Calcium and vitamin D in the prevention and treatment of osteoporosis – a clinical update

Authors


Dr Steven Boonen MD, PhD, Leuven University Center for Metabolic Bone Diseases and Division of Geriatric Medicine, University Hospital Leuven, Herestraat 49, B-3000, Leuven, Belgium.
(fax: +32 16 34 44 02; e-mail: steven.boonen@uz.kuleuven.ac.be).

Abstract.

Combined calcium and vitamin D supplementation is an essential component of the management of osteoporosis, supported by a strong scientific rationale. The types of individuals who should receive calcium and vitamin D supplements are those: (i) patients with documented osteoporosis receiving antiresorptive or anabolic treatment; (ii) patients receiving glucocorticoids; and (iii) individuals with or at high risk of calcium and/or vitamin D insufficiencies, in particular older women and men. This article describes the evidence base that supports targeting these groups. Benefits are most apparent when 800 IU day−1 vitamin D is complemented with a dose of 1000–1200 mg day−1 elemental calcium. Compliance is also key to optimizing clinical efficacy.

Introduction

Osteoporosis is a systemic disease caused by various pathological mechanisms that create a situation where bone mass is lost and the skeletal microarchitecture is weakened; this leads to an increased risk of fractures [1]. The different variants of the disease are primary osteoporosis including postmenopausal, age-related (senile) and idiopathic osteoporosis, and secondary osteoporosis, which is due to a specific underlying disease or drug therapy. Irrespective of the aetiology, osteoporotic fractures most commonly occur at the hip, spine and wrist. The consequences of these fractures can be serious, sometimes life threatening; the economic costs of treating fracture are also considerable [2–4]. Vertebral fractures can cause back pain, decreased pulmonary function, loss of height, deformity, and immobility [5–9], with resultant impact on patients’ health-related quality of life [10]. Hip fractures are associated with disability, chronic pain, decreased mobility, and dependence on long-term nursing care [11–13]. Mortality has also been found to increase following any major fracture [14–16], especially in men [17, 18].

Fracture prevalence increases dramatically with age [17, 19, 20]. With the number of people aged over 65 years growing considerably because of rising life expectancy, the prevalence of osteoporotic fractures is likely to increase; lifetime fracture-risk estimates taking this increased longevity into consideration are 47% for women and 22% for men [21]. Recent estimates show that there are currently 7.8 million women and 2.3 million men with osteoporosis in the US, which could rise by 35–45% over the next 15 years [22].

Clearly, the prevention and treatment of this highly prevalent and serious condition need to become healthcare priorities in the coming years. Effective pharmacological therapies for the prevention and treatment of osteoporosis have been developed, including bisphosphonates, raloxifene, teriparatide, parathyroid hormone (PTH) and calcium and vitamin D. Of these, calcium and vitamin D are often regarded as a first-line management strategy, either alone or in combination with other therapies [23]. This paper explores the rationale for using calcium and vitamin D supplementation in osteoporosis, the clinical evidence supporting this efficacious and inexpensive therapy, and the main patient groups who should receive calcium and vitamin D supplements.

Rationale for calcium and vitamin D use in osteoporosis

Mechanisms determining fracture risk

Osteoporosis is traditionally defined as a bone mineral density (BMD) [normally measured by dual-energy X-ray absorptiometry (DEXA)] at least 2.5 standard deviations below the mean of healthy young females [24]. Peak bone mass is achieved in the third decade, after which bone mass declines due to a decrease in bone formation. This decline is generally steady in men, as well as in women until the menopause, after which time the loss of endogenous oestrogen accelerates bone resorption [25]. However, the proportion of fractures attributable to DEXA-defined osteoporosis is modest, with a recent study suggesting that only <10–44% of fractures occur in patients meeting these criteria [26]. Other factors influencing fracture risk that are not revealed by measurements of BMD include changes in bone microarchitecture and remodelling [27]. Bone remodelling, where bone is resorbed by osteoclasts and reformed by osteoblasts, occurs throughout life. It is dependent on mechanical load and strain, with new bone deposits focused to strengthen bone under the most stress [28]; increased remodelling can have a negative effect on the microstructure of bones without affecting the BMD.

