Does Higher Serum 25-Hydroxyvitamin D Levels will harm Bone Mineral Density ?: a Cross-sectional Study

doi:10.21203/rs.3.rs-4824576/v1

Download PDF

Research Article

Does Higher Serum 25-Hydroxyvitamin D Levels will harm Bone Mineral Density ?: a Cross-sectional Study

https://doi.org/10.21203/rs.3.rs-4824576/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Objective: Vitamin D has long been acknowledged as vital in preventing and managing osteoporosis. However, there is controversy over which vitamin D supplementation strategies are best for maintaining optimal bone mineral density levels for adults. This study aimed to determine the association between serum 25-hydroxyvitamin D Levels and total bone mineral density.

Methods: Data were obtained from the National Health and Nutrition Examination Survey (NHANES) 2011 to 2008 cycles. The primary exposures were 25-Hydroxyvitamin D and Bone Mineral Density. A multivariable logistic regression approach was used to model and adjust the variables relevant to the associations.

Results: There is a U-shaped association between serum 25-hydroxyvitamin D Levels and bone mineral density among adults, and the concentration of serum 25-hydroxyvitamin D level was 81.1 nmol/L. Among males, those below 84.5 nmol/L showed a significant positive association (p= 0.0012), whereas above this threshold, the association was non-significant and slightly negative (p=0.0001). For females, those below 31.4 nmol/L exhibited a significant positive association (p=0.0034), whereas above this threshold, the association remained positive but weaker and marginally significant (p= 0.0002).

Conclusions: The optimal 25-hydroxyvitamin D level for adult males is 84.5 nmol/L. Higher serum 25-hydroxyvitamin D levels are not conducive to increased bone mineral density for males. For adult females, setting the threshold for vitamin D deficiency above 31.4 nmol/L is crucial, as higher levels of 25-hydroxyvitamin D are more beneficial for increasing bone density.

U-shaped association

25-hydroxyvitamin D

bone mineral density

NHANES

osteoporosis

adult

Vitamin D, akin to calcium, has long been acknowledged as vital in preventing and managing osteoporosis. Inadequate vitamin D levels can precipitate secondary hyperparathyroidism and accelerate bone demineralisation [1]. Serum 25-hydroxyvitamin D (25[OH]D) levels are known to mitigate the risk of osteoporosis[2]. Bone mineral density (BMD) is recognised as the most important predictor of osteoporosis, and fracture is the ultimate manifestation[3]. Epidemiologic studies have shown that a 10% increase in PBM at the population level reduces the risk of fracture later in life by 50%[4].

However, observational studies have yielded varying conclusions regarding the relationship between vitamin D levels and bone mineral density (BMD)[5]. Some studies suggest that vitamin D supplementation does not prevent fractures or falls nor has a clinically meaningful effect on bone mineral density[6]. The discussion persists regarding the optimal levels of 25-hydroxyvitamin D necessary for achieving the highest possible skeletal health outcomes[7, 8]. Indeed, there does not appear to be a causal relationship between higher 25(OH)D concentrations and higher BMD in the general healthy population[9]. Furthermore, a weak relationship between serum 25(OH)D and bone microarchitecture was observed in a population with a mean serum 25-hydroxyvitamin D level as high as 137 nmol/L [10].

This inconsistency in findings highlights the complexity of the relationship between vitamin D and bone health. It suggests the possibility of a U-shaped association, where both low and high levels of serum 25-hydroxyvitamin D may be associated with adverse bone health outcomes[7, 11, 12]. Such a pattern would imply an optimal range for serum 25-hydroxyvitamin D levels, within which bone mineral density is maximised, and deviations on either side could lead to reduced bone health. This study aims to clarify these relationships by examining a large, nationally representative sample of adults in the US, using data from NHANES collected between 2011–2018.

Study design and participants

The National Health and Nutrition Examination Survey (NHANES) serves as a crucial tool for understanding the health and nutrition landscape of the United States[13]. Employing a sophisticated multistage and probabilistic sampling approach, the NHANES ensures its findings represent the diverse US population. By meticulously selecting participants and gathering comprehensive data, the NHANES provides invaluable insights into various health and nutrition trends, guiding policymakers, healthcare professionals, and researchers in formulating evidence-based strategies and interventions.

The data were obtained from the NHANES, covering 2011 to 2018, focusing on the availability of 25-hydroxyvitamin D and total bone mineral density measures. After eliminating respondents who lacked information on demographics, examination, and examination data, 11,375 adults were included in the analysis (Figure 1).

Figure 1. Flowchart of participant selection

Abbreviations :25(OH)D=25-hydroxyvitamin D. BMD=bone mineral density. NHANES =National Health and Nutrition.

Measurement of BMD

Dual-energy X-ray absorptiometry (DXA) is the most widely accepted method for measuring total bone mineral density (BMD) due in part to its speed, ease of use, and low radiation exposure[14]. Participants aged 8-59 years were eligible. The reasons for exclusion from the DXA examination were as follows: 1. Pregnancy (positive urine pregnancy test or self-reported at the time of the DXA examination);2. Self-reported radiographic contrast material (barium) use history in the past seven days; 3. Self-reported weight over 450 pounds or height over 6,5” (DXA table limitation)[15].

Measurement of 25(OH)D

The test principle for the CDC method utilised from NHANES 2011 to 2018 was liquid chromatography-tandem mass spectrometry (LC-MS/MS) for the quantitative detection of 25-hydroxyvitamin D3 (25OHD3), 3-epi-25-hydroxyvitamin D3 (epi-25OHD3), and 25-hydroxyvitamin D2 (25OHD2) in human serum. The 25-hydroxyvitamin D data analysis focused on the total 25-hydroxyvitamin D, which comprises the sum of 25-hydroxyvitamin D2 and 25-hydroxyvitamin D3 but excludes epi-25-hydroxyvitamin D3.

Other covariates

Poverty may be a barrier to obtaining adequate nutrition levels to prevent osteoporosis. A cross-sectional study has shown that individuals over the age of 50 who live below the poverty line have a higher relative risk of osteoporosis than those who live above the line[16]. A retrospective study from the UK, which analysed data from 2005 to 2015, noted that Vitamin D deficiency is associated with economic problems[17]. In this study, we incorporated the ratio of family income to poverty. Values of the ratio greater than 5.00 were recorded as 5.00.

