https://orcid.org/0009-0007-5047-5628
Figures
Abstract
Objective
Accumulating evidence showed that exposure to heavy metals was harmful to human health. Little is known regarding the mixing effects of multiple metal exposures on vertebral compression fracture (VCF) and femoral neck bone mineral density (BMD). This study aimed to explore the individual and joint effects of four heavy metals [manganese (Mn), lead (Pb), cadmium (Cd) and mercury (Hg)] on VCF risk and femoral neck BMD.
Methods
This cross-sectional study included 1,007 eligible individuals with vertebral fractures from National Health and Nutrition Examination Survey 2013–2014. The outcome was the risk of VCF and femoral neck BMD. Weighted multivariate logistic regression was used to explore the individual effect of four heavy metals on the VCF risk, separately. Weighted multivariate linear regression was used to explore the individual effect of four heavy metals on the femoral neck BMD, separately. Adopted bayesian kernel machine regression (BKMR) model and quantile-based g computation (qgcomp) to examine the joint effects of four heavy metals on the VCF risk and femoral neck BMD.
Results
Among the population, 57 individuals developed VCF. After adjusting covariates, we found no statistical differences regarding the individual effects of four heavy metals on the risk of VCF. BKMR model and qgcomp indicated that there were no statistical differences regarding the joint effects between four heavy metals on the VCF risk. In addition, we found that Cd was associated with femoral neck BMD, and an increase in the mixture of heavy metal exposures was associated with a decreased risk of femoral neck BMD.
Citation: Yang X, Li L, Nie L (2024) Associations between co‑exposure to heavy metals and vertebral compression fracture, as well as femoral neck bone mineral density: A cross-sectional study from NHANES data. PLoS ONE 19(5): e0303418. https://doi.org/10.1371/journal.pone.0303418
Editor: Ahmed Mancy Mosa, Al Mansour University College-Baghdad-Iraq, IRAQ
Received: November 17, 2023; Accepted: April 6, 2024; Published: May 22, 2024
Copyright: © 2024 Yang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The datasets used and/or analyzed during the current study are available from NHANES database, https://www.cdc.gov/nchs/nhanes/index.htm.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Vertebral compression fracture (VCF) is a type of fracture in middle-aged and elderly population, and its incidence increases with age [1]. VCF was associated with the risk of osteoporosis, affecting 700,000 cases each year in the United States [2]. VCF causes chronic back and neck pain, reduced lung volumes, impaired quality of life and increased mortality [2, 3]. Previous studies have indicated that VCF may increase the risk of subsequent vertebral and non-vertebral fractures [4, 5]. Therefore, early implementation of appropriate prevention is important to reduce the risk of VCF, improve quality of life and increase life expectancy.
Heavy metals are commonly found in the surrounding air, soil, food and water [6]. Chronic exposures to the heavy metals may affect human health, such as female reproductive health [7], bone mineral density (BMD) loss [8], and cancer [9]. A meta-analysis conducted by Cheng X et al., reported that high cadmium (Cd) exposure may constitute a potential risk factor for fractures of any nature [10]. Wang C, et al., found that people exposed to manganese (Mn) may be a risk factor of osteopenia in a cross-sectional study with 9,732 subjects [11]. Most existing studies have only focused on the relationship between individual metal exposures and bone health [10, 11]. However, heavy metals are often co-present in the environment, and people are more likely to be exposed to multiple exposures at the same time [12]. Nowadays, a study with 2,545 United States adults suggested that multiple metals exposure was associated with reduced BMD by using weighted quantile sum and bayesian kernel machine regression (BKMR) models [13]. To the best of our knowledge, little is known about the mixing effects of multiple metal exposures on VCF risk or femoral neck BMD.
Herein, in the present study, we aimed to explore the individual and joint effects of four heavy metals [Mn, lead (Pb), Cd and mercury (Hg)] on the VCF risk and femoral neck BMD. The findings may provide new evidence for the link between heavy metal exposure and VCF risk/ femoral neck BMD.