Falling constitutes a significant fracture risk in the elderly, with over 90% of hip fractures caused in this way [29]. Many factors, including muscular weakness, contribute to the risk of falling in older individuals [30], leading to around one-third of people aged ≥65 years falling at least once a year [31]. Up to 50% of falls in older people cause a minor trauma [32], with 5–6% resulting in a fracture [33].

Role of calcium and vitamin D in bone and osteoporosis

Bone mass contains a high proportion of calcium. Peak bone mass is largely dependent on calcium intake during adolescence, and maintaining a high intake can slow age-related bone loss [34–37]. Conversely, inadequate dietary calcium intake increases the prevalence of osteoporosis [38]. In addition to its structural role, bone serves as the body's nutrient reserve of calcium [39].

Ingested calcium is absorbed from the intestine in two ways: passively and by vitamin D-mediated active transport, both of which become less efficient with age [40]. The efficiency of calcium absorption decreases further in the presence of vitamin D insufficiency. Vitamin D status is generally determined by measuring the serum concentration of the major circulating form of vitamin D, 25-hydroxyvitamin D [25(OH)D] [41]. Serum 25(OH)D levels tend to show a gradual decline with age, although absolute levels do vary across studies, according to geographical location, dietary vitamin D intake and the assay method used [42]. As people age, the skin's capability to synthesize vitamin D declines [43], whilst intestinal vitamin D absorption also becomes less efficient [44]. These changes may be compounded by lower exposure to the sun, as a result of diminished physical activity, and reduced dietary vitamin D intake; people who live at northern latitudes [45] or who are housebound or living in an institution are at particular risk of vitamin D insufficiency [42].

The presence of low calcium levels stimulates the secretion of PTH in order to increase production of 1,25(OH)2D (the physiologically active form of vitamin D), i.e. causes secondary hyperparathyroidism. PTH is a major hormone that maintains calcium homeostasis and normocalcaemia and regulates bone remodelling; it is itself regulated through levels of 1,25(OH)2D and calcium. Elevated PTH levels trigger bone metabolism, causing bone resorption and bone loss and releasing calcium ions in the blood [46–52]. When bone resorption occurs at an elevated rate, fracture risk is increased regardless of the individual's BMD [53], as the bone is eroded, weakening the skeleton [54–56].

The role of vitamin D in muscle strength.  In addition to its pivotal role in calcium homeostasis, vitamin D has also been linked with muscular strength. Highly specific receptors for vitamin D are expressed in human skeletal muscle [57] and myoblast cell lines [58], with expression decreasing significantly with age [59]. When vitamin D binds to these receptors, protein synthesis and cellular growth may be upregulated, a theory supported by a study in elderly women that showed increases in the relative size and number of muscle fibres after 3-month supplementation with 1-alphahydroxyvitamin D (alphacalcidol) [60].

Clinical studies have demonstrated a relationship between vitamin D and muscle strength in older people [61–66]. For example, better lower-extremity function (as measured by the sit-to-stand and 8-foot walk tests) was associated with higher serum 25(OH)D in 4100 ambulatory people aged ≥60 years [62]. Likewise, lower serum 25(OH)D and higher PTH levels were significant determinants of loss of muscle mass and strength in a group of 1008 people aged ≥65 years [64]. Finally, a recent study in 969 women found 25(OH)D levels correlated with physical activity, balance, gait speed and thigh muscle strength [63].

In summary, the age-related increase in fracture risk is influenced by changes in bone remodelling and BMD and an increased propensity to falls, attributable in part to the loss of muscular strength. Adequate dietary calcium intake is critical in maintaining bone mass, whilst vitamin D plays a key role in modulating calcium homeostasis and maintaining muscular strength. There is therefore a strong rationale for giving both calcium and vitamin D supplements to prevent and treat osteoporosis.

Target groups for calcium and vitamin D

Patients treated for osteoporosis using antiresorptive and anabolic agents

Pharmacological therapies for the prevention of fractures in patients with osteoporosis include antiresorptive therapies such as the bisphosphonates (alendronate, etidronate, ibandronate, risedronate) and the selective oestrogen receptor modulator raloxifene, and the anabolic agents teriparatide (PTH 1-34) and PTH. Antiresorptive therapies reduce osteoporotic fracture risk by increasing BMD and suppressing bone remodelling [67], initially increasing lumbar spine BMD by 4–5%, followed by a longer-lasting steady gain of 0.5–1% each year [68]. For this to occur, the amount of calcium absorbed must exceed the amount excreted, making adequate intake a necessity. Anabolic therapies stimulate bone formation, repairing defects at a microarchitectural level [69], leading to a substantial increase in BMD (9–13% per year at the lumbar spine, 3–6% per year at the femoral neck) in the presence of adequate calcium and vitamin D status [70]; such increases in bone mass cannot be achieved in patients with insufficient calcium intake.

