This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lips, P. Search for Related Content PubMed PubMed Citation Articles by Lips, P.

Vitamin D Deficiency and Secondary Hyperparathyroidism in the Elderly: Consequences for Bone Loss and Fractures and Therapeutic Implications

Department of Endocrinology, Institute for Endocrinology, Reproduction and Metabolism (EVM-Institute) and Institute for Research in Extramural Medicine (EMGO-Institute), Vrije Universiteit Medical Center, 1007 MB Amsterdam, The Netherlands

Correspondence: Address all correspondence and requests for reprints to: Paul Lips, M.D., Ph.D., Afdeling Endocrinologie, Vrije Universiteit Medical Center, Postbus 7057, 1007 MB Amsterdam, The Netherlands. E-mail: p.lips{at}vumc.nl

	Abstract

Top Abstract I. Introduction II. Vitamin D Status... III. Consequences of Vitamin... IV. Other Causes of... V. Functional Classification and... VI. Prevention and Treatment VII. Public Health Aspects VIII. Future Prospects IX. Conclusion References

 
Vitamin D deficiency is common in the elderly, especially in the housebound and in geriatric patients. The establishment of strict diagnostic criteria is hampered by differences in assay methods for 25-hydroxyvitamin D. The synthesis of vitamin D₃ in the skin under influence of UV light decreases with aging due to insufficient sunlight exposure, and a decreased functional capacity of the skin. The diet contains a minor part of the vitamin D requirement. Vitamin D deficiency in the elderly is less common in the United States than elsewhere due to the fortification of milk and use of supplements. Deficiency in vitamin D causes secondary hyperparathyroidism, high bone turnover, bone loss, mineralization defects, and hip and other fractures. Less certain consequences include myopathy and falls. A diet low in calcium may cause an increased turnover of vitamin D metabolites and thereby aggravate vitamin D deficiency. Prevention is feasible by UV light exposure, food fortification, and supplements. Vitamin D₃ supplementation causes a decrease of the serum PTH concentration, a decrease of bone turnover, and an increase of bone mineral density. Vitamin D₃ and calcium may decrease the incidence of hip and other peripheral fractures in nursing home residents. Vitamin D₃ is recommended in housebound elderly, and it may be cost-effective in hip fracture prevention in selected risk groups.

I. Introduction

A. History; association of hip fractures with osteomalacia

B. Physiology of vitamin D and bone mineralization

II. Vitamin D Status in the Elderly

A. Assessment of vitamin D status

B. Determinants of vitamin D status

C. Prevalence of vitamin D deficiency

D. Changes in vitamin D metabolism with aging

III. Consequences of Vitamin D Deficiency

A. Secondary hyperparathyroidism and high bone turnover

B. Bone histology in patients with hip fractures

C. Vitamin D status and bone mineral density

D. Vitamin D deficiency as a risk factor for fractures in epidemiological studies

E. Vitamin D deficiency and myopathy

F. Other consequences of vitamin D deficiency

G. Racial differences in vitamin D and PTH metabolism

IV. Other Causes of Secondary Hyperparathyroidism in the Elderly

A. Renal function

B. Estrogen deficiency

C. Low calcium nutrition

V. Functional Classification and Diagnosis of Vitamin D-Deficient States

A. Vitamin D deficiency and insufficiency

B. Diagnostic criteria

VI. Prevention and Treatment

A. Sunshine and UV irradiation

B. Oral vitamin D supplementation

C. Side effects and risks

VII. Public Health Aspects

A. Recommended dietary allowance and adequate intake

B. Risk groups

C. Prevention strategies

D. Cost-effectiveness

VIII. Future Prospects

IX. Conclusion

	I. Introduction

Top Abstract I. Introduction II. Vitamin D Status... III. Consequences of Vitamin... IV. Other Causes of... V. Functional Classification and... VI. Prevention and Treatment VII. Public Health Aspects VIII. Future Prospects IX. Conclusion References

 
A. History; association of hip fractures with osteomalacia
VITAMIN D is widespread in nature and photosynthesized in most plants and animals exposed to sunlight (1, 2). Its major role in vertebrate animals and humans is to increase the absorption of calcium and phosphate for the mineralization of the skeleton. In the case of vitamin D deficiency in children, the cartilage is not calcified, causing rickets. In adults, the newly formed bone matrix (the osteoid) is not mineralized, causing osteomalacia (3). The first classical description of rickets traces back to Whistler and Glisson in the middle of the 17th century (1). The association between lack of sunshine and rickets was first recognized in the beginning of the 19th century (1, 4). A hundred years later, around 1920, rickets was experimentally cured by exposure of children to sunshine or UV light (1, 2, 5). Since then, rickets has almost disappeared in the western world because of the use of cod liver oil or vitamin D preparations and by adequate sunshine exposure. About 50 yr later, it was recognized in 1967 by Chalmers et al. (6) that osteomalacia was more common than expected, especially in elderly women. They also reported that hip fractures often were often associated with osteomalacia (7). While rickets and osteomalacia are rare outside certain risk groups, less severe vitamin D deficiency is quite common, especially in elderly people (8, 9). This less severe form, often described as vitamin D insufficiency or inadequacy (10, 11), causes stimulation of the parathyroid glands, which may lead to high bone turnover, bone loss, and hip fractures (8, 12). Over the last 20 yr, numerous papers have been published on vitamin D deficiency and its consequences in elderly people. The aim of this paper is to review the prevalence of vitamin D deficiency in the elderly, its consequences, and its prevention and treatment, including public health aspects.

B. Physiology of vitamin D and bone mineralization
Vitamin D₃, or cholecalciferol, is synthesized in the skin. Its precursor, 7-dehydrocholesterol, is converted by the UV light of the sun (UVB 290–315 nm) into previtamin D₃, which is slowly isomerized to vitamin D₃ (13). Vitamin D binding protein (DBP) binds vitamin D and its metabolites and transports them in the bloodstream (14). Some nutrients also contain vitamin D₃, e.g., fatty fish, eggs, and dairy products. Vitamin D₂, or ergocalciferol, originates from irradiation of ergosterol, a major plant sterol (13), and has been added to dairy products and (multi)vitamin preparations. More and more, however, it is being replaced by vitamin D₃. Vitamin D₂ is also transported in the circulation by DBP, and its metabolism is similar to that of vitamin D₃.

