Osteoporosis is a systemic metabolic condition marked by diminished bone mass, deterioration of bone microstructure, and heightened fracture risk
[1]. Globally, osteoporosis accounts for nearly 9 million fractures annually
[2]. These fractures lead to significant disability, diminished quality of life, and heightened financial strain on healthcare systems
[3]. Sex, age, alcohol consumption, smoking, and insufficient physical activity are widely recognized as critical contributors to bone health
[4].
Cadmium (Cd), lead (Pb), mercury (Hg), selenium (Se), and manganese (Mn) are prevalent in the human environment, with exposure occurring through water, cosmetics, air, and food
[5-7]. Evidences
[8-10] suggest that these trace elements buildup in mammals is associated with various health problems, including cancer, metabolic disorders, and respiratory diseases. A recent study
[11] suggests that specific trace element may influence osteoporosis development. Elevated blood levels of Cd, Pb, and Mn, alongside reduced levels of Se and Hg, have been linked to osteoporosis risk. However, not all study results are consistent. A study
[12] based on a cohort of Chinese women didn’t find associations between blood Cd and osteoporosis, while research
[13] on adolescents showed an insignificant negative relationship between blood Hg and bone mineral density (BMD). Similarly, another analysis
[14] concluded that blood Mn concentrations were not connected to BMD in males.
The reason for the difference in results may be associated with the variety in age, areas and sample size of the population between different studies, but the key factor could be sexual difference and the focus on individual metal exposure, because different combinations of metals may have synergistic or antagonistic effects. Thus, it has an impact on the estimation of the effect of its involvement in osteoporosis. Additionally, dairy products, a common source of bone-supporting nutrients like protein,phosphorus, calcium, potassium, and vitamins A, D, B
2, B
3, and B
12[15], can help delay osteoporosis and lower fracture risk
[16]. However, whether dairy intake mitigates the negative effects of heavy metals on bone health is worth exploring.
This study investigates the association between exposure to Cd, Pb, Hg, Se, and Mn and osteoporosis using data from the National Health and Nutrition Examination Survey (NHANES) for 2013-2014 and 2017-2018. This study aims to systematically assess the sexual difference in mixed exposure to this 5 metals and interactions between metals. Additionally, this study explores the relationship between dairy intake frequency and osteoporosis. This study also aims to provide a deeper understanding of how Cd, Pb, Hg, Se, and Mn exposure relates to osteoporosis and lay the groundwork for personalized prevention and treatment strategies.
1 Materials and methods
1.1 Ethics statement
The study protocol was approved by the Research Ethics Board (REB) of the National Center for Health Statistics (NCHS), and all participants provided written informed consent.
1.2 Study population and design
NHANES, conducted by the NCHS of the Centers for Disease Control and Prevention (CDC), is a cross-sectional health survey designed for the non-institutionalized U.S. population. It uses a multi-stage, hierarchical sampling method to ensure representativeness. The survey protocol has undergone rigorous review and approval by CDC-affiliated committees, with all participants providing informed consent. This study integrated and analyzed questionnaire, laboratory, examination, demographic, and examination data from NHANES participants. Specifically, this study focused on a subset of 5 715 participants-males aged 50 and older and postmenopausal females-from the 19 429 respondents who participated in NHANES between 2013 and 2018. To maintain data integrity, we excluded respondents with incomplete sociological information, missing results for blood Cd, Pb, Hg, Se, and Mn (2 031 individuals), and those lacking bone mineral density measurements (2 418 individuals). Finally, this study involved 1 266 individuals, including 604 males and 662 women (
Figure 1).
1.3 Blood metals
The handling, preservation, and transportation of whole blood samples to the NCHS were carefully managed. Blood levels of Cd, Pb, Hg, Mn, and Se were quantified using quadrupole inductively coupled plasma mass spectrometry (ICP-MS). For measurements below the lower detection limit (LOD), values were calculated as the LOD divided by the square root of 2. According to statistical analysis standards, blood levels of Cd, Pb, Hg, Mn, and Se values are subjected to logarithmic transformation during subsequent processing of the data based on statistical descriptive results.