Methods
Study population
In this cross-sectional study, we analyzed data from National Health and Nutrition Examination Survey (NHANES) 2013–2014, the study year in which vertebral fractures were only assessed in participants aged ≥40 years. NHANES database is a representative cross-sectional survey of non-institutionalized U.S. population. The survey examines approximately 5,000 people each year. These people are located in counties throughout the country, and 15 of which are visited annually [14]. (https://www.cdc.gov/nchs/nhanes/about_nhanes.htm). The requirement of ethical approval for this was waived by the Institutional Review Board of Jiangle General Hospital of FuJian Province, because the data was accessed from NHANES (a publicly available database). The need for written informed consent was waived by the Institutional Review Board of Jiangle General Hospital of FuJian Province due to retrospective nature of the study.
3,224 individuals aged ≥40 years who had the evaluable vertebral fracture assessment (VFA) were included [14] (https://wwwn.cdc.gov/Nchs/Nhanes/2013-2014/DXXVFA_H.htm#DXXVFAST). Then, we further excluded some individuals who (1) had missing metal information, (2) combined with malignant tumors, (3) had missing possible covariates, such as waist circumference, body mass index (BMI) and poverty-to-income ratio (PIR). Finally, 1,007 eligible individuals were included in the current study. Fig 1 shows the process of participants selection.
Exposures: Measurement of heavy metal
For individuals enrolled in the NHANES database, whole blood specimens were collected. The concentrations of Mn, Pb, Cd and Hg in whole blood samples were measured directly by using mass spectrometry. The lower limit of detection (LLOD) for Mn, Pb, Cd and Hg was 0.99 μg/L, 0.07 ug/dL, 0.10 μg/L and 0.28 μg/L, respectively. For values below LOD, it would be calculated as the LLOD divided by the square root of 2.
(https://wwwn.cdc.gov/Nchs/Nhanes/2013-2014/PBCD_H.htm#Analytic_Notes).
Outcomes: Assessment of vertebral compression fracture and femoral neck BMD
The primary outcome of this study was whether VCF occurred. For NHANES individuals, VFA is obtained by utilizing dual energy x-ray absorptiometry (DXA) to conduct a lateral scan of the thoraco-lumbar spine. The DXA examinations were conducted by trained and certified radiology technologists. All scans were analyzed by Optasia Spinalizer software using Genant’s semiquantitative technique. Each vertebra was graded as normal (Grade 0), mild (Grade 1), moderate (Grade 2), or severe (Grade 3) fracture [15]. Grade 1 and above were diagnosed as VCF [16, 17]. (https://wwwn.cdc.gov/Nchs/Nhanes/2013-2014/DXXVFA_H.htm). The secondary outcome was femoral neck BMD. Femoral neck BMD was measured by DXA. The DXA examinations were conducted by certified radiologic technologists using Hologic QDR-4500A fan-beam densitometers (Hologic; Bedford, MA), and the data analysis was performed utilizing the Hologic APEX software, version 4.0. (https://wwwn.cdc.gov/Nchs/Nhanes/2013-2014/DXXFEM_H.htm).
Data collection
We extracted some data of all eligible individuals from the NHANES 2013–2014: age, gender, race/ethnicity, educational level, marital status, waist circumference (cm), BMI (kg/m2), PIR, smoking, drinking, physical activity [metabolic equivalent (MET)·min/week], previous fracture, parental fracture, history of glucocorticoid use, history of anti-osteoporosis medication use, diabetes, hypertension, total femur BMD (g/cm2), femoral neck BMD (g/cm2), menopause status, total energy (kcal), total calcium intake (mg), total vitamin D intake (mcg), Mn (ug/L), Pb (ug/dL), Cd (ug/L) and Hg (ug/L). Physical activity was considered as energy consumption, which was calculated as the number of minutes of activity per week multiplied by the MET score for each activity and summed [18]. The intake of total energy/total calcium intake/total vitamin D intake was calculated by adding up the dietary and supplemental intake in this study.