Vitamin D is hydroxylated in the liver into 25-hydroxyvitamin D [25(OH)D], which is the major circulating metabolite (15, 16). Further hydroxylation into 1,25-dihydroxyvitamin D [1,25-(OH)₂D] occurs primarily in the kidney. The hydroxylation in the kidney is stimulated by PTH and suppressed by phosphate. While 25(OH)D has limited biological activity, 1,25-(OH)₂D is the most active metabolite stimulating the absorption of calcium and phosphate from the gut. The production of 1,25-(OH)₂D is under tight feedback control, directly by serum calcium and phosphate and indirectly by calcium via a decrease of serum PTH (15, 16, 17). When serum calcium and phosphate are sufficiently high, the production of 1,25-(OH)₂D diminishes in favor of another metabolite, 24,25-dihydroxyvitamin D. The function of this metabolite in humans is still unclear (15). The free serum 1,25-(OH)₂D concentration is very low, as 1,25-(OH)₂D is more than 99% bound to DBP and albumin (14). The active metabolite 1,25-(OH)₂D acts through the vitamin D receptor (VDR), a specific nuclear receptor, related to the T₄ and steroid hormone receptors (15, 17, 18). The VDR is present in the intestine where 1,25-(OH)₂D, after binding to the VDR, stimulates the synthesis of several proteins in the intestinal cells, which participate in the transport of calcium from the intestinal lumen into the bloodstream (16, 17). The VDR is also present in many other organs such as bone, muscle, pancreas, and pituitary (19). The active metabolite 1,25-(OH)₂D influences muscle function and stimulates cell differentiation and immunological function in general (19, 20). Rapid nongenomic effects of 1,25-(OH)₂D (not involving the VDR) have been observed in the intestine, the osteoblast, the parathyroid gland, and other tissues (21). The action of 1,25-(OH)₂D on bone is not well understood. It stimulates the osteoblasts to produce osteocalcin and alkaline phosphatase and decreases the production of type I collagen by fetal rat calvaria (17). On the other hand, 1,25-(OH)₂D stimulates bone resorption in vitro (15, 16). The effects of 1,25-(OH)₂D on bone mineralization appear to be indirect by stimulating the calcium and phosphate supply, mainly by absorption from the gut.

The bone remodeling sequence by which new osteons are formed starts with osteoclasts resorbing existing bone (22). Thereafter, osteoblasts appear and construct the new unmineralized bone matrix, the osteoid. Subsequently, the osteoid is mineralized. The mineralization of the osteoid occurs in two phases. During primary mineralization, about half of the bone mineral accumulates within a few days, increasing the density to 1.4 g/cm³. The secondary mineralization proceeds more slowly during 6 months or more and increases the density to 1.9 g/cm³. When mineralization is normal, the mineral content of an osteon depends on its age. Young, low-density bone is more prevalent when bone turnover is high. Older, completely mineralized high-density bone is associated with low bone turnover (23).

	II. Vitamin D Status in the Elderly

Top Abstract I. Introduction II. Vitamin D Status... III. Consequences of Vitamin... IV. Other Causes of... V. Functional Classification and... VI. Prevention and Treatment VII. Public Health Aspects VIII. Future Prospects IX. Conclusion References

 
A. Assessment of vitamin D status
Before serum measurement of vitamin D metabolites became feasible, vitamin D deficiency was suspected in patients with symptoms of bone pain and muscle weakness and was diagnosed by low serum calcium and phosphate levels and elevated alkaline phosphatase activity (3, 24). In addition, urine calcium excretion in these patients was low. Clinical signs used in the screening of elderly people were those pointing to proximal muscle weakness, such as standing up from a chair. During the last two decades, measurement of serum 25(OH)D has become common practice for the assessment of vitamin D status and the detection of vitamin D deficiency (25). Although this practice has made the diagnosis comparatively easy, the assays for 25(OH)D still lack sufficient standardization as indicated by international comparative studies (26, 27, 28). In the most recent study (28), serum 25(OH)D was measured in two laboratories in 104 samples from a vitamin D supplementation study by a competitive protein binding assay (CPB), RIA, and a CPB after purification by HPLC. The mean serum 25(OH)D measured by CPB was 80% higher than that measured by HPLC, while the RIA gave intermediate values. The correlation between CPB and HPLC was moderate (r = 0.69, P < 0.01). Of the serum 25(OH)D values measured by HPLC in the lowest quartile, 25% and 21% were not recognized to be in the lowest quartile by CPB and RIA, respectively. Similar discrepancies were observed in the highest quartile. In the second part of the same study, serum 25(OH)D was measured in up to eight serum samples by five laboratories using their routine assay methods (CPB or RIA). The difference between the mean values of the highest and the lowest laboratory was 38%. The individual values are shown in Fig. 1. It can be concluded from these interlaboratory comparisons that serum 25(OH)D levels from different regions or countries cannot be compared satisfactorily as long as the assays have not been cross-calibrated. Another problem is that different investigators have used various reference populations. Reference values have been based on healthy adults, such as blood donors, or on more or less representative samples of the population.

View larger version (20K): [in this window] [in a new window]   Figure 1. International comparison of serum 25(OH)D measurements in three European and two US laboratories. The points represent individual values of serum 25(OH)D measured by four laboratories in eight serum samples and by two laboratories in seven other serum samples. [Reproduced with permission from P. Lips et al.: Osteoporos Int 9:394–397, 1999 (28 ). © International Osteoporosis Foundation and National Osteoporosis Foundation.]

B. Determinants of vitamin D status
Vitamin D₃ (cholecalciferol) is synthesized in the skin under the influence of UV light (13). The UV range stimulating the formation of vitamin D₃ is rather small (290–315 nm). During winter at northern latitudes, the sunlight must pass a much longer distance through the atmosphere, and most UV light is absorbed. In the northern part of the United States and Canada, as well as in northwestern Europe, vitamin D production is virtually absent between October and March (29), but this may be different in more southern countries. UV radiation is effectively absorbed by glass and most plastics (1). Clothing prevents the photosynthesis of vitamin D₃ (30).

Sunscreens such as p-aminobenzoic acid also interfere with the cutaneous production of vitamin D₃ (31). The mean serum concentration of 25(OH)D was much lower in chronic sunscreen users than in control subjects (40 vs. 91 nmol/liter, Ref. 32). The cutaneous synthesis of vitamin D₃ is much lower in highly than in lightly pigmented skin (33). It is positively correlated with skin thickness (34).