1.4 Osteoporosis
Osteoporosis, as per World Health Organization (WHO) guidelines, is defined as BMD values 2.5 standard deviations (SDs) or more below the mean of a young adult reference group. BMD was assessed using dual-energy X-ray absorptiometry (DXA) with a Hologic QDR-4500A fan-beam bone densitometer. The left hip’s proximal femur was the primary focus, but the right hip was scanned if the participant had implants or metal objects in the left leg. Participants were excluded if they were pregnant, weighed over 300 pounds, or had a history of radiographic contrast material use, fractures, prostheses. The study evaluated osteoporosis in the femoral neck and vertebrae.
1.5 Covariates
Demographic and lifestyle information was gathered by trained personnel in accordance with the NHANES website statement. Demographic variables included age (years), race/ethnicity, sex (male or female), educational level. Smoking status was categorized into never smoke or current smokers. Alcohol consumption was classified as <12 or ≥12 drinks per year. Body mass index (BMI; kg/m²) categories were <25, 25-30, or ≥30. Rheumatoid arthritis was defined as a physician diagnosis of rheumatoid arthritis. Hypertension was defined as a previous diagnosis or systolic blood pressure >140 mmHg (1 mmHg=0.133 kPa)/diastolic blood pressure >90 mmHg. The income to poverty ratio (PIR) was classified as <1, 1-3, or ≥3. The weekly metabolic equivalent (MET)-min for each activity was determined by multiplying its standard MET value by the total weekly minutes. The total weekly MET-min was then the sum of these values. From 2007 to 2018, PA was assessed using the Global Physical Activity Questionnaire (GPAQ) by WHO, which evaluates usual PA in three domains (work/domestic, transport/travel, leisure time) for continuous 10-minute periods. Intensity (moderate or vigorous) was considered, and participants were categorized into low (<600 MET-min/week), moderate (600-3 000 MET-min/week), and high (≥3 000 MET-min/week) PA levels, aligning with WHO’s recommendation of 150 minutes of moderate PA per week. NHANES collected data on dairy product intake through a 30-day dietary recall questionnaire, milk product consumption was categorized into <7 times/week and ≥7 times/week.
1.6 Statistical analysis
Analyses were conducted using R software (4.1.0, R Core Team). Following NHANES guidelines, mobile exam center weights were applied to represent the U.S. population. Blood metal concentrations were log-transformed for normality. Weighted mean±SD were calculated for continuous variables and analyzed using weighted t-tests or Kruskal-Wallis tests. Categorical variables were statistically described using frequencies and percentages. For unordered categorical variables, the weighted chi-square test was used to compare the differences between the 2 groups, and for ordered categorical variables, the weighted rank sum test was used to compare the differences. Survey-weighted multivariate logistic regression models evaluated the association between blood metals and osteoporosis.
Firstly, weighted generalized linear regression models were employed within a complex sampling framework to investigate the link between osteoporosis and blood Cd, Pb, Hg, Se, Mn levels, adjusting for confounding factors
[17]. Due to potential nonlinearity, both continuous and categorical variables were considered. Dominance ratios and 95% confidence interval (CI) summarized the results. Adjustments were made for age, race/ethnicity, marital status, education, PIR, BMI, physical activity, rheumatoid arthritis, hypertension, smoking, and alcohol intake. To study the associations of blood metal mixtures on osteoporosis, we employed weighted quantile sum (WQS) models and Bayesian kernel machine regression (BKMR) models. WQS calculates the multiple exposures by constructing a weighted index and assessing its relationship with outcomes. The weights indicate each chemical’s contribution, helping identify significant substances. BKMR provides flexible estimates of the multivariate exposure-response function. We determined the relationship of mixed exposure to Cd, Pb, Hg, Se and Mn on osteoporosis by estimating osteoporosis risk at the 25th to 75th percentiles (in 5th percentile increments) compared to the 50th percentile. The posterior inclusion probability (PIP) for each metal was calculated to identify the primary contributor to osteoporosis. After full covariate adjustment, the model was iterated 25 000 times. Statistical tests were two-tailed, with significance set at
P<0.05.