Statistical analysis
In descriptive analysis, continuous variables were described by mean and standard error (S.E), and weighted t-test was used for comparison between two groups. Categorical variables were described by the number of cases and constituent ratio [n (%)], and χ2 test was performed to compare the difference between groups. In the weighted univariate logistic regression analysis, we screened some possible confounding factors related to VCF (P<0.05, S1 Table). A weighted univariate linear regression model was established to identify the confounding factors potentially associated with femoral neck BMD (P<0.05, S1 Table). Then, our study was conducted in two phases. The first phase was to explore the individual effect of four heavy metals on the VCF risk and femoral neck BMD, separately by weighted multivariate logistic regression. Odds ratio (OR) with a 95% confidence interval (CI) was calculated. Weighted multivariate linear regression analysis was used to assess the individual effect of four heavy metals on the femoral neck BMD, separately. The second phase was to examine the joint effects of four heavy metals on the VCF risk and femoral neck BMD by BKMR model and quantile-based g computation (qgcomp). P<0.05 was considered as statistically significant difference.
Bayesian kernel machine regression model
BKMR is a semi-parametric method to assess the nonlinear and/or interaction associations of exposure-outcome [19]. BKMR model underwent 10,000 iterations using Markov chain Monte Carlo algorithm. Individual metal exposures were grouped using Pearson correlation analysis (S1 Fig) [20], and we found higher correlation for Pb and Cd compared to correlations for other metal exposures. Pb and Cd were classified as group 1, while Mn and Hg were classified as group 2. The group posterior inclusion probability (Group PIP) and conditional posterior inclusion probability (Cond PIP) were calculated to select the key metals that had an impact on VCF. The combined effect on VCF was calculated by comparing mixed heavy metals at or above the 60th percentile to the 50th percentile. The exposure-response relationship for each heavy metal and VCF risk were explored by taking all other metabolites at median levels. The interaction associations were assessed by taking exposure 1 heavy metal at its 25%, 50%, and 75% levels and other metabolites fixed at their median levels. R bkmr package was used for this analysis.
Quantile-based g computation
qgcomp is a parameterized model based on g-computation to for evaluating the effect of increasing all exposures in the mixture [21]. In this study, we adopted Bootstrapping method to calculate the population average exposure effect. R qgcomp package was used for this analysis.
Results
Characteristics of participants
Table 1 shows the characteristics of all eligible individuals (n = 1007). The average age was 55.98 (0.24) years. 48.67% were female. Of these participants, 6.77% are Mexican American, 71.50% are non-Hispanic white, 10.82% are non-Hispanic black, 4.18% are other Hispanic, and 6.73% are other races. In addition, among the population, 57 individuals developed VCF. We compared the differences in characteristics between VCF group (n = 57) and non-VCF group (n = 950). Compared with the non-VCF group, individuals with VCF were older (P<0.001). Individuals with VCF showed lower BMI, total femur BMD and femoral neck BMD than individuals with non-VCF (P<0.05).
Individual effect of four heavy metals on the VCF risk
We adopted a weighted multivariate logistic regression to explore the individual effect of four heavy metals on the VCF risk, separately. As shown in Table 2, after adjusting age, marital status, BMI, history of anti-osteoporosis medication use, total femur BMD, femoral neck BMD and menopause status, we found no statistical differences regarding the individual effects of four heavy metals on the risk of VCF (P>0.05).
Joint effects of four heavy metals on the VCF risk
We further analyzed the joint effects of four heavy metals on the VCF risk by the BKMR model. Table 3 displays the Group posterior inclusion probability (Group PIP) and Conditional posterior inclusion probability (Cond PIP) obtained by the BKMR model. The Group PIP of group 2 was greater than group 1 (0.7444 vs 0.3592). In the group 2, the Cond PIP of Mn was greater than Hg (0.9154 vs 0.0846), which indicated that Mn contributed most to the model for the VCF risk. Subsequently, Fig 2 reveals the overall association between heavy metal mixtures and VCF risk. We observed that the ln-transformed VCF risk tended to increase when the blood mixed heavy metals at their 60th and above percentile compared to their 50th percentile. The result of exposure-response relationship indicated that a slightly inverse link between Mn and ln-transformed VCF risk (S2 Fig).