All clinical trials of the antiresorptive and anabolic therapies were carried out in calcium and vitamin D replete individuals. Indeed, the consensus amongst osteoporosis experts is that patients receiving antiresorptive and anabolic treatment for osteoporosis must have adequate calcium intake and optimal vitamin D status to maximize benefits [71–74]. A high proportion of patients with osteoporosis are at significant risk of calcium and/or vitamin D insufficiencies unless they receive supplementation [75, 76]. For example, mean baseline calcium intake was considered inadequate (<1200 mg day−1) in 85% of participants in six major osteoporosis trials [75]. Likewise, more than half of a sample of 1536 women with postmenopausal osteoporosis had serum 25(OH)D levels below those associated with secondary hyperparathyroidism [76].

Whilst the impact of calcium and vitamin D insufficiency has not been studied for all the available osteoporosis drugs, the available evidence does suggest that without sufficient calcium and vitamin D, the effects of osteoporosis therapies will be blunted. The bisphosphonate etidronate was significantly more effective (as assessed by BMD at the lumbar spine and femoral neck) in patients with serum 25(OH)D ≥40 nmol L−1 than in those with levels <40 nmol L−1 [77]. Another study showed that the combined etidronate and vitamin D therapy was associated with a significantly higher mean increase in BMD at 1 year than etidronate alone at both the lumbar spine (5.2% vs. 2.7%) and femoral neck (2% vs. −0.4%) [78]. Further, in a meta-analysis of 31 trials, hormone replacement therapy (HRT), which has antiresorptive properties, was also associated with greater BMD increases when combined with calcium than when used alone (3.3% vs. 1.3% per year) [79].

In conclusion, therefore, in order that patients with documented osteoporosis derive the expected clinical benefits from antiresorptive or anabolic therapy, calcium and vitamin D supplementation should be given throughout treatment with these therapies.

Patients treated with glucocorticoids

Glucocorticoid therapy is generally the cause of osteoporosis in adults aged 20–45 years and is the most common cause of secondary osteoporosis at any age [80]. Glucocorticoid therapy is associated with rapid bone loss throughout the body, and a high fracture risk that is only partially explained by low BMD [81, 82]. Glucocorticoids have multiple effects on the body that increase fracture risk. Both intestinal and renal calcium absorption are suppressed, as is osteoblast function and bone formation, whilst urinary calcium excretion increases [83]. As calcium levels fall, secondary hyperparathyroidism develops, stimulating bone turnover and resorption [84]. In addition to the resultant effects on BMD, glucocorticoid osteoporosis is also thought to arise from microarchitectural deterioration, predisposing to bone fragility [83].

The increased fracture risk associated with glucocorticoids is dose related. For example, the risk of hip fracture is 0.99 for glucocorticoid doses below 2.5 mg day−1, 1.77 for doses of 2.5–7.5 mg day−1 and rising to 2.27 for doses >7.5 mg day−1 [85]. Whilst theoretically, therefore, prevention of bone loss can be achieved by reduction of glucocorticoid dose, this is not always possible clinically. According to current guidance from the American College of Rheumatology, all patients receiving glucocorticoids should also be treated with calcium, and either vitamin D (800 IU day−1) or an analogue (calcitriol or alfacalcidol), with antiresorptive therapy added for patients treated with higher glucocorticoid doses [86]. Current UK guidelines state that the BMD of all patients receiving glucocorticoids should be investigated, and adequate vitamin D and calcium intake should be maintained [87]. Recent guidelines from the Belgian Osteoporosis Society also recommend calcium and vitamin D supplements as first-line therapy for glucocorticoid-induced osteoporosis [88].