2 Results
2.1 Characteristics of participants and metals distribution
This study finally included 1 266 participants, with 604 (47.71%) males and 662 (52.29%) females. Among them, 553 (43.68%) were smokers and 1 107 (87.44%) were drinkers. There were 490 (38.70%) participants with rheumatoid arthritis and 752 (59.40%) with hypertension. Regarding physical activity levels, 346 (27.33%) participants had low-intensity physical activity (PA=0), 543 (42.89%) had moderate-intensity physical activity (PA=1), and 377 (29.78%) had high-intensity physical activity (PA=2) (Supplementary
Table 1,
https://doi.org/10.57760/sciencedb.xbyxb.00077).
The concentrations of Cd, Pb, Hg, Se, and Mn in the blood of male participants were (0.48±0.55) μg/L, (1.75±1.47) μg/L, (1.72±2.73) μg/L, (196.07±26.7) μg/L, and (9.29±3.13) μg/L, respectively. For female participants, the concentrations were (0.56±0.57) μg/L, (1.36±0.94) μg/L, (1.66±2.32) μg/L, (194.9±37.1) μg/L, (10.0±3.55) μg/L, respectively. There were 13 (2.15%) cases of femoral neck osteoporosis among males, and 57 (8.61%) among females; for vertebral osteoporosis, there were 39 (6.46%) cases among males and 114 (17.22%) among females. There were differences in marriage, smoking, drinking, physical activity and Cd, Pb and Mn exposure between men and women. Men were more likely to be married, smoking, drinking and taking high intensity physical activity. The number of men suffering from rheumatoid arthritis was lower, and the exposure levels of Cd and Mn were lower than women. The Pb exposure level was higher in women than in men (Supplementary
Table 1,
https://doi.org/10. 57760/sciencedb.xbyxb.00077).
Compared with men without femoral neck osteoporosis, men with femoral neck osteoporosis had higher Cd (
P=0.001) and Pb exposure (
P=0.004) and lower Se exposure (
P<0.001). Men with osteoporosis were older (
P=0.011) and had a lower proportion of overweight BMI (
P<0.001). Compared with women without femoral neck osteoporosis, women with femoral neck osteoporosis were older (
P<0.001), more likely to be non-Hispanic white (
P<0.001), less educated (
P=0.001), less likely to be overweight (
P=0.009), and more likely to have hypertension (
P=0.001; Supplementary
Table 2,
https://doi.org/10.57760/sciencedb.xbyxb.00077). Similarly, individuals with total osteoporosis were older and had lower levels of education and overweight BMI than those without the condition (Supplementary
Table 1,
https://doi.org/10. 57760/sciencedb.xbyxb.00077).