In addition, there might be a potential interaction between Pb and Mn for ln-transformed VCF risk (S3 Fig). Although there was no significant difference, we found that each additional quantile in the qgcomp index may be positively associated with the risk of VCF in the adjusted model (Table 4, S4 Fig).
Individual effects of four heavy metals on the femoral neck BMD
The present study additionally assessed the individual /joint effects of four heavy metals on the femoral neck BMD by using weighted univariate and multivariate linear models. As shown in S2 Table, after adjusting age, gender, race/ethnicity, educational level, marital status, drinking, parental fracture, BMI, waist circumference, history of glucocorticoid use, history of anti-osteoporosis medication use, diabetes, menopause status, and total energy, we found that Cd was associated with femoral neck BMD (β = -0.05, 95%CI: -0.03, -0.01, P = 0.043).
Joint effects of four heavy metals on the femoral neck BMD
Using femoral neck BMD as the outcome, the mixed effect model BKMR for heavy metal exposure was established. In the fully adjusted BKMR model, a decreasing trend in the overall impact of four heavy metals on femoral neck BMD was observed, indicating that an increase in the mixture of heavy metal exposures was associated with a decreased risk of femoral neck BMD (S5 Fig). In the fully adjusted qgcomp model, the qgcomp index demonstrated a positive correlation with a reduction in the risk of femoral neck BMD (S3 Table, OR = 0.97, 95%CI: 0.95–0.98, P = 0.001). In the qgcomp model, Cd received the highest negative weights for the femoral neck BMD (S6 Fig).
Discussion
In this study, we used some statistical methods to investigate the individual and joint effects of Mn, Pb, Cd and Hg on the VCF risk. There were no statistical differences regarding the individual effects of Mn, Pb, Cd and Hg on the risk of VCF by weighted multivariate logistic regression, and the joint effects on the VCF risk by BKMR and qgcomp model. In addition, we found that an increase in the mixture of heavy metal exposures was associated with a decreased risk of femoral neck BMD.
To our knowledge, there have been several reports concerning the effects of heavy metals on bone diseases. In the meta-analysis study of Jalili C, et al., they found that exposure to Cd and Pb may be linked with an increased risk of osteopenia or osteoporosis [21]. Previously, several studies have stated possible mechanisms for the effects of heavy metals on bone structure. For example, Cd may affect BMD by stimulating osteoclast differentiation and activity [22]. In addition, the harmful effects of Cd on bone metabolism may also include oxidative stress, autophagy, apoptosis and mitochondrial dysfunction [10, 23, 24]. Nevertheless, no significant association was observed between individual heavy metal with VCF risk in this analysis.
At present, more and more studies have begun to focus on the mixed effects of metal exposure on the risk of bone diseases. A cross-sectional study of 627 Chinese adults aged ≥50 years examined the relationship of co-exposure to Mn, iron (Fe), copper (Cu) and selenium (Se) and osteoporosis risk, and BKMR model showed that co-exposure to Mn, Fe, Cu, and Se was associated with reduced osteoporosis risk [12]. Not only that, a population-based study of United States adults indicated that co-exposure to Mn, Pb, Cd, Hg Cu, Se, and zinc (Zn) may be risk factor for BMD [13]. However, few studies have been conducted to investigate the relationship between mixed exposures of Mn, Pb, Cd and Hg and the risk of VCF. In the present study, we found that VCF risk tended to increase when the blood mixed heavy metals at their 60th and above percentile in the BKMR model. In other words, the greater the level of mixed heavy metals exposure, the higher the risk of VCF.
Not only that, this study also assessed the individual and joint effects of four heavy metals on the femoral neck BMD. Several prior studies have suggested a potential association between heavy metals and femoral neck BMD [13, 25, 26]. One animal study indicated that chronic or even low-level exposure to Cd may potentially contribute to an increased likelihood of experiencing a fracture in the femoral neck [27]. Wang C et al., pointed out a negative association between blood Mn and femoral neck BMD was found in the fully adjusted model, especially for women aged 50–70 years [11]. In this study, after adjusting all potential confounding factors, we observed that only Cd level exposure was associated with femoral neck BMD, while no statistically significant differences were observed between other heavy metals and femoral neck BMD. The potential explanation is that the samples originate from diverse sources. The point worth noting is that, this study showed that mixture of heavy metal exposures was related to femoral neck BMD by using BKMR and qgcomp model. The findings also suggest a possible decrease in femoral neck BMD when individuals are simultaneously exposed to high levels of heavy metals.
