Vitamin D status and bone mineral density in the Chinese population: a review

Summary Low vitamin D status is associated with low bone mass which, in turn, is an important predictor of fracture. However, data on this relationship in non-Caucasian populations are scarce. This review shows such an association in the Chinese population in five of the 11 included studies. Introduction In the elderly population, the serum 25-hydroxyvitamin D [25(OH)D] concentration is often inadequate. This may cause a lower bone mineral density (BMD), which is an important predictor of fracture. It is estimated that by 2050 more than half of all hip fractures worldwide will occur in Asia. However, data on the relationship between vitamin D status and BMD in a non-Caucasian population are scarce. Therefore, this study reviews the literature on the relationship between serum 25(OH)D and BMD in the Chinese population. Methods A search was made in PubMed, EMBASE, Web of Science and Cochrane Library (up to December 2014) to identify relevant studies using the terms vitamin D status, bone mineral density, and Chinese. Results Of the 293 studies identified, 11 fulfilled the inclusion and exclusion criteria and were analyzed. Mean serum 25(OH)D concentrations ranged from 29-82 nmol/L. In 5 of the 11 studies, an association was found between vitamin D status and BMD in the Chinese population. Conclusion The evidence for a relationship between the serum 25(OH)D concentration and BMD in the middle-aged and elderly Chinese population living in Asia appears to be limited and inconsistent.


Introduction
Vitamin D is important for the development, growth, and maintenance of a healthy skeleton. Low vitamin D status is suggested to contribute to bone loss and a low bone mineral density (BMD) is an important predictor of osteoporotic fractures. Fractures cause impaired quality of life [1], are associated with increased mortality [2] and are accompanied by high cost. Bone loss also occurs with aging and the prevalence of osteoporosis is expected to increase as life expectancy rises. As a result of a large aging Asian population, it is estimated that by 2050 more than 50 % of all hip fractures worldwide will occur in Asia [3]. For example, in Hong Kong, the incidence of hip fractures in older men and women more than doubled between 1966 and 1985 [4]. In Beijing, the incidence rate of hip fracture showed a greater increase from 2002 to 2006 than from 1990 to 1992 [5]. Rapidly increasing industrialization and urbanization and reduction of weight-bearing activities are also associated with an increase in the incidence of fractures in various areas in Asia [6]. By contrast, during the past decade a decline has been reported in the incidence of hip fractures in the more developed regions of Asia (e.g., Hong Kong). An increase in BMD levels, and a healthier lifestyle with more physical activity and higher vitamin D status, may explain this decline in the incidence of osteoporosis and fractures [7]. Vitamin D 3 , or cholecalciferol, is primarily produced in the skin from 7-dehydrocholesterol when exposed to the ultraviolet light of the sun (UVB 290-315 nm). It can also be obtained from food, especially fatty fish such as salmon, mackerel, eel, and herring and may also be added to food as fortification, e.g., to margarine or milk. In the liver vitamin D 3 is rapidly hydroxylated into 25-hydroxyvitamin D [25(OH)D], which has low biological activity but is the major circulating metabolite. It is further hydroxylated in the kidney into 1,25dihydroxyvitamin D [1,25(OH) 2 D], the active form of vitamin D. This latter hydroxylation is stimulated by parathyroid hormone (PTH). The main effect of 1,25(OH) 2 D is to stimulate the intestinal absorption of calcium and phosphate for the mineralization of the skeleton and for neuromuscular action.
The vitamin D status is usually assessed by measuring the serum 25(OH)D concentration. Insufficient exposure to sunlight, covering of the skin, skin pigmentation, a diet that is low in vitamin D and calcium, and advanced age may negatively affect vitamin D status. Consequences of severe vitamin D deficiency include rickets in children and osteomalacia in adults. In addition, it may lead to secondary hyperparathyroidism, high bone turnover, bone loss, mineralization defects and it predisposes to fractures. Muscle weakness is another consequence of low 25(OH)D levels.