2.2 Association between blood metals and the risk of osteoporosis
Table 1 shows the results of regression analyses of the relationship between blood metal levels and osteoporosis. In males, univariate analysis showed that Cd level was significantly positively correlated with femoral neck osteoporosis [odds ratio (
OR)=4.037, 95%
CI 2.180 to 7.447,
P<0.001], Se level was significantly negatively correlated with femoral neck osteoporosis (
OR<0.001, 95%
CI 0 to 0.050,
P<0.05). Pb was also positively correlated with femoral neck osteoporosis (
OR=3.050, 95%
CI 1.842 to 5.031,
P<0.001). In multivariate analysis, adjusted Model 2 (adjusted for age, race, smoking, and drinking) showed that Cd level was significantly positively correlated with femoral neck osteoporosis (
OR=7.199, 95%
CI 2.588 to 20.022,
P<0.05), while Se was negatively correlated with femoral neck osteoporosis (
OR=0.001, 95%
CI 0 to 0.028,
P<0.05). Pb was also positively correlated with femoral neck osteoporosis (
OR=2.595, 95%
CI 1.393 to 4.835,
P<0.05). Model 3 (adjusted for race, age, smoking, marital status, alcohol consumption, education, BMI, rheumatoid arthritis, hypertension, physical activity, and annual household income) Cd levels were significantly positively correlated with femoral neck osteoporosis (
OR=8.563, 95%
CI 1.888 to 38.845,
P<0.05), while Se was still negatively correlated with femoral neck osteoporosis (
OR<0.001, 95%
CI 0 to 0.15,
P<0.05). No association was observed between metals and vertebral osteoporosis in men (all
P>0.05).
In females, univariate analysis showed that Cd levels were significantly positively correlated with femoral neck osteoporosis (OR=1.863, 95% CI 1.160 to 2.990, P<0.05). In multivariate analysis, after adjusting for age, race, smoking, and drinking, Cd remained positively associated with femoral neck osteoporosis (OR=2.610, 95% CI 1.325 to 5.142, P<0.05) and was also significantly positively associated with vertebral osteoporosis (OR=1.810, 95% CI 1.058 to 3.095, P<0.05). After further adjusting for education, marital status, BMI, rheumatoid arthritis, hypertension, physical activity, and annual household income, Cd levels were still significantly positively associated with femoral neck osteoporosis (OR=2.148, 95% CI 1.038 to 4.443, P<0.05), and Se was negatively correlated with femoral neck osteoporosis (OR=0.027, 95% CI 0.001 to 0.986, P<0.05).
2.3 WQS model to assess the correlation of blood metals co-exposure with osteoporosis
WQS model investigated the single direction association of mixed exposure of Cd, Pb, Hg, Se, and Mn with osteoporosis. The WQS model analysis confirmed the positive association of 5 blood metals with vertebral osteoporosis in men (
OR=1.452, 95%
CI 0.277 to 2.626,
P<0.05). As for the negative correlation, the model did not show statistically significant results (Supplementary Table 3,
https://doi.org/10.57760/sciencedb. xbyxb.00077). Figure
2 and
3 showed the weights of each metal during the mixed exposure. It can be seen that Cd (0.40) and Hg (0.36) play a dominant role in the relationship between the index of metal mixture and male vertebral osteoporosis.
2.4 BKMR model to assess the correlation of mixed exposure to blood metals with osteoporosis
BKMR model result shows: In males, significant positive associations were observed between exposure to these 5 metals and the risk of osteoporosis at both the vertebral and femoral neck. When all metals reached or exceeded the 55th percentile, the correlation of mixed exposure to blood metals with osteoporosis was statistically significant compared to exposure at the 50th percentile. However, there was no association observed between metal exposure and osteoporosis in females (
Figure 4).
Supplementary Table 4 (
https://doi.org/10.57760/sciencedb.xbyxb.00077) shows that, among the metal mixtures, for males, the highest PIPs for femoral neck osteoporosis were estimated for Cd (PIP=0.90) and Mn (PIP=0.84), followed by Se (PIP=0.80), Hg (PIP=0.77), and Pb (PIP=0.76). For vertebral osteoporosis, the highest PIPs were estimated for Hg (PIP=0.91), followed by Cd (PIP=0.70), Se (PIP=0.66), Mn (PIP=0.65), and Pb (PIP=0.58).For females, all metal PIP values for the femoral neck were above 0.5 (Cd: 0.58; Pb: 0.54; Hg: 0.56; Se: 0.57; Mn: 0.54), while all were below 0.5 for the vertebral.