The benefits of combined calcium and vitamin D supplementation in the prevention of glucocorticoid-induced osteoporosis have been demonstrated clinically. For example, in a 2-year randomized trial, rheumatoid arthritis patients receiving prednisone therapy (mean dose 5.6 mg day−1) lost BMD at a rate of 2.0% and 0.9% per year in the lumbar spine and trochanter respectively. In contrast, patients randomized to calcium (1000 mg day−1) and vitamin D (500 IU day−1) gained BMD at an annual rate of 0.72% and 0.85% respectively (P = 0.005 and 0.024 respectively) [89]. There is a paucity of studies directly comparing the efficacy of the vitamin D analogues versus native vitamin D in glucocorticoid-induced osteoporosis, with the available studies often underpowered for the fracture endpoint [90, 91]. No difference between these approaches was identified with respect to either bone loss or fracture prevention in a recent indirect comparative meta-analysis [92]. The relative side effect profile of these therapies is an important consideration that may drive treatment choice.

In summary, patients prescribed glucocorticoid therapy are at high risk of developing osteoporosis. Taking into account the pathophysiological mechanisms of glucocorticoid-induced osteoporosis, the clinical evidence of benefit of calcium and vitamin D and expert opinion, supplementation should be initiated at the time glucocorticoid therapy is commenced.

Individuals with documented or likely calcium and vitamin D insufficiency

Insufficiencies of both calcium and vitamin D appear to be widespread in adults, particularly older and institutionalized adults [93]. A study of elderly ambulatory women living in institutions found 66% had daily intake of calcium <800 mg day−1 and low 25(OH)D levels (<30 nmol L−1) [94]. Elderly independent French women were also found to have a mean daily calcium intake of just 569 mg, and 39% had serum 25(OH)D ≤30 nmol L−1 [95]. Elderly and institutionalized individuals, therefore, represent a further important target group for calcium and vitamin D supplementation.

Effects of calcium and vitamin D on secondary hyperparathyroidism in the elderly.  Calcium and vitamin D supplements attenuate secondary hyperparathyroidism in the elderly. PTH levels were reduced by 31% in subjects receiving vitamin D or an active vitamin D metabolite (alphacalcidol or calcitriol) compared with controls [50, 96–100]. This decreased serum PTH is associated with reduced bone turnover [100] and increased BMD [101]. When calcium and vitamin D are administered together, a greater reduction in serum PTH is observed than with either component alone [50], particularly in patients who had severe vitamin D deficiency at baseline or a low calcium intake. Increases in BMD of the femoral neck, lumbar spine and/or hip of 4–6% have also been observed in patients receiving a combination of calcium and vitamin D, accompanied by a decrease in bone resorption markers and serum PTH levels [102–104].

Effects of calcium and vitamin D on muscular strength and the risk of falling.  Compared with calcium alone, calcium and vitamin D supplementation has been associated with significant improvements in musculoskeletal function, such as knee flexor and extensor strength and grip strength, in elderly women after just 3 months [105], with the risk of falling decreasing by 49%. The effects on falls have been described in several other randomized, controlled trials conducted in at-risk individuals [105–110]. Vitamin D supplementation (10 000 IU once weekly plus 600 mg elemental calcium once daily) was shown to reduce risk of falling by 27% at 2 years in elderly people who had pretreatment serum 25(OH)D levels of 25–90 nmol L−1 [110]. Combined treatment has also been shown to reduce body sway, a major risk factor for falling, by 9% [108]. Other studies, however, have failed to demonstrate a significant reduction in falls with vitamin D (for example, [109]).

In order to clarify the effects of vitamin D in respect of fall prevention, a meta-analysis of double-blind, randomized, controlled trials of vitamin D (any form) on falls was conducted [111]. In an analysis of five trials involving subjects in stable health states that stated how falls were ascertained and defined, the risk of falling was reduced by 31% in subjects receiving vitamin D compared with those receiving calcium alone or placebo; the corrected risk reduction was 22%. In a sensitivity analysis that included a further five trials that did not meet the inclusion criteria for the primary analysis, the effect size decreased to 13%, but remained statistically significant.

Effects of calcium and vitamin D on fracture risk in the elderly.  The effects of calcium and vitamin D on fracture risk in postmenopausal women and elderly men who do not have documented osteoporosis and are not at risk of secondary osteoporosis have been studied extensively. Different populations, including community-dwelling subjects and ambulatory people living in institutions, have been studied, as have varying doses of calcium and vitamin D (Table 1) [94, 103, 109, 112–116].