The formation of vitamin D₃ in the skin is much less efficient in the elderly than in younger people. Total body irradiation with artificial UV light in six white young adults and six white elderly people with the same skin type showed that the increase of the serum vitamin D₃ concentration in young adults was about 4 times more than that in the elderly (35) (Fig. 2). However, even in the elderly, cutaneous vitamin D₃ production remains very effective. UV irradiation of 1,000 cm² of the skin on the back increases serum 25(OH)D from 20 to 60 nmol/liter in 3 months time, with an exposure time of 3–7 min three times per wk (36). This suggests that a 10- min exposure of head and arms (unprotected) three times per week, is adequate to prevent vitamin D deficiency.

View larger version (14K): [in this window] [in a new window]   Figure 2. Serum vitamin D3 concentration after total body exposure to artificial sunlight (UV 260–360 nm) in six white young adults (20–30 yr) and six white elderly people (62–80 yr) with skin type III. Serum vitamin D3 concentration was measured for 7 d. The area under the curve for serum vitamin D3 suggests that the production of vitamin D3 in the skin in the elderly is about 25% of that in young adults. [Reproduced with permission from M. F. Holick et al.: Lancet 2:1104–1105, 1989 (35 ). © The Lancet Ltd.]

C. Prevalence of vitamin D deficiency
Vitamin D deficiency can be defined according to population-based reference limits for serum 25(OH)D or biological indices, e.g., hypocalcemia and elevated alkaline phosphatase or PTH levels (health-based limits). The former will depend on the reference population and country, determined by sunshine exposure and nutrition. We have defined reference limits in a population of healthy blood donors in Amsterdam with nonparametric estimation of 95% confidence limits (44, 37). The lower reference limits for serum 25(OH)D were 20 nmol/liter in winter (October-March) and 30 nmol/liter in summer (April-September).¹ Population-based reference limits will be higher when sunshine exposure is higher or when the diet contains more vitamin D. A similar lower reference limit of 30 nmol/liter (12 ng/ml) for serum 25(OH)D was reported in the Euronut SENECA (Survey in Europe on Nutrition and the Elderly, a Concerted Action) study (45) and SUVIMAX study (10). A somewhat higher reference limit of 37 nmol/liter was reported in the United States (40).

Extensive reviews on vitamin D status in young adults and the elderly have been published (9, 46). An overview of studies on vitamin D status in the elderly is presented in Table 1. The mean values of these studies are graphically presented according to geographical region or to subject/patient category in Fig. 3. These studies were performed in postmenopausal women, independent healthy elderly subjects, outpatients of general or geriatric clinics, hospital inpatients, geriatric patients, residents of homes for the elderly or nursing homes, and patients with hip fracture. The selection of study subjects is not always clear. Some studies were done in a population sample, e.g., Framingham (81), Baltimore Aging Study (78), EPIDOS (62), Euronut (45). The serum 25(OH)D concentrations vary widely between different studies. Only part of this variation may be explained by assay differences. Serum 25(OH)D concentrations are lower in European countries than in the United States. The levels are lower in hospital patients and residents of nursing homes than in independent elderly subjects. Low serum levels were also observed in patients with severe Parkinson’s and Alzheimer’s disease in Japan (73, 74) and in patients with hip fracture in various countries. Serum 25(OH)D measured with one assay (HPLC followed by CPB assay) (84) in several groups in Amsterdam showed a gradual decline from healthy adults, to independent elderly, institutionalized elderly, and patients with hip fractures (Fig. 4) (37, 56, 85). While the picture in Fig. 3B may not be entirely reliable because of assay differences, Fig. 4, with all data from one assay method, gives a similar impression, which is reassuring. The lowest serum 25(OH)D concentrations were observed in geriatric patients in the United Kingdom and Ireland (48, 49, 53), but vitamin D deficiency also appears to be very common in institutionalized elderly patients in Switzerland (67). However, low levels were also found in Southern Europe. The European Euronut SENECA study was done in 15 centers in 11 countries using a central laboratory facility. Mean serum 25(OH)D varied from 22 nmol/liter in a study center in Greece to 46 nmol/liter in a center in Norway (45). Unexpectedly, serum 25(OH)D correlated positively with the degree of northern latitude (Fig. 5). The prevalence of vitamin D deficiency depends on the study population and the lower reference limit. In some studies this limit was set at 10 or 12.5 nmol/liter, and the investigators stated that this was the reference limit for osteomalacia (55, 58). In one of these studies, four subjects had hypocalcemia and osteomalacia (49). Vitamin D intake was around 100 IU/d or less in European studies, but twice as high or more in the United States Other remarkable observations are the high prevalence of vitamin D deficiency (serum 25(OH)D < 30 nmol/liter) in healthy adults (14%) and independent elderly (39%) in France (10, 62). Very low levels of serum 25(OH)D were observed in inhabitants of Saudi Arabia, who tend to avoid sunlight and remain fully covered outside (69). Vitamin D deficiency also occurred in Australian hip fracture patients (70).

View larger version (15K): [in this window] [in a new window]   Figure 3. Mean values of serum 25(OH)D from the studies in Table 1 according to geographical region (Fig. 3A) or to subject/patient/residence category (Fig. 3B).

View larger version (14K): [in this window] [in a new window]   Figure 4. Serum 25(OH)D (median, 5th-95th percentile) in 250 healthy adults (blood donors), 74 independent elderly subjects, 142 institutionalized elderly patients, and 125 patients with hip fracture. The samples in all groups were collected throughout the year. All measurements were performed by HPLC followed by competitive protein binding assay (data from Refs. 37 56 85 ). [Reproduced with permission from M. E. Ooms: Thesis. Vrije Universiteit Amsterdam, 1994 (105 ).]

View larger version (13K): [in this window] [in a new window]   Figure 5. Serum 25(OH)D measured in elderly people in 16 European centers participating in the Euronut SENECA Study. The points represent the mean values of each center for males and females according to northern latitude. The lowest values were found in Greece, Spain, and Italy. [Reproduced with permission from R. P. J. Van der Wielen et al.: Lancet 346:207–210, 1995 (45 ). © The Lancet Ltd.]

One may conclude that vitamin D deficiency is very common in patients with hip fracture and in institutionalized elderly in European countries, but it also occurs in hospital patients and homebound elderly in the United States (40, 41). Vitamin D intake is very low in European countries, but low intakes are also encountered in the United States.