BMI is an important indicator for measuring physical health[18]. Higher BMI is associated with low 25(OH) D levels[19] [20], while higher BMI may have a beneficial effect on BMD, masking the impact of reduced 25(OH) D levels[21]. One cross-sectional study showed a significant positive and saturated association between BMI and BMD in adolescents[22].

Smokers seem to have lower BMD values[23]. Smokers were associated with the prevalence of osteoporosis or osteopenia, and serum cotinine level, which is commonly utilised as a marker of tobacco exposure[24], has a nonlinear solid positive dose-response relationship with osteoporosis or osteopenia[25]. Clinical epidemiological studies have shown that tobacco smoking disrupts bone homeostasis and is an independent risk factor for the development of osteoporosis[23]. In our study, serum cotinine was used as an objective measure of the degree to which people were affected by tobacco use.

Alcohol intake, as well as smoking, were named as risk factors in the process of gaining peak bone mass[3]. However, the role of low-dose alcohol is uncertain because light drinkers have even higher BMD compared with abstainers[22]. A cross-sectional study showed that heavy alcohol intake may be associated with lower BMD in men[26]. A Mendelian randomisation study suggests a positive association between alcohol intake frequency and BMD, although the causal relationship was not significant[27]. Thus, alcohol intake was included as a covariate in our research. According to Alcohol Use Questionnaire, participants were divided into four groups: weekly drink, monthly drink, yearly drink, or no drink.

Physical activity (PA) and sedentary activity (SA) were associated with different effects on BMD [28, 29]. Physical activity is recommended for the management of osteoporosis by the guidelines[30]. Observational studies suggested a positive association between physical activity and bone health[31, 32]. On the contrary, the ones with sedentary activity(SA), such as bedridden households, showed a high prevalence of vitamin D deficiency without association with nutritional risk and level of dementia[33]. PA was based on the MET values of type, frequency, and duration of activities per week, which was calculated using the formula: PA (MET-min/wk) = MET × weekly frequency × duration of each PA[34]. SA was measured by counting the number of minutes per day that subjects were sedentary[35].

Statistical analysis

Baseline data for the study population are presented as the 25(OH)D subgroup, and continuous variables are presented as the means ± standard deviations and were analysed with weighted linear regression models. The “mice” package utilised the random forest algorithm for multiple interpolations of the missing data. The beta values and 95% confidence intervals CIs between 25(OH)D and BMD were calculated using multivariate linear regression analysis. We also conducted a log-likelihood ratio test comparing the one-line linear regression model with a two-piecewise linear model. Pearson's test was used to analyse the correlation between 25(OH)D and BMD (Figure 2). The statistical study used the statistical computing and graphics software R (version 4.4.0) and Empower Stats (version 4.1).

The multivariate test was built using three models: Crude Model: no variables adjusted; Model I: adjusted for gender, age and race; Model Ⅱ: adjusted for gender, age, race, education level, family income to poverty ratio, body mass index, alcohol consumption frequency, serum cotinine, physical activity, and sedentary activity. We performed tests for linear trends by entering the median value of each category of 25(OH)D as a continuous variable in the models. By changing the variables, smoothed curve fits were performed simultaneously. A threshold effects analysis model examined the relationship and saturation value between 25(OH)D and BMD. P < 0.05 was considered to indicate statistical significance. We used a weighting approach to reduce the significant volatility of our dataset.

Baseline characteristics

A total of 11,375 respondents were included in this study based on the inclusion and exclusion criteria (Figure 1), and the average age of the participants was 38.62 ± 12.23 years(Table 1). The mean (SD) concentrations of 25(OH)D and total BMD were 66.40 ± 26.10 nmol/L and 1.11 ± 0.11 g/cm², respectively. Table 1 lists the clinical features of the study participants, and column stratified grouping was based on 25(OH)D by dividing all participants equally into three groups by number. In comparison to those in the bottom tertile, those in the top tertile with higher 25(OH)D levels were more likely to be older and richer, have a greater proportion of Non-Hispanic White individuals, and have higher levels of education, and more alcohol intake, but lower levels BMI (P < 0.05). The higher 25(OH)D level groups seem to have more physical activity and less sedentary activity but with no statistical significance (P > 0.05).

Table 1. Baseline Characteristics of Participants (N =11 375)

Characteristic	25(OH)D(nmol/L)			P-value
Characteristic	Q1, ＜47.9	Q2, ≥47.9 to ＜67.9	Q3, ≥67.9	P-value
Genger, %				<0.0001
Male	52.39	53.84	46.09
Female	47.61	46.16	53.91
Age (years, mean ± SD)	36.52 ± 11.95	37.56 ± 12.08	40.53 ± 12.21	<0.0001
Race, %				<0.0001
Mexican American	18.15	13.68	4.78
Other Hispanic	8.61	9.54	5.14
Non-Hispanic White	31.94	58.6	78.88
Non-Hispanic Black	28.3	8.18	3.99
Other Race - Including Multi-Racial	13.01	9.99	7.2
Educational Level, %				<0.0001
Less than High school	17.42	15.55	10.26
High school graduate/GED or equivalent	26.04	22.34	20.85
Some college or AA degree or above	56.54	62.11	68.89
PIR,(, mean ± SD)	2.33 ± 1.58	2.74 ± 1.65	3.23 ± 1.67	<0.0001
BMI, (kg/m², mean ± SD)	30.37 ± 7.66	29.16 ± 6.66	27.53 ± 6.02	<0.0001
BMD, (g/cm², mean ± SD)	1.12 ± 0.11	1.11 ± 0.10	1.12 ± 0.11	0.0096
Serum Cotinine ,(ng/mL, mean ± SD)	64.99 ± 133.16	60.79 ± 129.95	59.20 ± 130.23	0.1791
Alcohol Consumption Frequency, %				<0.0001
Weekly Drink	29.57	33.55	41.33
Monthly Drink	24.36	25.78	23.94
Yearly Drink	19.27	18.58	17.81
No Drink	26.8	22.09	16.92
Physical Activity,(MET-min/week, mean ± SD)	72.53 ± 155.35	90.93 ± 274.16	86.84 ± 292.45	0.0156
Sedentary Activity,(min/day, mean ± SD)	395.21 ± 216.18	379.25 ± 203.64	387.23 ± 202.02	0.0089

Abbreviations: 25(OH)D=25-hydroxyvitamin D. BMD=bone mineral density. BMI=Body Mass Index. GED= General Educational Development. PIR= family income to poverty ratio. MET= Metabolic Equivalent of Task.