Table 1.   Overview of randomized controlled clinical trials of long-term calcium and vitamin D in the prevention of fractures
SourceStudy populationTrial design/dosingDuration (months)Incidence of fractures (P-value)
Any nonvertebralHip
  1. aAfter this initial analysis, patients were further followed to 36 months. Results are cited in text. bPatients had ≥1 prior low-trauma fractures. Results cited are new low-trauma fractures. cPatients had at least one self-reported risk factor for osteoporosis. Patients in both groups received a leaflet providing information on calcium intake and fall prevention. dControl patients were offered an environmental and health programme aimed at preventing fractures.

Studies in institutionalized patients
[103]3270 F, mean age 84 yearsCalcium 1200 mg day−1 + vitamin D 800 IU day−1 vs. placebo18a66 vs. 97 (0.015)21 vs. 37 (0.043)
[94]639 F, mean age 85.2 yearsCalcium 1200 mg day−1 + vitamin D 800 IU day−1 vs. placebo2417.8% vs. 17.9% (NS)6.9% vs. 11.1% (0.07)
Studies in community-dwelling patients
[109]3314 F, mean age 77 yearscCalcium 1000 mg day−1 + vitamin D 800 IU day−1 vs. no treatmentMedian 254.4% vs. 4.6%0.6% vs. 0.9%
[116]2638, 85% F, aged ≥70 yearsbCalcium 1000 mg day−1 + vitamin D 800 IU day−1 vs. placebo24–6212.6% vs. 12.7%3.5% vs. 3.1%
[115]5073 M and F, mean age 74 yearsCalcium 1000 mg day−1 + vitamin D 400 IU day−1 vs. placebod426.4% vs. 7.9%Not reported
[114]389 M and F, mean age 71 yearsCalcium 500 mg day−1 + vitamin D 700 IU day−1 vs. placebo365.6% vs. 12.9% (0.02)0 vs. 1
[112]36 282 F, mean age 62.4 yearsCalcium 100 mg day−1 + vitamin D 400 IU day−1 vs. placeboMedian 84Not reported1.0% vs. 1.1%

Four large-scale, randomized, placebo-controlled trials, two studying subjects within institutions and two involving elderly community residents, found positive effects of calcium and vitamin D supplementation (Table 1) [94, 103, 113–115]. In Decalyos I (Table 1), a study in elderly institutionalized women with severe calcium and vitamin D deficiencies, the risk of hip and all nonvertebral fractures had fallen by 43% and 32% respectively (P = 0.043 and 0.015 respectively) after 18 months of calcium and vitamin D therapy [103]. Prior to treatment, these women had a mean dietary calcium intake of 513 mg day−1 (<800 mg day−1 in all cases) and serum 25(OH)D levels <50 nmol L−1. In subjects still receiving treatment after 36 months, hip fracture and nonvertebral fracture risk continued to be reduced, by 29% and 24% respectively (Table 1) [113]. A similar, but smaller, group of individuals was enrolled in Decalyos II (n = 583), 79% of whom had a calcium intake <800 mg day−1 and a serum 25(OH)D <50 nmol L−1 [94]. A similar reduction in hip fracture risk to that found in Decalyos I was observed (Table 1), although it must be noted that the study was not adequately powered for the hip fracture endpoint.

A reduction in fracture risk with supplementation was also observed in a study of both men and women (n = 389) aged ≥65 years living in the community and who had less severe insufficiencies of calcium and vitamin D than in the Decalyos studies (Table 1) [114]; mean calcium intake was approximately 730 mg day−1, and baseline serum 25(OH)D levels 83.25 and 71.25 nmol L−1 for men and women respectively (Table 1) [114]. A significant difference (P = 0.02) in nonvertebral fractures was observed between the placebo group (13.3%) and those subjects receiving 500 mg day−1 calcium and 700 IU day−1 vitamin D (5.6%), although this difference was driven by the reduction in ankle and radius fracture rates [114]. A nonrandomized trial in elderly community residents (n = 9605) found a reduction in fracture risk of 16% with calcium (1000 mg) and vitamin D (400 IU) compared with controls offered an environmental and health programme aimed at preventing fractures [115].