D. Changes in vitamin D metabolism with aging
As stated above, the efficiency of vitamin D production in the skin from its precursor 7-dehydrocholesterol decreases with aging (Fig. 2). The absorption of vitamin D is adequate even at very advanced ages. Oral vitamin D₂, 50,000 IU, led to similar increases of serum vitamin D₂ concentration in elderly subjects and young adults (39). In a study in 142 elderly subjects (mean age 82 yr) in a home for the elderly and a nursing home, a vitamin D supplement of 400 or 800 IU/d increased serum 25(OH)D from 24 to 65 or 75 nmol/liter, respectively (56). The serum 25(OH)D concentrations increased to more than 40 nmol/liter in all subjects who received a supplement. This indicates that absorption of vitamin D₃ is very adequate in the elderly. Of course, malabsorption such as that found in celiac disease may compromise the absorption of vitamin D₃. On the other hand, the pathogenesis of osteomalacia in malabsorption also involves poor calcium absorption and increased vitamin D turnover due to secondary hyperparathyroidism (87).

Vitamin D is hydroxylated in the liver to 25(OH)D. This hydroxylation step is well preserved in old age, but may be compromised by liver disease (88, 89). Further hydroxylation to 1,25-(OH)₂D occurs in the kidney. The renal 1-hydroxylase activity may decrease with aging parallel to the decrease of renal function (90). The increase of serum 1,25-(OH)₂D after infusion of human PTH for 24 h (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) was lower in elderly patients with vertebral fractures than in younger adults (91). The increase of serum 1,25-(OH)₂D after PTH infusion correlated negatively with age and positively with glomerular filtration rate and was lower in patients with hip fracture than in healthy elderly patients (92). However, in the Baltimore Aging Study, serum 1,25-(OH)₂D was not lower in very healthy elderly subjects than in young and middle-aged adults (78). Experimentally, the age-related loss of 1-hydroxylase activity was reversible in old rats on a low phosphorus or low calcium diet when infused with IGF-I (93, 94). This indicates that aging in itself does not necessarily cause a decrease of serum 1,25-(OH)₂D.

The age-related decrease in intestinal calcium absorption has been attributed to the decrease in serum 1,25-(OH)₂D. However, it appears that the decrease of calcium absorption with aging is partly independent of vitamin D (95). The serum concentration of 1,25-(OH)₂D is tightly controlled by negative feedback. When serum 1,25-(OH)₂D increases, gut calcium absorption also increases. The relatively high serum calcium decreases the secretion of PTH, leading to a lower production of 1,25-(OH)₂D. In case of severe vitamin D deficiency, serum 25(OH)D is low, and the production of 1,25-(OH)₂D may be restricted by lack of substrate (37, 56, 58). In vitamin D-deficient elderly and in patients with hip fracture, a positive correlation between serum 25(OH)D and serum 1,25-(OH)₂D has been observed, indicating substrate-dependent synthesis of 1,25-(OH)₂D (56, 58, 96). Serum DBP decreases with aging similarly to serum albumin. The lower serum 25(OH)D and 1,25-(OH)₂D concentration in frail elderly people and in patients with hip fracture could be due to the decrease of DBP, because, for the most part, the vitamin D metabolites are bound to DBP (14, 97). However, the free 25(OH)D index and free 1,25-(OH)₂D index [ratio 25(OH)D/DBP and 1,25-(OH)₂D/DBP] were lower in patients with hip fracture than in healthy elderly controls, indicating real deficiency in patients with hip fracture (86). In conclusion, vitamin D metabolism is relatively normal in healthy elderly people, but chronic diseases may interfere with it. The formation of 1,25-(OH)₂D may be restricted by impairment of renal function and by lack of substrate in the case of vitamin D deficiency.

	III. Consequences of Vitamin D Deficiency

Top Abstract I. Introduction II. Vitamin D Status... III. Consequences of Vitamin... IV. Other Causes of... V. Functional Classification and... VI. Prevention and Treatment VII. Public Health Aspects VIII. Future Prospects IX. Conclusion References

 
A. Secondary hyperparathyroidism and high bone turnover
A low serum 25(OH)D concentration is the hallmark of vitamin D deficiency. The low serum 25(OH)D concentration leads to a small decrease of serum 1,25-(OH)₂D and calcium absorption. The lower serum calcium concentration causes an increase of PTH secretion, which stimulates the production of 1,25-(OH)₂D. By this mechanism serum 1,25-(OH)₂D is kept at (nearly) normal levels at the expense of a higher serum PTH concentration, which is referred to as "secondary hyperparathyroidism." It implicates that serum PTH is relatively high for the associated serum calcium concentration, although it may still be within normal reference limits. As a consequence of the seasonal variation of serum 25(OH)D, vitamin D deficiency, when present, is most marked at the end of the winter months. As can be expected, serum PTH was observed to exhibit an inverse seasonal variation with high levels at the end of winter and low levels at the end of summer when serum 25(OH)D is at its maximum (76, 98). The increased serum PTH causes an increase of bone turnover, which is usually associated with (primarily cortical) bone loss. As is known from studies in primary hyperparathyroidism, the trabecular bone is relatively preserved: the bone mineral density in most patients with primary hyperparathyroidism is normal in the lumbar spine, while it is lower in the femoral neck (99). This was confirmed by histomorphometric studies (100, 101).

Secondary hyperparathyroidism has been proposed as the principal mechanism whereby vitamin D deficiency could contribute to the pathogenesis of hip fractures. Many investigators have observed increased serum PTH concentrations in elderly people with or without hip fractures associated with vitamin D deficiency (37, 56, 61, 70, 102). Serum PTH correlated negatively with serum 25(OH)D in many studies (10, 37, 56, 57, 61), usually with a correlation coefficient between 0.20 and 0.30 (Fig. 6). The correlation coefficient may be somewhat higher when restricted to the low range of serum 25(OH)D and when confounding variables (e.g., serum creatinine) are controlled. The latter is important as many two-site intact PTH (1-84) assays may actually also measure inactive fragments such as PTH (7-84) accounting for more than 50% of the observed value in case of impaired renal function (103). The inverse relationship between serum 25(OH)D and serum PTH has been observed not only in the elderly but also in postmenopausal women aged 45–65 yr (10, 76, 102).

View larger version (28K): [in this window] [in a new window]   Figure 6. A, Negative relationship between serum PTH and serum 25(OH)D in 330 elderly women (>70 yr) in Amsterdam. The best fit was obtained by a linear regression model with a threshold for serum 25(OH)D at 25 nmol/liter with a negative correlation below the threshold (P = 0.02) and no significant correlation above the threshold. B, Negative relationship between serum PTH and serum 25(OH)D in 1,569 adults (50 ± 6 yr) from the SUVIMAX study. The best fit was obtained by nonlinear regression analysis (P < 0.01). A plateau for serum PTH was reached at serum 25(OH)D above 78 nmol/liter. Below this level, serum PTH started to rise. [Panel A reproduced with permission from M. E. Ooms: Thesis. Vrije Universiteit Amsterdam, 1994 (105 ); panel B reproduced with permission from M. C. Chapuy et al.: Osteoporos Int 7:439–443 1997 (10 ). © International Osteoporosis Foundation and National Osteoporosis Foundation.]