The Univariate analysis of BMD

This single-factor analysis table presents statistical insights into the relationships between various factors and bone mineral density (BMD). The findings revealed significant correlations between BMD and factors such as 25 (OH)D, gender, age, body mass index (BMI), race, education level, family income to poverty ratio, physical activity, and sedentary Activity (Table 2).

Table 2. The results of univariate analysis for BMD(g/cm²)

Covariate	Statistics	β (95%CI)	P-value
Genger, %
Male	5647 (49.6440%)	Reference
Female	5728 (50.3560%)	-0.0682 (-0.0719, -0.0645)	<0.000001
Age, year	37.7509 ± 12.3740	-0.0007 (-0.0008, -0.0005)	<0.000001
Race, n (%)
Mexican American	1792 (15.7538%)	Reference
Other Hispanic	1198 (10.5319%)	0.0003 (-0.0088, 0.0094)	0.946123
Non-Hispanic White	3897 (34.2593%)	0.0157 (0.0094, 0.0220)	0. 000001
Non-Hispanic Black	2351 (20.6681%)	0.0763 (0.0681, 0.0844)	<0.000001
Other Race - Including Multi-Racial	2137 (18.7868%)	0.0001 (-0.0084, 0.0086)	0. 985001
Educational Level, n (%)
Less than High school	1200 (21.25%)	Reference
High school graduate/GED or equivalent	1424 (25.22%)	0.0093 (0.0004, 0.0182)	0.039701
Some college or AA degree or above	3023 (53.53%)	0.0170 (0.0092, 0.0249)	0.000022
PIR(, mean ± SD)	2.4055 ± 1.6457	0.0033 (0.0021, 0.0045)	<0.000001
BMI(kg/m², mean ± SD)	28.7293 ± 6.9396	0.0023 (0.0020, 0.0025)	<0.000001
25(OH)D (nmol/L, mean ± SD)	60.0717 ± 24.9127	-0.0001 (-0.0002, -0.0000)	0. 033972
Serum Cotinine(ng/mL, mean ± SD)	58.6813 ± 126.6446	0.0000 (0.0000, 0.0000)	0.018797
Alcohol Consumption Frequency, %
Weekly Drink	3346 (29.4154%)	Reference
Monthly Drink	2684 (23.5956%)	-0.0072 (-0.0123, -0.0022)	0.005241
Yearly Drink	2179 (19.1560%)	-0.0153 (-0.0209, -0.0098)	<0.000001
No Drink	3166 (27.8330%)	-0.0325 (-0.0378, -0.0272)	<0.000001
Physical Activity(MET-min/week, mean ± SD)	84.6093 ± 272.5475	0.0000 (0.0000, 0.0000)	<0.000001
Sedentary Activity (min/day, mean ± SD)	371.9213 ± 203.7819	-0.0000 (-0.0000, -0.0000)	0.033240

Notably, a surge in family income to poverty ratio(PIR) and an increase in BMI were strongly positively correlated with bone mineral density (BMD), with regression coefficients of β= 0.0033 (95% CI: 0.0021, 0.0045) and β= 0.0023 (95% CI: 0.0020, 0.0025), respectively. Age was negatively correlated with BMD (β = -0.0007, 95% CI: -0.0008, -0.0005). Female showed lower BMD than male( β =-0.0682,, 95% CI: -0.0719, -0.0645) .Significant variations in BMD were detected among the different racial groups, with Non-Hispanic White and Non-Hispanic Black individuals exhibiting notably greater BMD than did the reference group (β =0.0157, 95% CI: 0.0094, 0.0220; β = 0.0763, 95% CI: 0.0681, 0.0844). Moreover, higher education levels are associated with increased BMD (β = 0.0170, 95% CI: 0.0092, 0.0249) (Table 2). More physical activity and less sedentary activity are associated with increased BMD levels.

Association between 25(OH)D and total BMD

Table 3 shows the results of the multivariate regression analysis to examine the relationship between 25(OH)D levels and total BMD (Bone Mineral Density) across three models: Crude, Model I, and Model II. 25(OH)D is negatively associated with BMD in the Crude model. After adjustment, the relationship is positive, with β values increasing from 0.0003 in Model Ⅰ to 0.0004 in Model II, all showing highly significant P-values (<0.000001). This indicates a robust positive effect of higher 25(OH)D levels on BMD across adjusted models.

Table 3. Relationship between 25(OH)D(nmol/L) and total BMD(g/cm²)