In contrast to these trials, three recent randomized trials of calcium and vitamin D supplementation in elderly community-dwelling people did not find a reduction in fracture risk [109, 112, 116]. In the RECORD trial, which involved a total of 5292 elderly men and women who had previously suffered a low-trauma fracture, new low trauma fractures occurred in 13% of all subjects, but with no significant difference between the calcium (1000 mg day−1) and vitamin D (800 IU day−1) group and the placebo group (12.6% and 13.4% respectively) (Table 1) [116]. In the second trial, women aged ≥70 (n = 3314) with at least one risk factor for hip fracture were randomized to receive daily calcium (1000 mg) and vitamin D (800 IU) supplements or a leaflet providing information on calcium intake and fall prevention (Table 1) [109]. Again, low clinical fracture rates were found in both groups after a median follow-up period of 25 months (range 18–42), and there was no significant difference between groups for overall or hip fracture risk. Participants in the third trial were enrolled in either the Women's Health Initiative (WHI) Dietary Modification trial or the WHI Hormone Therapy trials, and were invited to join the calcium with vitamin D trial [112]. The 36 282 postmenopausal women recruited were randomized to receive daily doses of either calcium (1000 mg) with vitamin D (400 IU) or placebo. After a median follow-up of 7 years, no significant difference was found between the treatment and placebo arms in terms of fracture rate for both hip fractures (1% vs. 1.1% respectively) or total fractures (11.6% vs. 11.9% respectively) (Table 1) [112].

In addition to the anti-fracture studies of calcium and vitamin D described in Table 1, several studies have evaluated vitamin D alone [117–120], again with conflicting results. For example, whilst a large trial of vitamin D 100 000 IU every 4 months reduced the risk of first hip, wrist or forearm, or vertebral fractures by 33% [119], daily vitamin D 400 IU did not reduce fracture risk versus placebo in a study of similar size involving 2578 men and women aged ≥70 years [117].

Factors determining the clinical efficacy of calcium and vitamin D

While several large-scale, randomized, placebo-controlled trials, studying individuals within institutions and/or involving elderly community residents, found positive effects of calcium and vitamin D supplementation on the risk of hip and all nonvertebral fractures [94, 103, 113–115], some trials in community-dwelling people (including the RECORD trial performed in the UK and, most recently, the WHI clinical trial in the US) did not find a significant reduction in fracture risk during calcium and vitamin D supplementation [109, 112, 116]. Similarly, studies evaluating vitamin D alone have yielded conflicting results [116, 117, 119].

The inconsistent results of the anti-fracture trials of calcium and vitamin D in individuals without documented osteoporosis suggest that there are additional considerations in ensuring the clinical efficacy of calcium and vitamin D in this setting. These include the dose of vitamin D, the addition of calcium to vitamin D, the targeting of the supplementation to those with insufficiencies, and the need for good compliance.

Dosing of vitamin D.  Recent meta-analyses of vitamin D supplementation have provided evidence that a dose of 800 IU day−1 vitamin D is required for optimal benefit [111, 121]. To reduce fracture risk, 700–800 IU day−1 of vitamin D is needed, whereas 400 IU day−1 vitamin D was ineffective (Fig. 1) [121]. Likewise, a meta-analysis of the effects of vitamin D on risk of falling demonstrated that 800 IU day−1 of vitamin D (plus 0–1200 mg day−1 calcium) was associated with a fall risk reduction of 29–35%, whilst 400 IU day−1 did not seem to confer benefit [111]. Inadequate vitamin D dosing is likely to have contributed to the failure to show antifracture efficacy during treatment with (400 IU day−1) vitamin D and calcium in the recent WHI clinical trial [112]. As the WHI authors recognized in their recent publication, the ‘effect of calcium with vitamin D supplementation might require higher doses of vitamin D than were used in the WHI’ [112].

Figure 1.

 Forest plots showing differences in hip and nonvertebral fracture risk between vitamin D doses (700–800 and 400  IU day−1) and the control groups [121]. Reproduced with permission.

Combining vitamin D and calcium.  The addition of calcium is one major difference between the Decalyos I study [103, 113], which reported benefits for combined supplementation, and that of Lips et al. [117], which reported no benefit with vitamin D alone at a dose of 400 IU day−1, although other factors may also have determined the different outcomes (see Fig. 2a,b). A Cochrane review in 2001 was unable to reach a firm conclusion regarding the efficacy of vitamin D alone, whilst concluding that some older people taking calcium and vitamin D supplements do benefit in terms of a reduction in fracture risk [122].