Although the exact mechanism is unknown, these results help to reveal the harmful effect of mixed exposure to heavy metals on VCF risk. However, we also need to acknowledge some limitations of this study. First, due to a cross-sectional design in this study, the causal association between heavy metals and VCF is uncertain. More longitudinal studies are needed to confirm this finding. Second, we used the data of heavy metal was single-point measurements, which did not fully reflect the actual exposure levels of individuals [28]. Third, the nature of retrospective studies may introduce potential recall bias. Lastly, the NHANES database only is a representative of the United States population, and the results of this study were unlikely to be generalizable to other populations. More prospective epidemiological researches are required to confirm the current findings.
Conclusion
In this analysis, no significant correlation was observed between co-exposure to Mn, Pb, Cd and Hg and VCF risk. But we found that an increase in the mixture of heavy metal exposures was associated with a decreased risk of femoral neck BMD. However, more prospective studies are needed to validate these findings and explore potential biological mechanisms.
Supporting information
S1 Fig. Individual metal exposures were grouped using Pearson correlation analysis.
https://doi.org/10.1371/journal.pone.0303418.s001
(PDF)
S2 Fig. The exposure-response relationship of four heavy metals and ln-transformed VCF risk.
https://doi.org/10.1371/journal.pone.0303418.s002
(PDF)
S3 Fig. The interaction between four heavy metals and ln-transformed VCF risk.
https://doi.org/10.1371/journal.pone.0303418.s003
(PDF)
S4 Fig. The relationship of each additional quantile in the qgcomp index and the risk of VCF.
https://doi.org/10.1371/journal.pone.0303418.s004
(PDF)
S5 Fig. Combined effects of four heavy metals mixtures and femoral neck BMD by BKMR analysis.
https://doi.org/10.1371/journal.pone.0303418.s005
(PDF)
S6 Fig. Qgcomp model regression index weights of four heavy metals mixtures on femoral neck BMD.
https://doi.org/10.1371/journal.pone.0303418.s006
(PDF)
S1 Table. The selection of covariates related to VCF or femoral neck BMD.
https://doi.org/10.1371/journal.pone.0303418.s007
(DOCX)
S2 Table. Individual effects of four heavy metals on the femoral neck BMD.
https://doi.org/10.1371/journal.pone.0303418.s008
(DOCX)
S3 Table. Qgcomp model to assess the joint effects of four heavy metals on the femoral neck BMD.
https://doi.org/10.1371/journal.pone.0303418.s009
(DOCX)
References
- 1. Malik AT, Retchin S, Phillips FM, Xu W, Peters K, Yu E, et al. Declining trend in osteoporosis management and screening following vertebral compression fractures—a national analysis of commercial insurance and medicare advantage beneficiaries. Spine J. 2020;20(4):538–46. Epub 2019/11/05. pmid:31683068.
- 2. Alsoof D, Anderson G, McDonald CL, Basques B, Kuris E, Daniels AH. Diagnosis and Management of Vertebral Compression Fracture. Am J Med. 2022;135(7):815–21. Epub 2022/03/22. pmid:35307360.
- 3. Cho MJ, Moon SH, Lee JH, Lee JH. Association between Osteoporotic Vertebral Compression Fractures and Age, Bone Mineral Density, and European Quality of Life-5 Dimensions in Korean Postmenopausal Women: A Nationwide Cross-sectional Observational Study. Clin Orthop Surg. 2021;13(2):207–15. Epub 2021/06/08. pmid:34094011; PubMed Central PMCID: PMC8173246.