Univariate exposure analysis (
Figure 5) showed the dose-response relationship between each metal exposure and osteoporosis when all other metals were fixed at their median values. In males, Cd and Se had a non-linear relationship with femoral neck osteoporosis, with Cd positively associated and Se negatively associated. Pb exhibited an “inverted S-shaped” curve, Mn showed a U-shaped relationship, and Cd, Pb, Hg, Se, and Mn all had non-linear relationships with vertebral osteoporosis. Specifically, Cd, Hg, and Mn were positively associated with vertebral osteoporosis, especially Hg, while Se was negatively associated. No dose-response relationship was observed in females.
Furthermore, we observed the change in osteoporosis risk and confidence interval for each interquartile range (IQR) increase in a single metal element when other metals were set at the 25th, 50th, and 75th percentiles. The results showed no correlation (Supplementary
Figure 1,
https://doi.org/10.57760/sciencedb. xbyxb.00077).
Finally, we tested the binary associations between each pair of metal mixtures and osteoporosis using the BKMR model (Supplementary
Figure 2,
https://doi.org/10. 57760/sciencedb.xbyxb.00077). When the metal mixture in the row was fixed at its 10th, 50th, and 90th percentiles, and the remaining metals were fixed at their 50th percentile, we found potential interactions between Pb and Hg on femoral neck osteoporosis in men. Obvious interactions also existed between Mn and Hg, Mn and Pb, Se and Pb, and Hg and Se. For vertebral osteoporosis, we found interactions between Cd and Pb, Cd and Se, Mn and Pb, Mn and Se, and Pb and Se. No interaction relationships were observed in women.
2.5 Regulation of milk product consumption on relationships between Cd and osteoporosis
Subgroup analyses examined the relationship between Cd exposure and osteoporosis at different consumption frequencies of milk products (
Table 2). For women with femoral neck osteoporosis, we observed a reduced
OR in the low-frequency milk consumption group (
OR=3.12, 95%
CI 1.84 to 5.30,
P<0.05) compared to the high-frequency milk consumption group (
OR=1.19, 95%
CI 0.51 to 2.78,
P>0.05). The
P-interaction value was 0.01, indicating that daily milk consumption frequency was linked with a lower prevalence of femoral neck osteoporosis in women under the Cd exposure.
3 Discussion
This study sought to determine whether there is an association between osteoporosis and the 5 metals in blood, Cd, Pb, Hg, Se, Mn, and their combinations, and whether there are sex differences in this association. Using NHANES data (2013-2014, 2017-2018), it was found that Cd was positively linked to femoral neck osteoporosis in men over 50 and postmenopausal women, while Se showed a negative association with femoral neck osteoporosis in both groups. Combined exposure to metal mixture was positively correlated with the risk of osteoporosis in different parts of the body in men, and Cd was the main contributor to the mixture. However, no association was observed in women. The positive association of mixed exposure to 5 blood metals with osteoporosis may be partially offset by Se. Consumption of dairy products once or more daily alleviated the positive relationship between Cd and femoral neck osteoporosis in women, while no significant association was found in men. This is the first study to assess sex-specific correlations of mixed exposure to Cd, Pb, Hg, Se, and Mn on osteoporosis.
Human exposure to various different environmental chemicals simultaneously can lead to interactions between them that interfere with the study of effects. People inevitably come into contact with multiple metals simultaneously, which can interact with each other. Previous epidemiological studies
[18-19] have mainly concentrated on the relationship between individual metal exposure and osteoporosis, and they examining the effects of metal exposure has predominantly relied on single-pollutant models, potentially overlooking the influence of combined exposure. Metal combinations may have synergistic or antagonistic effects, as different metals may enhance or inhibit the absorption of others
[20].