	Crude Model		Model Ⅰ		Model Ⅱ
	β (95%CI)	P-value	β (95%CI)	P-value	β (95%CI)	P-value
25(OH)D(nmol/L)	-0.0001 (-0.0002, -0.0000)	0.033972	0.0003 (0.0002, 0.0004)	<0.000001	0.0004 (0.0003, 0.0004)	<0.000001
Subgroup by 25(OH)D (nmol/L)
Q1, ＜47.9	Reference		Reference		Reference
Q2, ≥47.9 to ＜67.9	-0.0060 (-0.0113, -0.0007)	0.025534	0.0068 (0.0017, 0.0118)	0.008376	0.0079 (0.0030, 0.0129)	0.001687
Q3, ≥67.9	0.0007 (-0.0043, 0.0057)	0.791956	0.0217 (0.0166, 0.0267)	<0.000001	0.0259 (0.0208, 0.0309)	<0.000001
P value for trend	<0.001		< 0.001		<0.001
Stratified by Gender
Male	0.0003 (0.0002, 0.0004)	<0.000001	0.0006 (0.0005, 0.0007)	<0.000001	0.0007 (0.0005, 0.0008)	<0.000001
Female	-0.0001 (-0.0002, -0.0000)	0.018309	0.0001 (-0.0000, 0.0002)	0.177592	0.0001 (0.0000, 0.0002)	0.002754
Stratified by Age (year)
18-39.9	-0.0001 (-0.0002, 0.0000)	0.107554	0.0003 (0.0002, 0.0004)	< 0.000001	0.0004 (0.0003, 0.0005)	<0. 000001
40-59	-0.0000 (-0.0001, 0.0001)	0.991721	0.0003 (0.0001, 0.0004)	0.000003	0.0003 (0.0002, 0.0004)	<0. 000001
Stratified by Race
Mexican American	0.0003 (0.0000, 0.0005)	0.024763	0.0003 (0.0001, 0.0005)	0.008778	0.0005 (0.0002, 0.0007)	0.000134
Other Hispanic	0.0001 (-0.0002, 0.0004)	0.386483	0.0003 (0.0000, 0.0005)	0.044297	0.0003 (0.0001, 0.0006)	0.011908
Non-Hispanic White	0.0000 (-0.0001, 0.0002)	0.519281	0.0003 (0.0002, 0.0004)	0.000007	0.0004 (0.0002, 0.0005)	<0.000001
Non-Hispanic Black	-0.0001 (-0.0003, 0.0001)	0.419328	0.0002 (-0.0000, 0.0004)	0.088385	0.0003 (0.0001, 0.0005)	0.003132
Other Race - Including Multi-Racial	0.0002 (-0.0000, 0.0004)	0.067969	0.0004 (0.0002, 0.0006)	0.000044	0.0004 (0.0002, 0.0006)	0.003132

Data were presented as β(95%CI) P value.

Abbreviations: 25(OH)D=25-hydroxyvitamin D. BMD=Bone Mineral Density. BMI=Body Mass Index.

Model Ⅰ adjusted for gender, age and race.

Model Ⅱ adjusted for gender, age, race, education level, family income to poverty ratio, body mass index, alcohol consumption frequency, serum cotinine, physical activity, and sedentary activity.

In the subgroup analysis stratified by gender, age, race, and BMI, the model is not adjusted for gender, age, race, and BMI, respectively.

Subgroup analysis based on 25(OH)D levels further highlights the relationship. Compared to the reference group (Q1, <47.9 nmol/L), higher tertiles of 25(OH)D showed progressively stronger positive associations with BMD. In Model II, Q2 (47.9-67.9 nmol/L) had a β of 0. 0079 (P = 0.001687), Q3 (≥67.9 nmol/L) had a β of 0. 0259 (P < 0.000001). The trend analysis confirmed that increasing levels of 25(OH)D were significantly associated with higher BMD across all models, with P-values for trends all below 0.001.

The multiple regression analysis further examines the relationship between 25(OH)D levels and total BMD by stratifying the data according to gender, age, and race. Stratified analysis by gender indicates that males consistently exhibit a strong positive association between 25(OH)D levels and BMD across all models, with the strongest association observed in Model II (β = 0.0007, 95% CI: 0.0005, 0.0008). For females, a slight negative association is noted in the crude model (β = -0.0001, 95% CI: -0.0002, -0.0000), which becomes a weak and non-significant positive association in Model I, and slightly significant in Model II (β = 0.0001, 95% CI: 0.0000, 0.0002). Age stratification shows that younger adults (18-39.9 years) have a weak negative association in the crude model, which turns significantly positive in adjusted models (β = 0.0004, 95% CI: 0.0003, 0.0005). Older adults (40-59 years) display no significant association in the crude model but a significant positive association in adjusted models (β = 0.0003, 95% CI: 0.0002, 0.0004). Racial stratification reveals that Mexican Americans and non-Hispanic whites consistently demonstrate positive and significant associations in adjusted models. Non-Hispanic Blacks and other racial groups exhibit variable associations, with substantial positive results primarily in adjusted models, emphasising the importance of considering demographic factors when analysing the impact of 25(OH)D on BMD.

Nonlinearity and Saturation Effect Analysis between 25(OH)D and 25(OH)D

Figure 2 illustrates the dose-response relationship between 25-Hydroxyvitamin D (25(OH)D) and Bone Mineral Density (BMD) using a generalised additive model (GAM). Figure 2A shows the association without any adjustments, while Figure 2B depicts the relationship after total adjustment for variables such as. A nonlinear association is evident in both figures, with the solid red line representing the smooth curve fit between 25(OH)D levels and BMD. The blue bands indicate the 95% confidence intervals for the fit. The adjusted model (Figure 2B) shows a more pronounced and evident relationship, highlighting the significance of these covariates in understanding the association between 25(OH)D and BMD.

Figure 2. Dose-response relationship between 25-Hydroxyvitamin D and Bone Mineral Density. (A: no adjustment, B: full adjustment, C: full adjustment, stratified by gender )

A nonlinear association between 25-Hydroxyvitamin D and Bone Mineral Density in a generalised additive model (GAM). The solid red line represents the smooth curve fit between variables. Blue bands represent the 95% confidence interval from the fit.

Total adjustment was made for gender, age, race, education level, family income to poverty ratio, body mass index, alcohol consumption frequency, serum cotinine, physical activity, and sedentary activity.

Table 4 shows the threshold effect analysis of 25-hydroxyvitamin D (25(OH)D) on bone mineral density (BMD) using piece-wise linear regression. In the crude model, 25(OH)D levels below 96.4 nmol/L were significantly positively associated with BMD (β = 0.0004, 95% CI: 0.0003, 0.0006). In the adjusted model, a significant positive association is observed for 25(OH)D levels below 81.1 nmol/L (β = 0.0010, 95% CI: 0.0008, 0.0012), but levels at or above 81.1 nmol/L also show a non-significant negative association (β = -0.0002, 95% CI: -0.0005, 0.000). Likelihood ratio tests indicate significant threshold effects for both models (p = 0.003 and p <0.001, respectively).