- 4. Chanplakorn P, Lertudomphonwanit T, Daraphongsataporn N, Sritara C, Jaovisidha S, Sa-Ngasoongsong P. Development of prediction model for osteoporotic vertebral compression fracture screening without using clinical risk factors, compared with FRAX and other previous models. Arch Osteoporos. 2021;16(1):84. Epub 2021/06/05. pmid:34085107; PubMed Central PMCID: PMC8175310.
- 5. Suzuki N, Ogikubo O, Hansson T. Previous vertebral compression fractures add to the deterioration of the disability and quality of life after an acute compression fracture. Eur Spine J. 2010;19(4):567–74. Epub 2009/09/18. pmid:19760437; PubMed Central PMCID: PMC2899832.
- 6. Jannetto PJ, Cowl CT. Elementary Overview of Heavy Metals. Clin Chem. 2023;69(4):336–49. Epub 2023/03/23. pmid:36945128.
- 7. Dutta S, Gorain B, Choudhury H, Roychoudhury S, Sengupta P. Environmental and occupational exposure of metals and female reproductive health. Environ Sci Pollut Res Int. 2022;29(41):62067–92. Epub 2021/09/25. pmid:34558053.
- 8. Ximenez JPB, Zamarioli A, Kacena MA, Barbosa RM, Barbosa F Jr., Association of Urinary and Blood Concentrations of Heavy Metals with Measures of Bone Mineral Density Loss: a Data Mining Approach with the Results from the National Health and Nutrition Examination Survey. Biol Trace Elem Res. 2021;199(1):92–101. Epub 2020/05/02. pmid:32356206.
- 9. Budi HS, Catalan Opulencia MJ, Afra A, Abdelbasset WK, Abdullaev D, Majdi A, et al. Source, toxicity and carcinogenic health risk assessment of heavy metals. Rev Environ Health. 2022. Epub 2022/10/02. pmid:36181731.
- 10. Cheng X, Niu Y, Ding Q, Yin X, Huang G, Peng J, et al. Cadmium Exposure and Risk of Any Fracture: A PRISMA-Compliant Systematic Review and Meta-Analysis. Medicine (Baltimore). 2016;95(10):e2932. Epub 2016/03/11. pmid:26962791; PMCID: PMC4998872.
- 11. Wang C, Zhu Y, Long H, Ou M, Zhao S. Relationship between blood manganese and bone mineral density and bone mineral content in adults: A population-based cross-sectional study. PLoS One. 2022;17(10):e0276551. Epub 2022/10/22. pmid:36269752; PubMed Central PMCID: PMC9586363.
- 12. Wei M, Huang Q, Dai Y, Zhou H, Cui Y, Song W, et al. Manganese, iron, copper, and selenium co-exposure and osteoporosis risk in Chinese adults. J Trace Elem Med Biol. 2022;72:126989. Epub 2022/05/06. pmid:35512597.
- 13. Wei MH, Cui Y, Zhou HL, Song WJ, Di DS, Zhang RY, et al. Associations of multiple metals with bone mineral density: A population-based study in US adults. Chemosphere. 2021;282:131150. Epub 2021/09/03. pmid:34470175.
- 14. Cosman F, Krege JH, Looker AC, Schousboe JT, Fan B, Sarafrazi Isfahani N, et al. Spine fracture prevalence in a nationally representative sample of US women and men aged ≥40 years: results from the National Health and Nutrition Examination Survey (NHANES) 2013–2014. Osteoporos Int. 2017;28(6):1857–1866. Epub 2017/02/09. 2017 Feb 7. pmid:28175980; PMCID: PMC7422504.
- 15. Genant HK, Wu CY, van Kuijk C, Nevitt MC. Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res. 1993;8(9):1137–48. Epub 1993/09/01. pmid:8237484.
- 16. Yu H, Tao Z, Luo X, Huang B, Zhou L. Role of metabolic equivalent between calcium intake and vertebral fractures: a cross-sectional study of NHANES 2013–2014. BMC Geriatr. 2022;22(1):986. Epub 2022/12/21. pmid:36539709; PubMed Central PMCID: PMC9768999.