This study showed a positive correlation between Cd and femoral neck osteoporosis in men over 50 and postmenopausal women. This result aligns with previous study
[21], which has found that Cd exposure is associated with an increased risk of osteoporosis by up to 23%. An NHANES study
[11] showed a negative correlation between blood Cd levels and BMD at the total femur, femoral trochanter, femoral neck, and intertrochanteric regions, particularly in women. Interestingly, another NHANES analysis
[22] revealed blood Cd was inversely related to lumbar BMD in young females, but the opposite was true for males. The possible mechanisms between Cd and osteoporosis include Cd impairing the viability, proliferation
[23-24], stimulating osteo-clastogenesis by increasing RANKL expression
[25], and inducing osteoblast apoptosis through reactive oxygen species (ROS)
[26]. Cd exposure also interferes with parathyroid hormone, leading to bone loss
[27]. As for Cd, it has only been found to be associated with osteoporosis of the femoral neck. We think that at the current environmental exposure levels, the femoral neck is a more sensitive and earlier site for damage to occur. The impact of Cd on the vertebral may only become apparent at a higher threshold
[28].
This study showed a negative correlation between Se and femoral neck osteoporosis in men over 50 and postmenopausal women, with no observed association with vertebral osteoporosis. This aligns with a prior study
[29] that found a positive association between Se and BMD in older European men. Additionally, several studies
[30-31] have shown a negative correlation between Se and fractures caused by osteoporosis. The potential mechanisms include Se can enhance the osteogenic differentiation of BMMSCs, downregulate the differentiation and formation of mature osteoclasts
[32]. An animal experiment
[33] has also found that Se deficiency can cause changes in bone metabolism.
For other heavy metals, including Pb, Hg, and Mn, results of this study did not show a clear association with osteoporosis. There are inconsistencies with previous studies. An NHANES study
[34] reported an inverse relationship between blood Pb levels and BMD in individuals aged 50 and older. However, another study
[35] showed no correlation between blood Pb and BMD, which aligns with our findings. Additionally, NHANES data linked low blood Hg levels to an increased osteoporosis risk in young adults
[36]. A KHANES study
[37] observed a positive correlation between high blood Hg levels and osteoporosis in males, but not in females. The link between osteoporosis and blood Mn remains inconclusive. One study
[14] found no association between blood Mn levels and BMD, while another
[38] indicated that Mn exposure was associated with an increased osteoporosis risk. For mixed exposure, metals can have different health effects due to incorporation or undesirable interactions. The protective role of Se may attenuate or even counteract the negative correlation between other metals and osteoporosis, potentially explaining the lack of observed relationships in this study. Future research should investigate these metals’ roles in bone health and underlying mechanisms in depth.
The proportions and contents of metal mixtures vary significantly among different countries and regions, and the effects may differ under different proportions and dosages
[39]. Therefore, we employed the WQS and BKMR model, which is specifically designed to handle the complex interactions between chemical substances and this study is the first to use the BKMR statistical model and WQS model to explore the sex-specificity association between mixed exposure to 5 metals (Cd, Pb, Hg, Se, Mn) and osteoporosis in 2 populations (men over 50 and postmenopausal women).
This study first explored the unidirectional combined association of Cd, Pb, Hg, Se and Mn with osteoporosis using the WQS model. The results showed that the 5 blood metals had a positive correlation with vertebral osteoporosis in men (
OR=1.45, 95%
CI 0.28 to 2.63,
P<0.05). The negative WQS model did not find statistically significant results. Slightly different from previous studies
[14, 22, 34-38], we found that for both sexes, Hg and Cd were positively correlated with a higher risk of osteoporosis, while Se, Pb, and Mn were negatively correlated. This may be related to differences in the study population and weighting methods.