Table 4. Threshold Effect Analysis of 25(OH)D and BMD using Piece-wise Linear Regression.

Crude model			Adjusted model
25(OH)D, (nmol/L)	β(95% CI)	P value	25(OH)D, (nmol/L)	β(95% CI)	P value
< 81	0.0002 (0.0001, 0.0003)	0.0027	< 81.1	0.0007 (0.0006, 0.0008)	<0.0001
≥81	-0.0005 (-0.0006, -0.0003)	<0.0001	≥81.1	-0.0001 (-0.0003, 0.0000)	0.1182
Likelihood Ratio test		<0.001			<0.001

Data were presented as β(95%CI) P value.

Abbreviations :25(OH)D=25-hydroxyvitamin D. BMD=bone mineral density.

The adjusted model is adjusted for gender, age, race, education level, family income to poverty ratio, body mass index, alcohol consumption frequency, serum cotinine, physical activity, and sedentary activity.

Table 3 shows that Stratified analysis revealed that females are unique when studying the relationship between 25(OH)D and BMD. Then, Figure 2C illustrates the nonlinear relationship between 25(OH)D levels and BMD, stratified by gender. The red points represent males, showing a rise in BMD up to a certain level of 25(OH)D, followed by a plateau and slight decline. The blue points represent females, showing a more gradual increase in BMD with increasing 25(OH)D levels, with less pronounced changes than males. The separation of the two curves highlights the different patterns in the 25(OH)D and BMD relationship between males and females.

Table 5 shows the threshold effect analysis of 25(OH)D and BMD using piece-wise linear regression stratified by gender. For males below 84.5 nmol/L, there is a significant positive association (β = 0.0010, 95% CI: 0.0008, 0.0012), while above this threshold, the association is non-significant and slightly negative (β = -0.0002, 95% CI: -0.0005, 0.0001). For females below 31.4 nmol/L, there is a significant positive association (β = 0.0021, 95% CI: 0.0009, 0.0034), and above this threshold, the association remains positive but weaker and slightly significant (β = 0.0001, 95% CI: 0.0000, 0.0002). The likelihood ratio test confirms the statistical significance of these threshold effects for both males (P < 0.001) and females (P = 0.002).

Table 5. Threshold Effect Analysis of 25(OH)D and BMD using Piece-wise Linear Regression(stratified by gender).

Male			Female
25(OH)D, (nmol/L)	β(95% CI)	P value	25(OH)D, (nmol/L)	β(95% CI)	P value
< 84.5	0.0010 (0.0008, 0.0012)	<0.0001	< 31.4	0.0021 (0.0009, 0.0034)	0.0007
≥84.5	-0.0002 (-0.0005, 0.0001)	0.2678	≥31.4	0.0001 (0.0000, 0.0002)	0.0474
Likelihood Ratio test		<0.001			0.002

Data were presented as β(95%CI) P value.

Abbreviations :25(OH)D=25-hydroxyvitamin D. BMD=bone mineral density.

This study investigated whether serum 25-hydroxyvitamin D Levels are potentially associated with total bone mineral density in adult US adults. To clarify this association, we utilised multivariate regression and nonlinearity analyses to analyse the data from the NHANES. Collectively, our data demonstrate a U-shaped association between serum 25-hydroxyvitamin D Levels and bone mineral density in an adult US population.

As shown in Fig.2, we found that the concentration of serum 25(OH)D associated with the highest bone mineral density was 81.1 nmol/L after total adjustment (Table 4) through threshold effect analysis, which following previous reports (Wang et al.[36], 2023; Nakamura et al.[37], 2008; Swanson et al.[38]2015). However, unlike our study, a prior trial did not detect any significant correlation between serum 25(OH)D levels and BMD at any anatomical site[5]. The discrepancies between studies may be attributed to differences in sample sizes, which affect the statistical power, and variations in analytical methods. Another study found no association between serum 25(OH)D levels and BMD in postmenopausal women [39]. These conflicting results could stem from differences in study populations or the failure to account for important confounding factors such as physical activity[30] and socioeconomic status [17]. Addressing these confounders is crucial for accurately assessing the relationship between serum 25(OH)D levels and BMD.

As shown in Table 4, there was a notable correlation: bone mineral density decreased as serum 25(OH)D levels increased (β, -0.0001; 95% CI, (-0.0003, 0.0000) in participants with serum 25(OH)D levels ≥ 81.1 nmol/L. This finding aligns with prior studies (Burt et al.[2], 2019). Specifically, research has indicated that tibial bone mineral density is markedly lower in individuals treated with a daily dose of 10,000 IU of vitamin D[2]. This phenomenon may be related to an observed increase in the plasma marker of bone resorption (CTx) and suppression of parathyroid hormone (PTH) associated with high doses of vitamin D intake[40]. Research has shown that intact PTH (iPTH) levels did not decrease rapidly when serum 25(OH)D levels were ≥ 46.25 nmol/L (18.9 ng/mL) [41]. A study revealed that segmented regression identified two critical 25(OH)D thresholds: at 55 nmol/L, PTH levels start to plateau, while at 22 nmol/L, PTH levels rise sharply[42]. It is noteworthy that circulating 25(OH)D concentrations are nearly 1000-fold higher than 1,25(OH)2D[43]; hence, 25(OH)D may also accumulate in parathyroid cells in concentrations sufficient to activate vitamin D receptors (VDR) and thus directly inhibit PTH mRNA synthesis[44].In an animal study, administering 1α,25(OH)2D3 (5 μg/kg body weight/day) to wild-type mice for four days increased the number of osteoclasts in bone and serum concentrations of C-terminal crosslinked telopeptide of type I collagen (CTX-I, a bone resorption marker) [45]. This study also demonstrated that the pro-resorptive, hypercalcemic, and toxic actions of high-dose 1α,25(OH)2D3 are mediated by vitamin D receptors (VDR) in osteoblast-lineage cells [45].

