medication, vitamin D with calcium reduced the hip fracture RR by 29% (vs. 12% in the ITT analysis) (15).
Thus, RECORD and WHI raise two crucial patient-management issues in the prevention of hip (and nonvertebral) fractures with vitamin D and calcium: appropriate targeting and maximizing treatment compliance. Certainly, although the institutionalized elderly are an appropriate target group, unrestricted supplementation in the community seems unnecessary. However, specific community-based subgroups may benefit; for example, individuals with documented vitamin D and/or calcium deficiency, and patients receiving antiresorptives; RCTs in these subgroups are warranted. When supplementation is indicated, regular patient follow-up and monitoring by physicians can promote long-term compliance (42). Physicians must also ensure patients understand their risk level and treatment goals.
Although evidence from RCTs is generally preferred when evaluating the relative efficacy of treatment options, the indirect approach used here improves statistical power when few direct comparative RCTs are available (16, 43). The strengths of our investigation include the restriction of eligible RCTs to those published in full and reporting hip fracture and adequate statistical power. Causes of noncompliance and study withdrawal were balanced across treatment groups in all trials. No significant heterogeneity between RCTs was demonstrated. Moreover, our findings were robust to different statistical strategies. Finally, all trials were conducted within the past 15 yr, reducing the potential impact of secular trends. However, our study does have limitations. The validity of an adjusted indirect comparison depends on the internal validity of the RCTs involved. The methodology assumes similarity in trial design and methodological quality (16, 43). Another assumption is that the magnitude of the treatment effect is consistent in patients across different trials. In this review, differences in study population, for example, baseline prevalent fracture rates or comorbidity, might explain the extent of fracture reduction, and we acknowledge that we could adjust only for information that was aggregated at the trial level. Importantly, however, sensitivity metaregression analysis showed that our results were not influenced by any baseline characteristic or the trial methodology. Another limitation is the influence of the Chapuy trial (17) on the outcome, although the one-way sensitivity analysis should be interpreted cautiously because exclusion of one study is arbitrary and may introduce unintentional bias. Because most studies did not stratify their data by gender, we could not calculate gender-specific estimates. Finally, our analyses were designed to extend the metaanalysis of vitamin D studies by Bischoff-Ferrari et al. (12) so we did not compare the effects of calcium alone vs. calcium with vitamin D. Therefore, it is not possible to establish the extent to which calcium and vitamin D individually contribute to the treatment effects of combined supplementation. However, as outlined above, we would contend that neither vitamin D nor calcium alone provides a significant benefit in terms of hip fracture reduction in osteoporotic patients but that combination therapy is required to maximize clinical efficacy.
Based on the pooled-risk difference, 276 people need to receive vitamin D plus calcium to prevent one hip fracture. The corresponding NNT for all nonvertebral fractures was 72. Although the NNT for both hip and nonvertebral fractures is relatively high, the low cost of supplements vs. the high economic burden of fractures means that supplementation is cost effective. In clinical practice, we suggest that the vitamin D dose should be 700–800 IU/d. This is in accordance with the recommendations of Bischoff-Ferrari et al. (12) and supported by our exploratory analysis of combined supplementation RCTs using the 700–800 IU/d dose only (comparison of Fig. 3, A and B). With respect to calcium supplementation, we suggest 1000–1200 mg/d of elemental calcium, as used in the four RCTs (10, 14, 17, 18) that contributed more than 99% weighting to the pooled-risk estimate. Most patients with osteoporosis will derive additional benefit in terms of fracture prevention from an antiresorptive or anabolic agent.
Our results for hip fracture may not be generalizable to other types of nonvertebral fractures. Vitamin D alone may be sufficient for the primary prevention of nonvertebral fractures in ambulatory older individuals, as demonstrated by Trivedi et al. (26) in their study of 100,000 IU cholecalciferol every 4 months in community-dwelling older people (RR for fracture at any site = 0.78, 95% CI 0.61, 0.99; RR hip, wrist or forearm = 0.67; 95% CI 0.46, 0.99). One explanation may be that hip fractures tend to occur indoors in frail individuals with a high risk for both calcium and vitamin D deficiency (44), whereas other nonvertebral fractures tend to occur in more active older individuals with a higher dietary calcium intake plus higher 25(OH)D levels (45). However, as many older individuals are at risk for both calcium and vitamin D deficiency, combined supplementation is still warranted for nonvertebral fracture prevention unless dietary assessment indicates a person is calcium-replete.
In summary, our analyses, designed to extend the findings of Bischoff-Ferrari et al. (12), suggest that oral vitamin D appears to reduce the risk of hip (and any nonvertebral) fractures only when calcium is added. Thus, to optimize clinical efficacy, vitamin D 700–800 IU/d should be complemented with calcium, using a dose of 1000–1200 mg/d of elemental calcium. Universal supplementation in the elderly is unnecessary. Further research should identify those individuals who benefit most from vitamin D and calcium supplementation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
S.B. and D.V. are both senior clinical investigators of the National Fund for Scientific Research (Flanders, Belgium) (FWO-Vlaanderen).