Then results showed that when the mixture concentration exceeds the 55th percentile, a significant positive correlation is observed in the male population, indicating that the coexistence and accumulation of multiple metals may have a synergistic effect and is associated with an increased risk of osteoporosis, also indicating a dose-response trend for the cumulative risk. Although the relative proportion may vary in other populations, we found that the risk reaches statistical significance after exceeding the median exposure level. This finding provides a crucial threshold reference standard for the public health field. Although the proportion of the mixture changes, the bone toxicity of certain metals (such as Cd) is a biological commonality across populations and regions. The analysis of PIP values further revealed the relative importance of different metals in the risk of osteoporosis. The PIP values for femoral neck osteoporosis showed the highest values for Cd, indicating that Cd is the core risk factor in femoral neck osteoporosis in both men and women. This means that as long as Cd occupies a dominant or important position in the mixture, regardless of how the ratios of other metals are adjusted, this positive correlation is robust in clinical practice. Cd has been confirmed in multiple international studies
[12, 29] to be related to bone damage. This indicates that regardless of changes in the mixture ratios, the independent pathogenicity of certain components is stable. For vertebral osteoporosis, Hg had the highest PIP value, suggesting its special role in vertebral bone loss. Se showed high PIP values in both types of osteoporosis, but its negative correlation indicates its relevance in protecting bone health. In contrast, no similar associations were observed in women, which may be related to differences in hormone levels, bone metabolism mechanisms, or lower sensitivity to metal exposure in women. Therefore, when formulating public health policies and environmental intervention measures, the issue of mixed exposure to multiple metals should be fully considered. Univariate exposure analysis revealed complex nonlinear relationships between metal exposure and osteoporosis. In men, Cd was positively correlated with both femoral neck and vertebral osteoporosis, while Se was negatively correlated, further supporting previous studies
[21, 29] on the protective effect of Se on bone health. Pb and Mn exhibited different curve relationships, indicating that they may have different relationships on bone at low and high concentrations. This nonlinear relationship suggests that when assessing the risk of heavy metal exposure, we need to consider the relationship between exposure levels more carefully. The BKMR model also revealed interactions between multiple metals and found significant interactions between Pb and Hg, Mn and Hg, and other metals, which may jointly affect bone health through complex biological mechanisms. However, no similar associations were observed in women, further emphasizing the importance of sex differences in how metal mixture exposure affects bone health. This suggests that the relationship between metal exposure and bone health is not isolated, but rather the outcome of multiple factors interacting together.
The differences between men and women in terms of behavioral exposure, lifestyle variations, and gene expression can explain these specific sex-related associations. After menopause, the level of estrogen in women decreases, and the absorption of Cd, which increases in the state of low iron reserves, will no longer occur
[40]. However, during and after menopause, due to the reduced estrogen secretion, the accelerated bone loss may significantly increase the exposure of women to heavy metals such as Cd and Pb
[41-42]. Meanwhile, epidemiological and laboratory study
[43] indicates that males have a higher heavy metal load and are more susceptible to the harmful effects of heavy metals. Regarding the differences during mixed exposure, we believe that the protective effect of Se may have counteracted the apoptosis of bone cells induced by Cd. Se can form inert Cd- Se complexes within the body, thereby reducing the bioavailable level of Cd
[44]. Se can enhance the activity of glutathione peroxidase to eliminate the reactive oxygen species produced by Cd, and promote the synthesis of metallothionein, thereby facilitating the storage of Cd
[45]. Se metabolism and selenoprotein expression exhibit sex-specific biphasicity. Females produced more selenoproteins when given the same amount of Se
[46]. Moreover, Cd exhibits a strong binding affinity for isolated estrogen receptors (ERs). A vitro study
[47] has shown that Cd can effectively block the binding of estrogen to its receptor, disrupting estrogen’s regulatory function in bone metabolism. Sex differences can impact various processes, including accumulation, antioxidant capacity, nutritional requirements, and gene expression patterns. In addition, multiple interrelated factors contribute to metal-related toxicity and modulate sex-specific alterations in absorption and metabolism
[43, 48-49].