Table 5 suggests that when aiming to improve BMD, we should use different 25(OH) D-level strategies for males and females. For males, the significant positive association below 84.5 nmol/L suggests that increasing 25(OH)D levels within this range could benefit bone health. However, the lack of substantial association above this threshold indicates a potential plateau effect, where further increases in 25(OH)D may not yield additional BMD benefits. Maintaining 25(OH)D levels above 84.5 nmol/L for adult males may adversely affect bone density. The significant positive association below 31.4 nmol/L for females suggests that small increases in 25(OH)D could substantially impact bone health. In contrast, the weaker positive association above this threshold suggests diminishing returns. These findings underline the importance of maintaining adequate 25(OH)D levels, particularly in populations at risk for deficiency, and suggest that supplementation strategies might need to be gender-specific to optimise bone health outcomes. For adult females, it is necessary to set the threshold for vitamin D deficiency above 31.4 nmol/L. The higher the 25-hydroxyvitamin D levels, the more beneficial it is for increasing bone density for adult females.

Vitamin D deficiency is a global issue that has led to increased use of vitamin D supplements[17, 46]. According to SUCRA's findings, daily vitamin D supplementation ranked higher than intermittent supplementation when the same doses were. However, the daily and intermittent supplementation groups observed no statistically significant differences in 25(OH)D concentrations[47]. Vitamin D is available in 2 distinct forms: vitamin D2 and D3. Studies have indicated that vitamin D3 is superior to vitamin D2 in raising 25(OH)D concentrations [48]. The standard for vitamin D intake was 528.4 IU/day (~13.21 μg/day) to achieve the optimal serum 25(OH) D in Korean Postmenopausal Women [41]. A Randomized Clinical Trial investigating the effect of high-dose Vitamin D supplementation found that suitable daily doses of vitamin D3 intake range from 400 IU/day to 4000 IU/day[2]. Furthermore, A study involving 116 randomised controlled trials with 11,376 participants used network meta-analysis to evaluate the effectiveness of various vitamin D supplementation regimens. The analysis concluded that daily supplementation of 2,000 IU of vitamin D was the most effective in achieving optimal 25(OH)D concentrations (>75 nmol/L)[47]. Our research can provide a reference for selecting 25(OH)2D levels.

For the first time, our study showed a dose-response relationship between serum 25-hydroxyvitamin D levels and total bone mineral density in male and female adults. This finding will help explain why a higher dose of vitamin D intake might not be a favourable choice for attaining higher bone quality. Recommendations from agencies that assess the safety of a nutrient have set a tolerable upper limit of 4,000 IU daily[7]. The optimal concentration for maintaining bone health was identified at 81.1 nmol/L. This critical finding informs the optimal vitamin D levels necessary to maintain bone health and can guide dietary and supplementation guidelines to prevent and manage osteoporosis. To avoid excessive vitamin D intake in males, the criteria for determining vitamin D deficiency in adult women were referenced. These criteria can help establish safe and effective supplementation practices for both genders.

Our study investigated the association between serum 25-hydroxyvitamin D levels and bone mineral density (BMD) in a sample of 11,375 adults using NHANES data. A U-shaped relationship was found, with the highest BMD observed at a serum 25-hydroxyvitamin D concentration of 81.1 nmol/L. This association was significant even after adjusting for various covariates. The results highlight the complexity of vitamin D's impact on bone health, indicating that low and high serum levels of 25-hydroxyvitamin D can adversely affect BMD. The study provides valuable insights into optimal vitamin D levels for bone health, with implications for dietary and supplementation guidelines.

However, our study also had some limitations. First, as this is a cross-sectional study of disease prevalence rather than incidence, there is uncertainty in the degree to which exposure preceded the outcomes observed. Second, the NHANES database only assessed total bone mineral density in participants aged 8-59 years, leaving the impact of serum 25-hydroxyvitamin D on older inconclusive. Third, although we tried to adjust for numerous confounding factors, unmeasured confounders remain a possibility, and some covariates, such as alcohol consumption frequency, physical activity, and sedentary activity, were investigated based on the participant questionnaire survey.

Our study demonstrated a clear association between serum 25-hydroxyvitamin D levels and total bone mineral density among the adult US population. We also revealed a significant U-shaped relationship, indicating that the serum 25(OH)D concentration associated with the highest bone mineral density was 81.1 nmol/L. The optimal 25-hydroxyvitamin D level for adult males is 84.5 nmol/L. For adult females, setting the threshold for vitamin D deficiency above 31.4 nmol/L is crucial, as higher levels of 25-hydroxyvitamin D are more beneficial for increasing bone density. Moreover, they provide clinicians and researchers with a critical reference point for designing and implementing targeted interventions to optimise bone health outcomes in this population. Our study advances our understanding of the interplay between vitamin D status and bone health, paving the way for more effective strategies to combat osteoporosis and related conditions.

Acknowledgment

Not applicable.

Ethics statement: The NHANES protocols used were approved by the Research Ethics Review Board of the National Center for Health Statistics, and written informed consent was obtained from each participant.

Author contributions: Conceptualization, Bingcheng Xu and Hao Liu; methodology, Bingcheng Xu; software, Qiai Li; validation, Bo Luo; formal analysis, Qiai Li; investigation, Bingcheng Xu; data curation, Bingcheng Xu; writing—original draft preparation, Bingcheng Xu and Qiai Li; writing—review and editing, Hao Liu; supervision, Bo Luo; project administration, Hao Liu; funding acquisition, Hao Liu. and S.Y.L. All authors have read and agreed to the published version of the manuscript.

Funding: This research was supported by grants from the Scientific Research Project of the Wuhan Municipal Health Commission (WX23Z07)

Informed Consent Statement: Informed consent was obtained from all participants involved in the study.

Data availability statement: Publicly available datasets were analysed in this study. The survey data can be found at Http://www.cdc.gov/nchs/nhanes/.

Conflict of interest: The authors declare that the research was conducted without commercial or financial relationships that could create a conflict of interest.