B. Bone histology in patients with hip fractures
About 30 yr ago, Chalmers and colleagues (6, 7) pointed to the frequent occurrence of osteomalacia in elderly women, often occurring after gastrectomy and often associated with hip fractures. Since then, many reports have been published on increased osteoid tissue (hyperosteoidosis) and osteomalacia in patients with hip fracture. Osteomalacia can only be diagnosed reliably in a nondecalcified bone biopsy, labeled with tetracycline before the biopsy to assess mineralization. Osteomalacia is characterized by an increase of the bone surface covered by osteoid seams and an increase of the osteoid thickness, which usually measures more than 4 lamellae (3, 12). The cornerstone of the diagnosis of osteomalacia is the demonstration of a reduction in mineral apposition rate, mineralization surface, and bone formation rate, which can be measured after the administration of double tetracycline labels before the bone biopsy (107). Absence of double fluorescent labels and a low mineralization surface combined with increased osteoid thickness indicates severe osteomalacia (108). Bone biopsies have been extensively studied in patients with hip fracture because they can easily obtained during the operation. However, tetracycline labels cannot be administered before the hip fracture, implicating that the mineral apposition rate cannot be measured. Studies of bone biopsies in patients with hip fracture are summarized in Table 2. The incidence of osteomalacia in these studies ranges from 0 to 37%. However, comparison between these studies is hampered by the use of different criteria for osteomalacia. One may avoid this problem by referring to the increased amount of osteoid tissue as "hyperosteoidosis." Many studies show an increase of osteoid surface and eroded surface (resorption surface) compatible with high bone turnover associated with secondary hyperparathyroidism. In 89 biopsies of patients with hip fractures from The Netherlands, high turnover characterized by osteoid surface > 18% and eroded surface > 6% was observed in 22% of the biopsies (96). In this study osteoid thickness was not increased in any biopsy. Six studies report the measurement of osteoid thickness, and in three of these osteoid thickness was increased in 5 to 12% of the patients (113, 115, 120). In series from the United Kingdom, increased osteoid values appear more common than in series from other countries (126). The excess osteoid tissue shows a seasonal variation with the highest values in spring and the lowest in autumn, reversed to the seasonal variation of serum 25(OH)D, as may be expected (127). A more recent well documented study (115) reported osteomalacia in 12% of the patients with hip fracture, and these biopsies had increased osteoid volume (mean 12.5%) and osteoid thickness (>13 µm) associated with severe vitamin D deficiency [mean serum 25(OH)D 10.6 nmol/liter]. Frank osteomalacia with radiographically demonstrated Looser zones was reported in several series of patients with hip fracture (7, 110, 118, 121) or subtrochanteric fracture (128). A very sunny climate as in Qatar does not exclude osteomalacia where Looser zones were observed in 6 of 69 patients with hip fracture (121).

C. Vitamin D status and bone mineral density
In cross-sectional studies, a positive relationship has been observed between serum 25(OH)D and bone mineral density (BMD) of the hip. In 330 elderly women in Amsterdam, a threshold was observed (57). The positive correlation was significant when serum 25(OH)D was lower than 30 nmol/liter, but above this level the relationship was no longer significant (Fig. 7). The relationship appeared to be stronger for cortical bone (femoral neck) than for trabecular bone (trochanter). As can be seen in Fig. 7, the BMD at the femoral neck was 5 or 10% lower than average when serum 25(OH)D was 20 or 10 nmol/liter, respectively. A positive relationship between serum 25(OH)D and BMD of the hip was also observed in middle-aged women in the United Kingdom (45–65 yr) (51) and in elderly women in New Zealand (71). Similarly, a negative relationship has been observed between BMD of the hip and serum PTH (51, 130, 131).

View larger version (24K): [in this window] [in a new window]   Figure 7. Relationship between serum 25(OH)D and BMD of the femoral neck in 330 elderly women. The best fit was obtained by a linear regression model with a threshold for serum 25(OH)D at 30 nmol/liter. The correlation was significant (P < 0.001) when serum 25(OH)D < 30 nmol/liter. With higher values of serum 25(OH)D, the correlation with BMD was no longer significant. [Reproduced from M. E. Ooms et al.: J Bone Miner Res 10:1177–1184, 1995 (57 ) with permission from the American Society for Bone and Mineral Research.]

In addition, there is an irreversible component of bone loss. Remodeling balance per osteon usually is negative in the elderly. In a state of high turnover, the negative remodeling balance is multiplied by the high number of remodeling osteons, whereas in low turnover states, bone loss due to negative remodeling balance is much lower (23). Quantitative observations on the irreversible bone loss in osteomalacia were made by Parfitt et al. (133) in 28 patients who were treated with vitamin D and calcium. The osteoid volume decreased by 80% in cortical (from 6 to 1.5%) and trabecular bone (from 30 to 6%). Cortical porosity decreased from 10.3 to 7.8%. Mineralized bone volume increased by 7.5% in cortical and 40% in trabecular bone. The mineral deficits decreased after treatment from 42 to 36% in cortical and from 32 to 6% in trabecular bone. At the end, the irreversible cortical bone loss was due to cortical thinning (endosteal resorption) caused by secondary hyperparathyroidism (133). The bone mineral deficit in osteomalacia is much larger than that in milder degrees of vitamin D deficiency. The bone loss of 5 to 10%, due to secondary hyperparathyroidism in vitamin D-deficient elderly, is quite comparable to the bone loss in primary hyperparathyroidism, which is reversible after successful parathyroidectomy (134).

Another interpretation of the association between low serum 25(OH)D and low BMD is a sedentary life style. This would cause bone loss due to immobility as well as reduced exposure to sunlight. Indeed, frail elderly subjects are often vitamin D deficient and are physically not very active. However, immobility is associated with increased bone resorption, which causes suppression of parathyroid activity and a low serum PTH (135).