Studies among Caucasian populations reported the relationship between vitamin D status and BMD, with evidence of a positive association [8,9]. By contrast, low 25(OH)D concentrations were not correlated with low BMD among Black individuals [10] and among Black or Hispanic men [11]. These findings suggest that the relationship between 25(OH)D and BMD may differ by race and ethnicity. There are various reasons why the relationship between 25(OH)D and BMD might be different in the Chinese population compared to the Caucasian population. First, genetic factors, for example variants of vitamin D metabolizing enzymes CYP24A1 (24-hydroxylase), CYP2R1 (25-hydroxylase), vitamin D binding protein (DBP), and DHCR7 (7dehydrocholesterol reductase) can contribute to variation in 25(OH)D concentrations [12,13]. Second, the average dietary calcium intake in the Chinese population is low, leading to secondary hyperparathyroidism and increased turnover of vitamin D metabolites. This may also affect the relationship between 25(OH)D and BMD. In addition, some lifestyle factors, e.g., (outdoor) physical activity, may also influence vitamin D status and BMD, and other lifestyle factors (e.g., behavior in the sun) may differ between races as, for example, the Chinese population tend to have a sun-avoiding behavior.
Information on the relationship between 25(OH)D and BMD in the Chinese population is scarce but is relevant in view of the high occurrence of fractures in Asia. In addition, insight into the relationship may help formulate recommendations for the use of vitamin D supplementation to prevent osteoporosis. Therefore, the present study reviews the literature on the relationship between vitamin D status and BMD in the middle-aged and older Chinese population. The influence of the latitude of the area, season, gender, and age on serum 25(OH)D levels is also examined.

Methods
The following electronic databases were searched up to December 16 2014: PubMed, EMBASE, Web of Science, and Cochrane Library. Keywords included vitamin D status, bone mineral density, and Chinese. The search strategy included both truncated free-text and MeSH terms and was adjusted according to different databases. The complete search strategy is presented in Appendix A. Restriction was made to full-text articles and abstracts in the English and Dutch language. Chinese full-text articles with an English abstract were translated into English. Publications were eligible when reporting an association between serum 25(OH)D concentrations and BMD or bone mineral content (BMC) in the Chinese population.
All serum 25(OH)D concentrations were converted to nmol/L (1 ng/ml = 2.496 nmol/L). Figure 1 presents the results of the literature search; reasons for exclusion based on title, abstract, or full text are given. The characteristics of the selected studies are outlined in Table 1.

Results
Of the 11 included studies most had a cross-sectional design. Two studies combined a cross-sectional design with a 4and 5-year follow-up study [14,15]. Five studies included only women [16][17][18][19][20], five studies included both men and women [15,[21][22][23][24], and one study included only men [14]. The mean age of the subjects in most of the studies ranged from 60 to 76 years. The studies performed in Lanzhou [24] and Taipei [19] differed from the other studies in that they included men and women from age 40 years, and included both premenopausal and postmenopausal women, respectively.
Of the 11 studies, five reported an association between the serum 25(OH)D concentration and BMD [14,15,17,18,21] while six studies did not observe an association [16,19,20,[22][23][24]; however, various sites for BMD measurement were used. Most studies measured the BMD at the lumbar spine, femoral neck, and total hip [14,17,20,21]. Some studies measured the BMD at one or two of these three sites [15,16,19], or some of these sites were combined with the BMD of the trochanter [21,23], or different regions in the hip [22]. In addition, some studies measured the BMD of total body [18], or of the calcaneus [24]. Finally, one study used size-adjusted BMC rather than BMD because of the    [23]. All studies measured the BMD by dual energy X-ray absorptiometry (DXA), except one study which measured the calcaneal density by quantitative ultrasound (QUS) [24]. From the five cross-sectional studies with a mean baseline serum 25(OH)D concentration <50 nmol/L, two studies [17,18] showed an association between 25(OH)D and BMD, while three studies [20,23,24] did not. From the crosssectional studies with a mean baseline 25(OH)D concentration >50 nmol/L, two studies [14,21] revealed an association and three other studies [16,19,22] did not.
Although some studies with Caucasian populations observed a threshold serum 25(OH)D with regard to BMD [8], none of the Chinese studies reported a threshold for the association between serum 25(OH)D concentration and BMD.
Studies performed in north-east and north-west China showed a high prevalence of low vitamin D status [15,18,20,23]. In these studies the mean serum 25(OH)D concentration of around 30 nmol/L was remarkably lower than in studies performed in south-east Asia, which reported a mean serum 25(OH)D concentration >50 nmol/L (Fig. 2).