Figure 2.

 Hip fracture data from (a) the Decalyos I [103] and (b) the Lips et al. study [117]. Reproduced with permission. In the Decalyos I study, patients received calcium 1200 mg day−1 and vitamin D 800 IU day−1. In contrast, in the Lips et al. trial, patients received vitamin D 400 IU day−1 alone. In addition, it should be noted that the Lips et al. study had insufficient power to document fracture risk reductions <20%, an important limitation given that the hip fracture incidence was 25–30/1000 patient-years compared with 40/1000 patient-years in Decalyos I.

The fracture risk meta-analysis described above [121] was unable to establish the necessity for additional calcium, as calcium was given in all but one of the studies of the 800 IU day−1 dose. According to a recent adjusted indirect comparison of randomized controlled trials of vitamin D versus placebo and vitamin D plus calcium versus placebo, the addition of calcium is necessary to optimize the clinical efficacy of vitamin D supplementation in terms of reducing fracture risk (S. Boonen and P. Lips, unpublished observations). Thus, elderly patients should receive supplementation with both vitamin D 800 IU day−1 and an adequate dose of calcium.

Targeting calcium and vitamin D to those with insufficiencies.  Studies like the RECORD trial and the WHI trial have made it clear that universal supplementation in the community is not necessary. To maximize the health benefits of this strategy, calcium and vitamin D should, therefore, be targeted to those elderly individuals with known, or at most risk of, calcium and/or vitamin D insufficiencies [93, 123]. Whilst it is regrettable that, in the RECORD and WHI trials, vitamin D status across the entire trial population is unknown (levels generally being assessed in only a small subset, e.g. 1.1% of subjects in RECORD and 1.3% of subjects in WHI), the available baseline data do support the suggestion that the subjects living in the community in these studies were less severely vitamin D deficient than those living in institutions. The WHI clinical trial was targeted to a subset of healthy postmenopausal women and 82% of the study population was younger than 70 years of age [112]. Participants in the trial were living in the community and generally free of disability. Their average calcium intake at baseline exceeded 1000 mg day−1, close to current recommendations, and 42% had a daily vitamin D intake above 400 IU. In addition, more than half of the participants were receiving HRT on study entry, many as part of the active hormone-replacement programme of the WHI. The trial design assumed an 18% reduction in the risk of hip fracture and projected a hip fracture rate (approximately 34 per 10 000 persons per year) that was more than twice that observed (16 per 10 000). This lower-than-projected hip fracture rate reduced the power of the study to approximately 48% and is consistent with the assumption that a significant proportion of these women were not calcium or vitamin D insufficient.

The issue is, therefore, what level of dietary calcium intake or 25(OH)D levels define insufficiency? The US recommended daily allowance of calcium is 1000 mg day−1 for adults, rising to 1200 mg day−1 for those over 50 years of age [123], whilst adequate vitamin D intake for adults aged between 51 and 70 years is 400 IU day−1, and 600 IU day−1 for those over 70 [124]. European recommendations are lower, with calcium at 700–800 mg day−1, although it is 800 mg day−1 for women aged 50–65 years [125]. The European Commission recommended that adult intake of vitamin D should be between 0 and 400 IU day−1, depending on skin synthesis, rising to 400 IU day−1 for those over 65 years [125].

The relative importance of calcium intake and serum 25(OH)D status with respect to calcium homeostasis (in terms of serum PTH levels) has been assessed recently in a sample of 944 Icelandic adults [126]. The results suggested that it is more important to ensure vitamin D sufficiency than a high calcium intake. However, the findings should be interpreted with caution because of the cross-sectional design of the study with very few individuals over the age of 70 years and because the data had been obtained on a healthy subsample of the original random sample of the population. But the data do suggest that, as expected, vitamin D seems to have a calcium-sparing effect. There is currently a lack of consensus on the serum 25(OH)D threshold for defining optimal vitamin D status [42, 50, 93, 98, 127]. Different thresholds for vitamin D insufficiency have been suggested (30, 50 and 75 nmol L−1), based on a variety of endpoints, including PTH suppression and the prevention of bone loss, falls and fractures [42, 98, 127, 128]. Different 25(OH)D assays [42], variations in calcium intake between studies, and the fact that serum PTH levels do not necessarily rise above the upper limit of normal in the presence of a low 25(OH)D level [76] can help to explain these discrepancies.