D. Vitamin D deficiency as a risk factor for fractures in epidemiological studies
Epidemiological data on vitamin D deficiency as a predictor for bone loss or osteoporotic fractures are scarce. In the Study of Osteoporotic Fractures, a study in four US centers, risk factors for osteoporosis were studied prospectively in 9,704 elderly women. Vitamin D deficiency as a risk factor was studied using the case-cohort approach in 133 women with first hip fracture and 138 women with a new vertebral fracture (136). Mild vitamin D deficiency [serum 25(OH)D < 47 nmol/liter, prevalence 22%] was not associated with an increased risk for hip or vertebral fracture. A low serum 1,25-(OH)₂D ( 57 pmol/liter) was associated with an increased risk for hip fracture [risk ratio (RR) 2.1 adjusted for age and weight]. Serum PTH was not a risk factor in this study, but low estradiol and high SHBG were both associated with an increased risk for hip and vertebral fracture. In the Study of Osteoporotic Fractures, severe vitamin D deficiency was rare, and this may be the reason that vitamin D deficiency was not associated with an increased risk for osteoporotic fractures in this study. In Oslo, Norway, 246 patients with hip fracture and a similar number of controls were studied for risk factors. A vitamin D intake lower than 100 IU/d was associated with an increased risk for hip fracture [RR 3.9, confidence level (CI) 1.7–9.3] (137).

Bone loss from the femoral neck was prospectively studied during 16 yr in 304 women aged 30–49 yr from Rochester, Minnesota. Independent predictors of bone loss in elderly women included dietary vitamin D intake, hormone replacement therapy, and serum osteocalcin (138). However, vitamin D intake was not a risk factor for bone loss in pre- or postmenopausal women in this study. The efficacy of drugs preventing hip fracture was investigated in a retrospective case-control study (MEDOS) in 2,086 women with hip fracture and 3,532 controls (139). Estrogen, calcium, and calcitonin significantly reduced the risk, but vitamin D compounds did not. However, only 4% used vitamin D preparations. Subgroup analysis of these data showed that the relative risk of hip fracture was significantly lower in women aged above 80 yr when using vitamin D (RR 0.63, CI 0.40–0.98) and in women with a body mass index (BMI) below 20 kg/m² when using vitamin D (RR 0.45, CI 0.24–0.84) (140).

Vitamin D deficiency is less common in the elderly in the United States than in most European countries. This is an argument against vitamin D deficiency as a risk factor for hip fractures. However, within a country, hip fractures may be more frequent in vitamin D-deficient than in vitamin D-replete elderly subjects. More prospective epidemiological studies on risk factors for osteoporotic fractures including vitamin D deficiency are underway, e.g., the European Prospective Osteoporosis Study, the Rotterdam Study, and the Longitudinal Aging Study Amsterdam.

E. Vitamin D deficiency and myopathy
Vitamin D deficiency is associated with muscle weakness. Muscle contraction and relaxation are abnormal in vitamin D deficiency, and these are corrected by vitamin D independently of changes in mineral levels (20). Improvement of myopathy occurs after very low doses of 1,25-(OH)₂D. Treatment of patients with osteoporosis with 1-hydroxyvitamin D increased succinate dehydrogenase and phosphorylase activity and increased the number of IIA muscle fibers (141). Muscle weakness and hypotonia were observed as initial manifestation of severe vitamin D deficiency in an intensive care patient (142). Muscle strength correlated positively with serum 1,25-(OH)₂D in geriatric patients and with serum 25(OH)D in elderly male patients (143). Serum 25(OH)D was lower, and serum PTH was higher in Australian nursing home residents who fell than in other residents who did not fall (144). The association between falling and serum PTH persisted after multiple adjustments. A prospective study on risk factors for falls was done in 354 elderly subjects participating in a double blind clinical trial to evaluate the effect of vitamin D₃ supplementation on the incidence of hip fractures (Amsterdam Vitamin D Study). The incidence of falls and recurrent falls was similar in the vitamin D and placebo groups (145). The results of intervention studies with vitamin D on muscle strength are conflicting (see below).

F. Other consequences of vitamin D deficiency
During the last 20 yr many new actions of vitamin D metabolites, especially 1,25-(OH)₂D, have been discovered. The presence of the vitamin D receptor has been demonstrated in many organs, often without direct relevance for vitamin D action (17). New actions have been reviewed extensively (19, 21, 146). Some of these actions might decline in vitamin D-deficient elderly subjects. Although it is not the main subject of this review, some actions may be very relevant because of the frequent occurrence of vitamin D deficiency in the elderly. The active metabolite, 1,25-(OH)₂D, has been shown in numerous systems to decrease cell growth and induce cell differentiation (19, 147). It has been tested in the treatment of cancer and lymphoma, and systemic and topical 1,25-(OH)₂D has been successful in the treatment of psoriasis (13). Epidemiological studies have suggested that vitamin D deficiency is associated with colon and breast cancer (148, 149, 150).

Vitamin D status may influence immunological function. Vitamin D deficiency is also associated with impaired macrophage function (151); it is associated with infection in children with rickets and in adults with disseminated tuberculosis and with anergy to skin testing (146).Vitamin D status also influences insulin secretion. Vitamin D deficiency results in a decreased insulin response to glucose, which is corrected by 1,25-(OH)₂D (152). Treatment of a vitamin D-deficient patient with vitamin D improved glucose tolerance and ß-cell function (153). Inadequate vitamin D status has been implicated as a factor contributing to syndrome X, i.e., insulin resistance, obesity, hypertension, glucose intolerance, and dyslipidemia, but response to vitamin D treatment has been variable (154). Vitamin D is also involved in other endocrine organs such as the pituitary and the testis, but deleterious effects of vitamin D deficiency are not established for these organs. Vitamin D deficiency has also been associated with progression of osteoarthritis in the Framingham Study (155) and with an unusual pain syndrome characterized by severe hyperesthesia, which resolved after vitamin D treatment (156). Although vitamin D deficiency may have potentially important consequences along these pathways, the effects on health status in the elderly are uncertain.

G. Racial differences in vitamin D and PTH metabolism
Interesting observations have been made in American blacks compared with whites. The incidence of hip fractures is considerably lower in black than in white people (157, 158). A higher BMD was observed consistently in blacks compared with whites (159, 160). When adult black and white women and men were compared, serum 25(OH)D was significantly lower and serum 1,25-(OH)₂D was significantly higher in blacks (161). Similar observations were made when black and white adolescents were compared (162). In later studies, a higher serum 1,25-(OH)₂D in blacks than in whites was observed by some investigators (163, 164) but not by others (165, 166). A significant increase of serum PTH was found in some (161, 164, 165) but not in other studies (166, 167). Biochemical markers of bone turnover, such as serum osteocalcin, suggested a higher bone turnover in blacks than in whites (161) or a similar turnover (164). Urinary calcium excretion has consistently been lower in black than in white people (161, 162, 164, 166). A detailed study of calcium absorption in response to calcitriol (1,25-(OH)₂D) showed a lower response in blacks than in whites, suggesting a gut resistance to 1,25-(OH)₂D (164). Little data are available on vitamin D status and metabolism in elderly black people. Although the literature suggests changes in the vitamin D endocrine system, such as an increase of serum 1,25-(OH)₂D and serum PTH, conclusions on the relevance of these changes for the low incidence of hip fractures in blacks cannot yet be drawn.