Bargagli M, Arena M, Naticchia A, Gambaro G, Mazzaferro S, Fuster D, et al. The Role of Diet in Bone and Mineral Metabolism and Secondary Hyperparathyroidism. Nutrients. 2021;13:2328.
Burt LA, Billington EO, Rose MS, Raymond DA, Hanley DA, Boyd SK. Effect of High-Dose Vitamin D Supplementation on Volumetric Bone Density and Bone Strength: A Randomized Clinical Trial. JAMA. 2019;322:736–45.
Zhu X, Zheng H. Factors influencing peak bone mass gain. Front Med. 2021;15:53–69.
Gordon CM, Zemel BS, Wren TAL, Leonard MB, Bachrach LK, Rauch F, et al. The Determinants of Peak Bone Mass. J Pediatr. 2017;180:261–9.
Marwaha RK, Tandon N, Garg MK, Kanwar R, Narang A, Sastry A, et al. Bone health in healthy Indian population aged 50 years and above. Osteoporos Int. 2011;22:2829–36.
Bolland MJ, Grey A, Avenell A. Effects of vitamin D supplementation on musculoskeletal health: a systematic review, meta-analysis, and trial sequential analysis. Lancet Diabetes Endocrinol. 2018;6:847–58.
Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, et al. The 2011 Report on Dietary Reference Intakes for Calcium and Vitamin D from the Institute of Medicine: What Clinicians Need to Know. J Clin Endocrinol Metab. 2011;96:53–8.
Dawson-Hughes B, Heaney RP, Holick MF, Lips P, Meunier PJ, Vieth R. Estimates of optimal vitamin D status. Osteoporos Int J Establ Result Coop Eur Found Osteoporos Natl Osteoporos Found USA. 2005;16:713–6.
Larsson SC, Melhus H, Michaëlsson K. Circulating Serum 25-Hydroxyvitamin D Levels and Bone Mineral Density: Mendelian Randomization Study. J Bone Min Res. 2018;33:840–4.
Boyd SK, Burt LA, Sevick LK, Hanley DA. The relationship between serum 25(OH)D and bone density and microarchitecture as measured by HR-pQCT. Osteoporos Int. 2015;26:2375–80.
Anagnostis P, Bosdou JK, Kenanidis E, Potoupnis M, Tsiridis E, Goulis DG. Vitamin D supplementation and fracture risk: Evidence for a U-shaped effect. Maturitas. 2020;141:63–70.
Grant WB, Karras SN, Bischoff-Ferrari HA, Annweiler C, Boucher BJ, Juzeniene A, et al. Do studies reporting ‘U’-shaped serum 25-hydroxyvitamin D–health outcome relationships reflect adverse effects? Dermatoendocrinol. 2016;8:e1187349.
Taylor CL, Madans JH, Chapman NN, Woteki CE, Briefel RR, Dwyer JT, et al. Critical data at the crossroads: the National Health and Nutrition Examination Survey faces growing challenges. Am J Clin Nutr. 2023;117:847–58.
Messina C, Albano D, Gitto S, Tofanelli L, Bazzocchi A, Ulivieri FM, et al. Body composition with dual energy X-ray absorptiometry: from basics to new tools. Quant Imaging Med Surg. 2020;10:1687–98.
Yan Z, Xiong X, Tao J, Wang S. Association of bone mineral density with trichlorophenol: a population-based study. BMC Musculoskelet Disord. 2023;24:202.
Marshall K, Teo L, Shanahan C, Legette L, Mitmesser SH. Inadequate calcium and vitamin D intake and osteoporosis risk in older Americans living in poverty with food insecurities. PLoS ONE. 2020;15:e0235042.
Crowe FL, Jolly K, MacArthur C, Manaseki-Holland S, Gittoes N, Hewison M et al. Trends in the incidence of testing for vitamin D deficiency in primary care in the UK: a retrospective analysis of The Health Improvement Network (THIN), 2005–2015. BMJ Open. 2019;9:e028355.
Nuttall FQ. Body Mass Index: Obesity, BMI, and Health A Critical Review. Nutr Today. 2015;50:117–28.
Powe CE, Evans MK, Wenger J, Zonderman AB, Berg AH, Nalls M, et al. Vitamin D-binding protein and vitamin D status of black Americans and white Americans. N Engl J Med. 2013;369:1991–2000.
Walsh JS, Evans AL, Bowles S, Naylor KE, Jones KS, Schoenmakers I, et al. Free 25-hydroxyvitamin D is low in obesity, but there are no adverse associations with bone health. Am J Clin Nutr. 2016;103:1465–71.
Yuan C, Wang J, Zhang W, Yi H, Shu B, Li C et al. Effects of obesity with reduced 25(OH)D levels on bone health in elderly Chinese people: a nationwide cross-sectional study. Front Immunol. 2023;14.
Godos J, Giampieri F, Chisari E, Micek A, Paladino N, Forbes-Hernández TY, et al. Alcohol Consumption, Bone Mineral Density, and Risk of Osteoporotic Fractures: A Dose-Response Meta-Analysis. Int J Environ Res Public Health. 2022;19:1515.
Jing Z, Li Y, Zhang H, Chen T, Yu J, Xu X, et al. Tobacco toxins induce osteoporosis through ferroptosis. Redox Biol. 2023;67:102922.
Merianos AL, Stone TM, Jandarov RA, Mahabee-Gittens EM, Choi K. Sources of Tobacco Smoke Exposure and Their Associations With Serum Cotinine Levels Among US Children and Adolescents. Nicotine Tob Res Off J Soc Res Nicotine Tob. 2023;25:1004–13.
Hou W, Chen S, Zhu C, Gu Y, Zhu L, Zhou Z. Associations between smoke exposure and osteoporosis or osteopenia in a US NHANES population of elderly individuals. Front Endocrinol. 2023;14:1074574.
Cho Y, Choi S, Kim K, Lee G, Park SM. Association between alcohol consumption and bone mineral density in elderly Korean men and women. Arch Osteoporos. 2018;13:46.
Lai B, Jiang H, Gao R, Zhou X. Association between alcohol intake and bone mineral density: results from the NHANES 2005–2020 and two-sample Mendelian randomization. Arch Osteoporos. 2024;19:21.