First Published Online January 30, 2007
Abbreviations: CI, Confidence interval; ITT, intention-to-treat; NNT, number needed to treat; RCT, randomized controlled trial; RECORD, Randomized Evaluation of Calcium Or vitamin D; RR, relative risk; WHI, Women’s Health Initiative.
Received July 3, 2006.
Accepted January 23, 2007.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Shahnazari, B. R. Martin, L. L. Legette, P. J. Lachcik, J. Welch, and C. M. Weaver Diet Calcium Level but Not Calcium Supplement Particle Size Affects Bone Density and Mechanical Properties in Ovariectomized Rats J. Nutr., July 1, 2009; 139(7): 1308 - 1314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Lorincz, S. L. Manske, and R. Zernicke Bone Health: Part 1, Nutrition Sports Health: A Multidisciplinary Approach, May 1, 2009; 1(3): 253 - 260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Janssens, A. Lehouck, C. Carremans, R. Bouillon, C. Mathieu, and M. Decramer Vitamin D Beyond Bones in Chronic Obstructive Pulmonary Disease: Time to Act Am. J. Respir. Crit. Care Med., April 15, 2009; 179(8): 630 - 636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. A. Bischoff-Ferrari, W. C. Willett, J. B. Wong, A. E. Stuck, H. B. Staehelin, E. J. Orav, A. Thoma, D. P. Kiel, and J. Henschkowski Prevention of Nonvertebral Fractures With Oral Vitamin D and Dose Dependency: A Meta-analysis of Randomized Controlled Trials Arch Intern Med, March 23, 2009; 169(6): 551 - 561. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. A. van Staveren, J. M. Steijns, and L. C.P.G.M. de Groot Dairy Products as Essential Contributors of (Micro-) Nutrients in Reference Food Patterns: An Outline for Elderly People J. Am. Coll. Nutr., December 1, 2008; 27(6): 747S - 754S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Bouillon, G. Carmeliet, L. Verlinden, E. van Etten, A. Verstuyf, H. F. Luderer, L. Lieben, C. Mathieu, and M. Demay Vitamin D and Human Health: Lessons from Vitamin D Receptor Null Mice Endocr. Rev., October 1, 2008; 29(6): 726 - 776. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Scragg and C. A. Camargo Jr Scragg and Camargo Respond to "Physical Activity and Vitamin D" Am. J. Epidemiol., September 15, 2008; 168(6): 590 - 591. [Full Text] [PDF]
|
|
|
|
 |
|
 P. M Brannon, E. A Yetley, R. L Bailey, and M. F. Picciano Overview of the conference "Vitamin D and Health in the 21st Century: an Update" Am. J. Clinical Nutrition, August 1, 2008; 88(2): 483S - 490S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Dawson-Hughes Serum 25-hydroxyvitamin D and functional outcomes in the elderly Am. J. Clinical Nutrition, August 1, 2008; 88(2): 537S - 540S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Hadji, J. -J. Body, M. S. Aapro, A. Brufsky, R. E. Coleman, T. Guise, A. Lipton, and M. Tubiana-Hulin Practical guidance for the management of aromatase inhibitor-associated bone loss Ann. Onc., August 1, 2008; 19(8): 1407 - 1416. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W Nieves and R. Lindsay Calcium and fracture risk Am. J. Clinical Nutrition, December 1, 2007; 86(6): 1579 - 1580. [Full Text] [PDF]
|
|
|
|
|
|
H. A Bischoff-Ferrari, B. Dawson-Hughes, J. A Baron, P. Burckhardt, R. Li, D. Spiegelman, B. Specker, J. E Orav, J. B Wong, H. B Staehelin, et al. Calcium intake and hip fracture risk in men and women: a meta-analysis of prospective cohort studies and randomized controlled trials Am. J. Clinical Nutrition, December 1, 2007; 86(6): 1780 - 1790. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Smith, F. Anderson, H. Raphael, P. Maslin, S. Crozier, and C. Cooper Effect of annual intramuscular vitamin D on fracture risk in elderly men and women a population-based, randomized, double-blind, placebo-controlled trial Rheumatology, December 1, 2007; 46(12): 1852 - 1857. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Francis The Vitamin D Paradox Rheumatology, December 1, 2007; 46(12): 1749 - 1750. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Bouillon, A. W. Norman, P. Lips, R. C. Cava, A. N. D. Javier, W. R. Howe, R. Dellavalle, G. I. Baroncelli, and M. F. Holick Vitamin D Deficiency N. Engl. J. Med., November 8, 2007; 357(19): 1980 - 1982. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