	IV. Other Causes of Secondary Hyperparathyroidism in the Elderly

Top Abstract I. Introduction II. Vitamin D Status... III. Consequences of Vitamin... IV. Other Causes of... V. Functional Classification and... VI. Prevention and Treatment VII. Public Health Aspects VIII. Future Prospects IX. Conclusion References

 
A. Renal function
Renal function slowly decreases with aging. Population studies show a gradual decrease of the glomerular filtration rate from about 125 ml/min at age 20 to about 60 ml/min at age 80. This is associated with a gradual increase of serum PTH with age (168, 169). Serum PTH correlates positively with serum creatinine. Several mechanisms may explain this. The slight increase of serum phosphate directly increases parathyroid function. The lower formation of 1,25-(OH)₂D causes a small decrease in calcium absorption, and the lower serum calcium increases serum PTH. Multiple regression analysis in studies of elderly patients with hip fracture show both renal function and serum 25(OH)D as determinants of serum PTH (70, 170, 171). The commonly prescribed diuretic furosemide may also induce secondary hyperparathyroidism. In Australian nursing home residents, furosemide was a more important predictor of serum PTH than renal function and serum 25(OH)D (172). Furosemide increases calcium excretion and lowers ionized calcium by inducing alkalosis (173), thereby increasing serum PTH. Furosemide was a negative predictor for BMD of the hip in institutionalized elderly patients in Amsterdam (174).

B. Estrogen deficiency
Interactions between estrogen status and serum PTH have also been reported. The rise in serum PTH with aging does not occur in women receiving estrogen replacement therapy. In a cross-sectional study in 351 women from 20 to 90 yr of age, mean serum PTH increased from 1.8 to 3.2 pmol/liter. This increase was not observed in 54 women receiving estrogen replacement therapy (175). Estrogen status may also influence secondary hyperparathyroidism caused by vitamin D deficiency. In 348 elderly women participating in the Amsterdam Vitamin D Study, an interaction was observed between the serum concentration of sex hormone binding globulin (SHBG), which correlates negatively with the free estrogen concentration, and the relationship between serum PTH and 25(OH)D (57). Mean serum PTH was high in vitamin D-deficient elderly with high serum SHBG (associated with low free estrogen concentration), and normal in vitamin D-deficient elderly women with low serum SHBG (suggesting high free estrogen concentration). Thus, estrogen appeared to suppress secondary hyperparathyroidism in response to vitamin D deficiency. The interaction of serum SHBG with secondary hyperparathyroidism also was apparent in the BMD increase after treatment with vitamin D₃ (see below).

C. Low calcium nutrition
A low calcium intake also increases PTH secretion (176). In a study in postmenopausal women (mean age 70 yr), it was shown that increasing calcium intake from 800 to 2,400 mg/day caused a decrease of serum PTH of 30% during 24 h (177). It has been suggested that calcium intake modulates the age-related increase in serum PTH and bone resorption. In addition, a low calcium intake may influence vitamin D metabolism. It was observed that vitamin D deficiency occurs after partial gastrectomy even when sunshine exposure is normal (87). The low calcium intake (due to low dairy intake) after gastrectomy causes an increase of serum PTH and consequently of serum 1,25-(OH)₂D. Metabolic studies in rats on a low calcium intake demonstrated that these increases in serum PTH and 1,25-(OH)₂D were associated with increased metabolic clearance of 25(OH)D (178). In primary and secondary hyperparathyroidism, the half-life of serum 25(OH)D was inversely correlated to serum 1,25-(OH)₂D (87). The (relatively) high 1,25-(OH)₂D level in secondary hyperparathyroidism is associated with a high turnover of 25(OH)D (Fig. 8). A low calcium intake by causing a high serum 1,25-(OH)₂D thus may lead to an increased catabolism of 25(OH)D, thereby decreasing serum 25(OH)D and inducing or aggravating vitamin D deficiency. The reverse may also be true: a high calcium intake may suppress serum PTH, decrease serum 1,25-(OH)₂D, and thus have a vitamin D sparing effect (179, 180). A very low dietary calcium intake may cause histological osteomalacia. Three children of 4 to 13 yr presented with signs and symptoms of rickets. They had a normal serum 25(OH)D but a very low calcium intake of 125 mg/d (181). The bone biopsies showed high ostoid surface and thickness and a low bone formation rate. A clinical, biochemical, and histological cure was obtained by increasing calcium intake and calcium supplements. In conclusion, a low calcium intake aggravates vitamin D deficiency and its consequences, while a high calcium intake may reduce vitamin D requirement.

View larger version (19K): [in this window] [in a new window]   Figure 8. Relationship between the half-life of 3H-25(OH)D and the serum 1,25-(OH)2D concentration in 49 patients shown as the regression line and 95% confidence limits. Correlation coefficient r = -0.63, P < 0.001. The data are from patients after gastrectomy (O), patients with primary hyperparathyroidism before and after surgery (), and patients with other disorders of bone and mineral metabolism (). When serum 1,25-(OH)2D is high, the half-life of 25(OH)D is short, indicating an increased catabolism that may aggravate vitamin D deficiency. [Reproduced with permission from M. Davies et al.: J Clin Endocrinol Metab 82:209–212, 1997 (87 ) © The Endocrine Society.]

View larger version (21K): [in this window] [in a new window]   Figure 9. Schematic presentation of pathways from vitamin D deficiency and secondary hyperparathyroidism to osteoporotic fractures.

	V. Functional Classification and Diagnosis of Vitamin D-Deficient States

Top Abstract I. Introduction II. Vitamin D Status... III. Consequences of Vitamin... IV. Other Causes of... V. Functional Classification and... VI. Prevention and Treatment VII. Public Health Aspects VIII. Future Prospects IX. Conclusion References

Another point is the definition of secondary hyperparathyroidism. The increase of serum PTH is a compensatory or adaptive mechanism in response to a tendency for low serum calcium. Serum PTH changes during the day and between days in relationship with serum calcium, calcium intake, and calcium supplements (177, 179, 180). When serum PTH increases in winter after the decrease of serum 25(OH)D, this is in principle an adaptive physiological mechanism. This adaptive increase of serum PTH is usually called secondary hyperparathyroidism, although the mean values often are still in the normal reference range (76, 98). The seasonal variation of serum PTH occurs not only in the elderly but also in children (182). An inverse relationship between serum PTH and serum 25(OH)D occurs not only in elderly people but also in children in France and adults in Africa (183, 184).