- 17. Zhou F, Wu L, Shen G, Chen X, Liu C, Huang D, et al. Association between monocyte to high-density lipoprotein-cholesterol ratio and osteoporosis: An analysis of the National Health and Nutrition Examination Survey 2013–2014. J Investig Med. 2024;72(1):3–12. Epub 2023/09/20. pmid:37726952.
- 18. Chu NM, Hong J, Harasemiw O, Chen X, Fowler KJ, Dasgupta I, et al. Chronic kidney disease, physical activity and cognitive function in older adults-results from the National Health and Nutrition Examination Survey (2011–2014). Nephrol Dial Transplant. 2022;37(11):2180–9. Epub 2021/12/02. pmid:34850174; PubMed Central PMCID: PMC9585475.
- 19. Li H, Deng W, Small R, Schwartz J, Liu J, Shi L. Health effects of air pollutant mixtures on overall mortality among the elderly population using Bayesian kernel machine regression (BKMR). Chemosphere. 2022;286(Pt 1):131566. Epub 2021/07/23. pmid:34293557; PubMed Central PMCID: PMC8578302.
- 20. Wei C, Cao L, Zhou Y, Zhang W, Zhang P, Wang M, et al. Multiple statistical models reveal specific volatile organic compounds affect sex hormones in American adult male: NHANES 2013–2016. Front Endocrinol (Lausanne). 2022;13:1076664. Epub 2023/01/31. pmid:36714567; PubMed Central PMCID: PMC9877519.
- 21. Jalili C, Kazemi M, Taheri E, Mohammadi H, Boozari B, Hadi A, et al. Exposure to heavy metals and the risk of osteopenia or osteoporosis: a systematic review and meta-analysis. Osteoporos Int. 2020;31(9):1671–82. Epub 2020/05/04. pmid:32361950.
- 22. Li D, Lin H, Zhang M, Meng J, Hu L, Yu B. Urine Cadmium as a Risk Factor for Osteoporosis and Osteopenia: A Meta-Analysis. Front Med (Lausanne). 2021;8:648902. Epub 2021/05/04. pmid:33937289; PubMed Central PMCID: PMC8085254.
- 23. Ran D, Ma Y, Liu W, Luo T, Zheng J, Wang D, et al. TGF-β-activated kinase 1 (TAK1) mediates cadmium-induced autophagy in osteoblasts via the AMPK / mTORC1 / ULK1 pathway. Toxicology. 2020;442:152538. Epub 2020/07/22. pmid:32693121.
- 24. Luo H, Gu R, Ouyang H, Wang L, Shi S, Ji Y, et al. Cadmium exposure induces osteoporosis through cellular senescence, associated with activation of NF-κB pathway and mitochondrial dysfunction. Environ Pollut. 2021;290:118043. Epub 2021/09/04. pmid:34479166.
- 25. Engström A, Michaëlsson K, Suwazono Y, Wolk A, Vahter M, Akesson A. Long-term cadmium exposure and the association with bone mineral density and fractures in a population-based study among women. J Bone Miner Res. 2011;26(3):486–95. Epub 2010/08/25. pmid:20734452.
- 26. Kim ES, Shin S, Lee YJ, Ha IH. Association between blood cadmium levels and the risk of osteopenia and osteoporosis in Korean post-menopausal women. Arch Osteoporos. 2021;16(1):22. Epub 2021/02/03. pmid:33527234; PubMed Central PMCID: PMC7850996.
- 27. Brzóska MM, Majewska K, Moniuszko-Jakoniuk J. Mechanical properties of femoral diaphysis and femoral neck of female rats chronically exposed to various levels of cadmium. Calcif Tissue Int. 2005;76(4):287–98. Epub 2005/03/03. pmid:15742233.
- 28. Wang X, Mukherjee B, Park SK. Associations of cumulative exposure to heavy metal mixtures with obesity and its comorbidities among U.S. adults in NHANES 2003–2014. Environ Int. 2018;121(Pt 1):683–94. Epub 2018/10/14. pmid:30316184; PubMed Central PMCID: PMC6268112.