Seasonal variation in serum 25(OH)D concentration was not observed in Hong Kong [21], whereas in Taiwan statistically lower serum 25(OH)D levels were reported when blood samples were taken in winter [19] (Fig. 2). The remaining studies included in this review did not examine seasonal variation in 25(OH)D levels.
In three studies [16,20,22], markers of bone turnover, e.g., pro-collagen Type 1 N-terminal peptide (P1NP), osteocalcin (OC), and C-telopeptide β cross link of Type 1 collagen (CTX-β), were negatively correlated with BMD at various sites. In two studies, no association was found between bone turnover markers and serum 25(OH)D [19] or PTH [23].  In woman, one study found a positive association between physical activity (number of times per week walking with a load) and BMD at the femoral neck and trochanter [22]. In another study, low physical activity was associated with serum 25(OH)D concentrations <75 nmol/L [21]. Moreover, moderate physical activity was found to be protective against vitamin D deficiency in both men and women [24].

Discussion
The present review found no consistent relationship between the serum 25(OH)D concentration and BMD in the Chinese population: of the 11 included studies, five reported an association while six did not.
The majority of previous studies on this topic were performed in Caucasian populations. In a study among adults in the USA, higher 25(OH)D concentrations were associated with higher BMD among men and women [25]. Also, a positive association was revealed in the Longitudinal Aging Study Amsterdam (LASA) performed in the Netherlands, with a threshold of serum 25(OH)D concentration of about 50-60 nmol/L, i.e., no relation above this level [8]. In a multicenter trial among postmenopausal women with osteoporosis from 29 countries, of which 1.3 % were of Asian origin, a significant relationship was found between serum 25(OH)D levels and BMD [26]. In an Italian population of postmenopausal women a positive correlation was found between 25(OH)D concentrations and BMD at spine and hip [27], and a weak correlation was shown among ambulatory women living in Argentina [28]. A study performed in Japan among postmenopausal women reported that higher 25(OH)D concentrations were associated with a higher BMD of femoral neck [29]. In Manchester (UK), a study conducted among Pakistani women showed an association between 25(OH)D concentrations and BMD at femoral neck, total hip, lumbar spine, and distal radius BMD [30]. A modest positive correlation between the serum 25(OH)D concentration and BMD was found in White men but not in Black or Hispanic men [11]. By contrast, other studies did not show an association [31,32].
A clear conclusion regarding the outcome of the relationship between serum 25(OH)D and BMD may be hampered by heterogeneity in, for example: inclusion and/or exclusion criteria, selection of participants, adjustments for different variables, site of BMD measurement, study size, and mean baseline serum 25(OH)D concentration.
Adjustment for confounding variables, regarded as one of the quality criteria of observational studies [33], differed between the studies (Table 1). Physical activity, which may have an effect on BMD, was described in several studies. However, no study investigated the interaction between physical activity and the relation between 25(OH)D and BMD. Two large studies, with 1000 [14] and 10,000 [24] subjects, adjusted for physical activity. Only the study with higher baseline serum 25(OH)D concentrations [14] found a cross-sectional association between serum 25(OH)D and BMD, but not in the longitudinal design. This may be due to the protective action of higher serum 25(OH)D concentrations on bone loss, since more than 50 % of the subjects had serum 25(OH)D concentrations >75 nmol/L [14]. In contrast, another smaller longitudinal study [15], with 100 subjects and a mean baseline serum 25(OH)D concentration of 30 nmol/L, did find an association between 25(OH)D and bone loss.
Of the six cross-sectional studies with a smaller study population, four of them [16,19,22,23] did not show an association between 25(OH)D and BMD. However, a smaller study size may reduce the chance to detect a significant association. Three studies [16,19,22] had a mean baseline 25(OH)D concentration of >50 nmol/L and one study [23] of <50 nmol/L. On the other hand, by far the largest study [24], with a mean baseline 25(OH)D concentration of <50 nmol/L, also found no association. Apparently, neither the number of subjects nor the mean baseline 25(OH)D concentration seem to predict the outcome of the relationship between serum 25(OH)D and BMD.