Lin Z, Shi G, Liao X, Huang J, Yu M, Liu W et al. Correlation between sedentary activity, physical activity and bone mineral density and fat in America: National Health and Nutrition Examination Survey, 2011–2018. Sci Rep. 2023;13:10054.
Kim YA, Lee Y, Lee JH, Seo JH. Effects of physical activity on bone mineral density in older adults: Korea National Health and Nutrition Examination Survey, 2008–2011. Arch Osteoporos. 2019;14:103.
Brooke-Wavell K, Skelton DA, Barker KL, Clark EM, De Biase S, Arnold S, et al. Strong, steady and straight: UK consensus statement on physical activity and exercise for osteoporosis. Br J Sports Med. 2022;56:837–46.
Pinheiro MB, Oliveira J, Bauman A, Fairhall N, Kwok W, Sherrington C. Evidence on physical activity and osteoporosis prevention for people aged 65 + years: a systematic review to inform the WHO guidelines on physical activity and sedentary behaviour. Int J Behav Nutr Phys Act. 2020;17:150.
Nilsson M, Ohlsson C, Mellström D, Lorentzon M. Previous sport activity during childhood and adolescence is associated with increased cortical bone size in young adult men. J Bone Min Res Off J Am Soc Bone Min Res. 2009;24:125–33.
Leite NP, Alvarez TS, Fonseca FLA, Hix S, Sarni ROS. Vitamin D deficiency in bedridden elderly people at home. Rev Assoc Médica Bras. 2023;69:61–5.
Chen L, Cai M, Li H, Wang X, Tian F, Wu Y, et al. Risk/benefit tradeoff of habitual physical activity and air pollution on chronic pulmonary obstructive disease: findings from a large prospective cohort study. BMC Med. 2022;20:70.
Cao C, Friedenreich CM, Yang L. Association of Daily Sitting Time and Leisure-Time Physical Activity With Survival Among US Cancer Survivors. JAMA Oncol. 2022;8:395–403.
Wang D, Yang Y, Nakamura, et al. 13, 2008Hydroxyvitamin D Levels and Osteoporosis in Postmenopausal Women. Clin Interv Aging. 2023;18:619–27.
Nakamura K, Tsugawa N, Saito T, Ishikawa M, Tsuchiya Y, Hyodo K, et al. Vitamin D status, bone mass, and bone metabolism in home-dwelling postmenopausal Japanese women: Yokogoshi Study. Bone. 2008;42:271–7.
Swanson CM, Srikanth P, Lee CG, Cummings SR, Jans I, Cauley JA, et al. Associations of 25-Hydroxyvitamin D and 1,25-Dihydroxyvitamin D With Bone Mineral Density, Bone Mineral Density Change, and Incident Nonvertebral Fracture. J Bone Min Res. 2015;30:1403–13.
Hosseinpanah F, Rambod M, Hossein-nejad A, Larijani B, Azizi F. Association between vitamin D and bone mineral density in Iranian postmenopausal women. J Bone Min Metab. 2008;26:86–92.
Grimnes G, Joakimsen R, Figenschau Y, Torjesen PA, Almås B, Jorde R. The effect of high-dose vitamin D on bone mineral density and bone turnover markers in postmenopausal women with low bone mass—a randomized controlled 1-year trial. Osteoporos Int. 2012;23:201–11.
Shin HR, Lee YJ, Ly SY. Optimal Serum 25(OH)D Levels and Vitamin D Intake Comparison of the Effect of Daily Vitamin D2 and Vitamin D3 Supplementation on Serum 25-Hydroxyvitamin D Concentration. Nutrients. 2023;15:1856.
Gong M, Wang K, Sun H, Wang K, Zhou Y, Cong Y, et al. Threshold of 25(OH)D and consequently adjusted parathyroid hormone reference intervals: data mining for relationship between vitamin D and parathyroid hormone. J Endocrinol Invest. 2023;46:2067–77.
Tsuprykov O, Chen X, Hocher C-F, Skoblo R. Lianghong Yin null, Hocher B. Why should we measure free 25(OH) vitamin D? J Steroid Biochem Mol Biol. 2018;180:87–104.
Chen X, Chu C, Doebis C, Xiong Y, Cao Y, Krämer BK, et al. Vitamin D status and its association with parathyroid hormone in 23,134 outpatients. J Steroid Biochem Mol Biol. 2022;220:106101.
Mori T, Horibe K, Koide M, Uehara S, Yamamoto Y, Kato S, et al. The Vitamin D Receptor in Osteoblast-Lineage Cells Is Essential for the Proresorptive Activity of 1α,25(OH)2D3 In Vivo. Endocrinology. 2020;161:bqaa178.
Cashman KD, Dowling KG, Škrabáková Z, Gonzalez-Gross M, Valtueña J, De Henauw S, et al. Vitamin D deficiency in Europe: pandemic?12. Am J Clin Nutr. 2016;103:1033–44.
Zhuang Y, Zhu Z, Chi P, Zhou H, Peng Z, Cheng H, et al. Efficacy of intermittent versus daily vitamin D supplementation on improving circulating 25(OH)D concentration: a Bayesian network meta-analysis of randomized controlled trials. Front Nutr. 2023;10:1168115.
Van Den EGhm H, Lips P, Schoonmade LJ, Lanham-New SA, Van Schoor NM. Comparison of the Effect of Daily Vitamin D2 and Vitamin D3 Supplementation on Serum 25-Hydroxyvitamin D Concentration (Total 25(OH)D, 25(OH)D2, and 25(OH)D3) and Importance of Body Mass Index: A Systematic Review and Meta-Analysis. Adv Nutr. 2024;15:100133.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
06 Sep, 2024
Reviews received at journal
03 Sep, 2024
Reviews received at journal
01 Sep, 2024
Reviewers agreed at journal
27 Aug, 2024
Reviewers agreed at journal
25 Aug, 2024
Reviewers invited by journal
08 Aug, 2024
Editor invited by journal
05 Aug, 2024
Editor assigned by journal
05 Aug, 2024
Submission checks completed at journal
31 Jul, 2024
First submitted to journal
29 Jul, 2024

You are reading this latest preprint version

Does Higher Serum 25-Hydroxyvitamin D Levels will harm Bone Mineral Density ?: a Cross-sectional Study

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1