This study innovatively explores the impact of sex differences and dairy consumption frequency. Subgroup analysis showed that daily milk consumption frequency was linked with a lower prevalence of femoral neck osteoporosis in women under the Cd exposure. Among the various factors that may affect the prevalence of osteoporosis and fractures, dietary intake is crucial
[50]. Dairy products can prevent the occurrence of osteoporosis by regulating bone metabolism (such as inhibiting the secretion of PTH)
[51]. Previous study
[52] has shown that each additional 200 grams of daily dairy consumption reduces osteoporosis risk by 22% and 37%, respectively. Previous numerous studies
[16, 53-54] have confirmed that the intake of dairy products during middle and old age is associated with a reduced risk of fractures and a decreased incidence of osteoporosis. The interaction mechanism between Cd exposure and dairy consumption frequency on femoral neck osteoporosis in postmenopausal women is not fully understood. This may be related to inflammation, oxidative stress, and protein binding. Cd exposure can increase inflammation levels and oxidative stress, while dairy intake can scavenge free radicals and reactive oxygen species (ROS) induced by Cd, reducing oxidative stress damage to bone cells and renal tubular epithelial cells; it can also increase bone mineralization and bone mass, reduce bone loss, increase intestinal absorption of IGF-1 and calcium, and competitively bind heavy metal ions to exert its effects
[55].
This study boasts several strengths. Firstly, the data were from the NHANES study, which benefits from rigorous quality control measures, including standardized data collection protocols and professionally trained and certified data collection technicians. Furthermore, we employed 2 innovative models, WQS and BKMR, to estimate the relationships of mixed exposure to Cd, Pb, Hg, Se, and Mn with osteoporosis. The WQS model identifies the contribution of individual metals within a mixture, while the BKMR model captures exposure-response relationships by holding other metals at predetermined levels. Additionally, we explored the correlations of mixed metal exposure on osteoporosis across different sexes and the interactions of dairy consumption frequency. However, this study also has its limitations. Firstly, relying on a single blood metal measurement may not adequately capture long-term exposure levels. The sampling period has elapsed for nearly 10 years, and the environmental conditions may have undergone changes. Secondly, this study is based on a single population, the interaction effects may vary under different exposure conditions. While the use of WQS and BKMR mitigates this issue to some extent. Furthermore, the milk intake data, collected via a questionnaire asking about consumption over the past 30 days, may only reflect short-term intake rather than long-term dietary habits. Questionnaire-based data collection can also introduce recall bias. Future studies should incorporate biomarkers or genetic instrumental variables (MR studies) to further clarify the contribution of dairy products to bone microstructure at different life stages, thereby eliminating biases caused by short-term behavioral changes. Lastly, as a cross-sectional study, causal links between metal mixture exposure and osteoporosis cannot be established. Future multi-center studies involving diverse populations from different geographic regions are warranted to validate these associations under varying metal mixture profiles and exposure levels.
In summary, Cd is positively linked to femoral neck osteoporosis in males over 50 and postmenopausal females, Se is negatively linked to femoral neck osteoporosis in males over 50 and postmenopausal females. When exposed to multiple metals simultaneously, joint exposure to metal mixtures is positively correlated with femoral neck osteoporosis and vertebral osteoporosis risk in males, with Cd being the primary contributor in the mixture. However, no association between joint exposure to metals and osteoporosis risk was observed in females, while Se may serve as a protective factor. Consuming dairy products once or more daily can mitigate the adverse relationship between Cd and femoral neck osteoporosis in females, while no such interactions of dairy consumption was observed in males. Future research should further validate these findings and develop effective intervention strategies to mitigate the threat of heavy metal contamination to public health.
the Science and Technology Innovation Program of Hunan Province(2025RC3061)
the Natural Science Foundation of Hunan Province(2022JJ30828)
the Grants from Hunan Provincial Clinical Medicine Research Center for Intelligent Management of Chronic Disease(2023SK4042)
the Innovation Foundation For Postgraduate of Hunan Province(CX20240310)
the Postgraduate Innovative Project of Central South University(2023XQLH183)
©Journal of Central South University (Medical Science). All rights reserved.
,
京公网安备11010202008535号
PDF
(1687KB)