Other variables, such as calcium supplementation and dietary calcium intake, were described in several studies [16,19,21,23], but only one study adjusted for calcium supplementation [17]. Although low calcium intake may induce or aggravate vitamin D insufficiency [9], the influence of calcium is not clear from these studies.
The included studies did not explore genetic factors, although these may also influence serum 25(OH)D concentration [12,13] and, therefore, affect the relationship between vitamin D status and BMD.
The season of 25(OH)D sampling was not described in three studies [16,20,22]. However, seasonal variation of serum 25(OH)D did not influence BMD values for summer and winter in Taiwan [19]. This is not remarkable when taking into account that the mean serum 25(OH)D concentration in Taiwan is >70 nmol/L, even in the winter.
One study differed from the others by measuring calcaneal density by quantitative ultrasound; however, this method may predict fracture risk as well as DXA [34].
Previous studies reported a strong relationship between the latitude of the area and vitamin D. Serum 25(OH)D levels tend to increase with decreasing latitude [26,35]. Given the latitude of Beijing (40°N) and Shenyang (42°N) with a climate that is characterized as mild in spring and autumn, hot in summer, and very cold in winter with low exposure to sunlight, it is not surprising that serum 25(OH)D levels were lower compared with the southern regions of China, especially in late winter. Similar results for low serum 25(OH)D levels were reported in adolescent girls living in Beijing [36,37]. However, considering Malaysia's latitude of 3°N it is noteworthy that mean serum 25(OH)D concentrations of no more than 60 nmol/L were found in Kuala Lumpur [16]. In addition, in 50 % of the Malaysian Chinese women, serum 25(OH)D concentrations <50 nmol/L were reported. The reasons for this are unclear, but might be explained by sun-avoiding behavior, use of sunscreen, clothing habits, the hot weather, air pollution, and by limiting outdoor activities.
The relationship between vitamin D status and gender is also unclear. Serum 25(OH)D levels in Shenyang women were higher than in men [23]. In Hong Kong, male sex was associated with lower serum 25(OH)D levels [21]. By contrast, another study in Hong Kong and the study in Lanzhou observed higher serum 25(OH)D levels in men than in women [22,24]. These differences may be due to the type of clothing, working outside, and the reluctance of women to become tanned. Previous studies were also inconclusive. A metaanalysis showed higher serum 25(OH)D levels in women than in men [38], but another review focusing on the USA revealed higher vitamin D status in men than in women [39].
A negative correlation between serum 25(OH)D and age was reported in Beijing [20], which is consistent with the decrease of serum 25(OH)D levels with aging due to insufficient sunlight exposure, a decreased functional capacity of the skin, and reductions in intestinal calcium absorption and renal function. However, in Taipei no age-related changes in serum 25(OH)D in Chinese women aged 40-72 years were found [19]. Perhaps the abundant sunshine in Taiwan may have, in part, compensated for lower 25(OH)D levels, as found in younger women [40] and in the elderly. Also, the small sample size of that study may be a reason for the lack of a correlation between serum 25(OH)D levels and age.
The present study has several limitations. First, few studies were available and most of the 11 selected studies were crosssectional in design and susceptible to bias [33]. Moreover, most of the studies were not specifically conducted to examine a relationship between serum 25(OH)D concentrations and BMD, and not all studies used the same cut-off point to define vitamin D insufficiency and deficiency. Furthermore, studies included various bone sites for estimating BMD, and the methods used to measure BMD differed between the studies. Finally, comparison of vitamin D status may be hampered by the high variation of serum 25(OH)D measurement between laboratories [39,[41][42][43].
A strength of our review is that it included all studies identified by four different databases and included not only English but also Chinese language articles.
In conclusion, the evidence for a positive relationship between the serum 25(OH)D concentration and BMD in the middle-aged and older Chinese population living in Asia appears to be limited and inconsistent. Genetic factors, different lifestyle habits, including sun-avoiding behavior with the use of hats and umbrellas, the reluctance of women to become tanned, a low dietary calcium intake, and the absence of vitamin D-fortified foods may have an impact on vitamin D status and its relationship with BMD. Additional intervention studies are needed to investigate whether vitamin D supplementation can improve BMD in the Chinese population. Open Access This article is distributed under the terms of the Creative Comm ons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.