B. Diagnostic criteria
The most common approach is to define a sufficient vitamin D state according to reference limits for serum 25(OH)D in healthy adults from the population (e.g., blood donors) sampled throughout the year (85). However, this depends on climate, sunshine exposure, and clothing habits, leading to large differences between countries. A functional, health-based classification could be made using serum PTH. This is easy when serum PTH is increased to above the upper reference limit. However, the increases of serum PTH associated with vitamin D deficiency usually are within the normal reference ranges. In a study in Boston, the seasonal variation of serum PTH was no longer visible when serum 25(OH)D was higher than 90 nmol/liter, leading to the conclusion that serum 25(OH)D should be higher than this level to prevent secondary hyperparathyroidism (76). In a French population, serum PTH started to increase when serum 25(OH)D decreased below 78 nmol/liter, leading to a similar conclusion (10). However, in a large vitamin D study in Amsterdam, the negative relationship between serum PTH and serum 25(OH)D was only significant when serum 25(OH)D was lower than 30 nmol/liter (57, 105). Therefore, different data sets lead to different conclusions (Fig. 6). Of course, the differences may partly be due to differences in assays for 25(OH)D (28). Another modifying factor may be the dietary calcium intake. A relatively high calcium intake, as is usual in The Netherlands, may suppress serum PTH and influence the serum 25(OH)D level at which secondary hyperparathyroidism becomes manifest (179). Calcium nutrition by influencing serum PTH also influences the turnover of vitamin D metabolites. The half-life of serum 25(OH)D is shorter in hyperparathyroid states (87). In a vitamin D-deficient state, the synthesis of 1,25-(OH)₂D is substrate-dependent, i.e., dependent on sufficient 25(OH)D, as shown by a positive correlation between serum 25(OH)D and serum 1,25-(OH)₂D (56, 58, 96). Treatment of women with postmenopausal osteoporosis with 25(OH)D showed a very significant increase of serum 1,25-(OH)₂D in those who responded with an increase of intestinal calcium absorption (185). After vitamin D supplementation in the elderly, 1,25-(OH)₂D may increase in parallel with serum 25(OH)D (58). In a study of elderly people in Amsterdam, serum 1,25-(OH)₂D increased only when serum 25(OH)D was lower than 30 nmol/liter (56).

Another parameter of mild vitamin D deficiency (or insufficiency) may be the decrease of serum PTH after vitamin D supplementation. When serum PTH decreases more than 15–20% after vitamin D supplementation, this may point to clinically relevant vitamin D deficiency, leading to bone loss and osteoporosis. In vitamin D supplementation studies, the decrease of serum PTH was 30% in severely vitamin D- deficient psychogeriatric patients (36), 15% in institutionalized elderly (56, 186), and negligible in vitamin D-replete elderly (77). When vitamin D and calcium supplementation are combined, serum PTH may decrease up to 50% (187). Vitamin D supplementation (50,000 IU/wk) with calcium (1,000 mg/d) in 35 elderly patients with a serum 25(OH)D between 25 and 62 nmol/liter decreased serum PTH by 22% (188). The decrease in serum PTH was significant when baseline serum 25(OH)D was lower than 50 nmol/liter. This is about the serum 25(OH)D level below which serum PTH started to rise in the large study of hospital inpatients (40). Similar conclusions can be drawn from the results in the placebo group of the multicenter raloxifene study consisting of 2,529 women treated with vitamin D (400–600 IU/d) and calcium (500 mg/d). In this study, serum PTH decreased by 12% when baseline serum 25(OH)D was lower than 50 nmol/liter (189).

It may be concluded that it is difficult to delineate sharp diagnostic criteria for mild vitamin D deficiency or insufficiency. When the required serum 25(OH)D level is set too high, it will result in a clinically irrelevant diagnosis and unnecessary supplementation. When the required serum 25(OH)D level is set too low, unnecessary bone loss will occur in many patients. A proposal for staging is presented in Table 3. A serum 25(OH)D lower than 50 nmol/liter might be called "mild vitamin D deficiency" or "insufficiency." This is associated with a slightly elevated serum PTH concentration and a mild increase of bone turnover. When serum 25(OH)D is lower than 25 nmol/liter, "moderate" vitamin D deficiency is diagnosed. Serum PTH is moderately increased (up to 30%) and high bone turnover is observed. Severe vitamin D deficiency occurs when serum 25(OH)D is lower than 12.5 nmol/liter. In these cases, serum PTH may be increased 30% or more, and a mineralization defect may occur, ultimately leading to frank osteomalacia.

	VI. Prevention and Treatment

Top Abstract I. Introduction II. Vitamin D Status... III. Consequences of Vitamin... IV. Other Causes of... V. Functional Classification and... VI. Prevention and Treatment VII. Public Health Aspects VIII. Future Prospects IX. Conclusion References

A. Sunshine and UV irradiation
Irradiation with UV light was used in a study in a nursing home to increase vitamin D status. For this purpose, UV fluorescent lighting tubes were suspended from the ceiling at 3 meter height in the day wards. The patients received irradiation for 3 h/d, resulting in an increase of serum 25(OH)D of 25 nmol/liter in 8 wk (191). More recently, a randomized controlled study was done in a psychogeriatric nursing home in The Netherlands. The patients received UV irradiation with half of the minimal erythematous dose on 1,000 cm² of the back three times per week, oral vitamin D₃ 400 IU/d, or served as controls. Serum 25(OH)D increased from around 20 nmol/liter to 60 nmol/liter in both the UV group and the group that received oral vitamin D₃, while there was no change in the control group (Fig. 10). Serum PTH decreased about 30% in both treated groups (36).

View larger version (18K): [in this window] [in a new window]   Figure 10. Response of serum 25(OH)D to UV irradiation (UVB, half of the minimal erythematous dose, 3 times per week) on 1,000 cm2 of the back of elderly women with vitamin D deficiency in comparison with the response to oral vitamin D3 400 IU/d (vit D) and a control group (control). The study included 45 psycho-geriatric patients randomized in three groups. Data are expressed as median, 25th-75th percentile. [Reproduced from V. G. M. Chel et al. : J Bone Miner Res 13:1238–1242, 1998 (36 ) with permission of the American Society for Bone and Mineral Research.]