A substantial body of evidence supports a 25(OH)D threshold of 50 nmol L−1 as indicative of optimal vitamin D status, below which patients require supplementation. For example, when plots of the inverse relationship between PTH and 25(OH)D are examined, increases in PTH are minimal at 25(OH)D levels above 50 nmol L−1, both in healthy adults [93] and in postmenopausal women with osteoporosis [76]. Furthermore, vitamin D supplementation appears to cause much larger decreases in serum PTH of patients who had baseline 25(OH)D levels <50 nmol L−1 than in those whose baseline 25(OH)D was >50 nmol L−1 [50, 98]. A large study of the relationship between lower extremity function and 25(OH)D levels in older ambulatory adults also concurs with a threshold of 50 nmol L−1 [62]; virtually no change in function (measured by the 8-foot walk test and the sit-to-stand time) was observed when serum 25(OH)D was 50–75 nmol L−1, whereas function declined sharply when serum 25(OH)D fell below 50 nmol L−1. With respect to falls, no well-designed studies are available that establish a beneficial effect of vitamin D on falls in patients with baseline 25(OH)D levels >50 nmol L−1. Finally, in a Scottish study, 92.7% of elderly patients with hip fractures were found to have serum 25(OH)D levels <50 nmol L−1 [129].

Compliance.  The effects of pharmacological therapy for chronic conditions can be diluted if patient compliance and persistence with therapy is not optimal [130, 131]. Compliance and persistence with calcium and vitamin D supplementation are critical in osteoporosis as the effects do not last beyond the end of treatment [104, 132]. The differing outcomes of studies assessing the effects of calcium and vitamin D supplementation on fracture risk emphasize the importance of patient compliance. Compliance with calcium and vitamin D therapy in recent negative community-based fracture studies was reported to be only 40–60% [109, 112, 116]. In fact, the WHI investigators acknowledged that ‘it is plausible that there was a benefit amongst the women who adhered to the study treatment’ and showed that, in sensitivity analyses, there was a decrease of hip fracture amongst adherent patients (those who took >80% of their study medication) [112]. A nonsignificant reduction in hip fracture risk of 12% (hazard ration 0.88) was observed in women who received calcium and vitamin D compared with placebo, but when only adherent women were included, this rose to a significant 29% hip fracture risk reduction (hazard ratio 0.71) [112]. Individuals who live in institutions, where medication is administered by staff, achieve much higher rates of compliance, and positive benefits of calcium and vitamin D supplementation have been observed in all studies involving such patients [94, 103, 113]. Selective noncompliance could play a role in community-dwelling elderly, i.e. the frail who are vitamin D deficient might often forget to take the tablets. The role of the physician is important to maximize medication compliance in patients living in the community [133] which is why calcium and vitamin D supplementation should be prescribed as a medicinal product by the physician.

Conclusions

A wealth of evidence has shown that calcium and vitamin D supplementation has a beneficial effect on the bone health and fracture risks of certain groups of individuals. Calcium and vitamin D supplementation is most effective when targeted to those who: (i) are receiving antiresorptive or anabolic osteoporosis therapy; (ii) are being treated with glucocorticoids; and (iii) are likely to be calcium or vitamin D insufficient. These benefits are most apparent when 800 IU day−1 vitamin D is complemented with a dose of 1000–1200 mg day−1 elemental calcium. Adequate compliance and persistence with calcium and vitamin D therapy is crucial for patients to benefit from supplementation, with the physician playing a fundamental role, by both prescribing supplements and ensuring regular patient follow-up and education.

Conflict of interest statement

The authors have no conflict of interest.

Acknowledgements

Drs S. Boonen and D. Vanderschueren are both Senior Clinical Investigators for the Fund for Scientific Research–Flanders, Belgium (F.W.O.-Vlaanderen). This work was supported by a grant G.0171.03 from the Fund for Scientific Research–Flanders, Belgium (F.W.O.-Vlaanderen) to S. Boonen. The authors gratefully acknowledge the editorial assistance of Eleanor Steele in the preparation of this manuscript.

Ancillary