DOI: https://doi.org/10.21203/rs.3.rs-1574819/v1
Background and Aim:
Folate, a water-soluble vitamin B, plays an important role in the human body. An ever-increasing life expectancy and the rise of an aging society, has brought with it associated age-related health concerns, folate concentration is known to be significantly impacted by age. Osteoporosis, has many causes, but for the most part is an age-associated disease, that becomes more and more exasperated with age. Thus, the purpose of this article was to explore the relationship between folate which plays an essential role in bodily functions, and osteoporosis.
Methods:
Data collected over the last decade by the National Health and Nutrition Examination Survey were selected for our study. A total number of 8429 individuals (4030 females and 4399 males) were included in our final analysis. The individuals were divided into an osteoporosis group and a non-osteoporosis group (532 osteoporosis and 7897 non-osteoporosis). Logistic regression and restricted cubic spline model analysis were used to analyze the relationship between folate concentration and osteoporosis using R software.
Results:
Folate concentrations were found to be lower in the osteoporosis group than the non-osteoporosis group (44.15 ± 0.58 VS 57.9 ± 1.78, P < 0.001). Men with higher folate concentrations had a higher osteoporosis risk compared to men with low folate concentrations (OR 4.26, 95% CI = 1.81, 9.95). In patients aged 50 to 70 years, high levels of folate were associated with increased incidences of osteoporosis compared with low levels (OR 1.75, 95% CI = 1.14, 2.67), no differences were observed after adjusting for covariates such as gender. In the restricted cubic spline, the incidences of osteoporosis showed a downward trend with the increase of folate concentration, when folate concentration reached around 50 nmol/L, the incidences of osteoporosis began to increase.
Conclusion:
Folate concentration can affect the occurrence of osteoporosis: low levels of folate can play a protective role, while high levels of folate can be a risk factor. The specific cut-off point of folate concentration needs further investigation.
National Health and Nutrition Examination Survey (NHANES) is a long investigation project that dates back to 1959. This project selects different populations from various regions of the United States to conduct a statistical investigation, collects various personal information, and forms a representative population database for relevant scholars to study.
Osteoporosis is a disease that weakens bones and make them brittle, it occurs when bones density decreases and has various causes [1]. It occurs mainly in postmenopausal women and older individuals. With the fragility of the aging population, osteoporosis is significantly more detrimental to older individuals health [2–3]. A growing number of researchers are starting to pay attention to osteoporosis and its prevention and treatment have become a focal research point. Existing research shows that the following preventative methods against osteoporosis can be implemented: reasonable eating habits, moderate physical activity, and a healthy lifestyle [4].
Serum folate, as a water-soluble vitamin, is mainly manifested in the following aspects: participating in the metabolism of genetic material and protein, affecting the growth and development of animals, and improving the body’s immunity. Possible causes of folate deficiency include insufficient intake or increased requirement, intestinal malabsorption, vitamin deficiency, and liver disease, etc. [5–6]. Previous studies found that folate deficiency is related to the formation of Neonatal neural tube defects (NTDs) and Megaloblastic anemia (MA) [7–8]. It was later found that folate deficiency can affect the occurrence of cardiovascular disease, Alzheimer and depression [9–11]. However, folate has been shown to act on Homocysteine to cause bone mineral density (BMD) change [12]. Additionally, osteoporosis induced by high-fat diet in mice can be inhibited by folate [13]. Americans have higher levels of folate than other countries due to the long-term consumption of foods containing folate additives [14]. If there is a relationship between folate and osteoporosis, it is imperative we explore the underlying mechanisms.
Data used in our study was collected over the last decade (2007 to 2018) by the NHANES, one of the largest and most comprehensive databases in the United States. Due to the COVID-19 outbreak, data from 2019 to 2020 were not collected, as it was considered incomplete and was not nationally representative. Finally, data of four periods (2007–2008, 2009–2010, 2013–2014, and 2017–2018) were selected. Two periods (2011–2012, 2015–2016) without the osteoporosis project, were excluded. The data of 59 842 participants, over the age of 20, were assessed. Only 8429 participants were selected for our study and included in our final analysis. The NHANES database selects the representative population of each region, different weights are matched for each person, to better estimate the characteristics of the overall population data. We extracted the following demographic data: Body Mass Index (BMI), physical activity information, sedentary time per day (min), smoking, alcohol drinking, hypertension, diabetes, serum folate, osteoporosis of each participant, participants with any missing information were excluded. The survey protocol was approved by the National Center for Health Statistics (NCHS) Research Ethics Review Board, and all participants provided informed consent.
More than a decade ago, Serum folate was measured by using the Bio-Rad Laboratories "Quantaphase II Folate" radioassay kit. Later, Microbiologic assay (MBA) was developed to test folate content, this method is considered more accurate [15]. Since 2011, Serum total folate has been calculated as the sum of individual folate forms. Five folate forms, 5-methyltetrahydrofolate, folic acid, 5-formyltetrahydrofolate, tetrahydrofolate, and 5, 10-methenyltetrahydrofolate are measured by isotope-dilution high performance liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) [16]. We classified folate concentration into three grades (according to the interquartile range): low, medium, and high. With Q1 considered a low grade, Q2-Q3 considered an intermediate grade, and Q4 considered a high grade.
Information about osteoporosis was screened out from the questionnaires data and the dual-energy X-ray absorptiometry (DXA) test, responses "Yes" to family questions such as "Have you ever been diagnosed with osteoporosis/brittle bones", was diagnosed as osteoporosis. Additionally, DXA is the most commonly used method for bone mineral density measurement in a clinic, it has the advantages of high speed and low radiation. According to World Health Organization (WHO) criteria, osteoporosis should be considered if the density of the lumbar spine or femoral neck is less than 2.5 standard deviation of peers [17].
In the NHANES database, we selected age, race, sex, Body Mass Index, physical activity, sedentary time, smoking, alcohol consumption, hypertension, and diabetes as covariates. BMI adopts international units (kg/m2). Physical activity was defined as "How much time do you spend in a typical day doing moderate-intensity?". Sedentary time was defined as "How much time do you usually spend in a typical day sitting". Three levels distinguished the degree of smoking: never (less than a 100 cigarettes in their entire life), former smoker (smoked in the past but completely stopped), and current smoker (has smoked more than a 100 cigarettes in their life and currently smoke, either on some days or every day). Three levels distinguish the degree of drinking: never (has hardly consumed any alcohol in their life), light (consumes alcohol occasionally throughout the year), and moderate to heavy (weekly or daily consumption). The diagnostic criteria for hypertension was "Have you ever been diagnosed with hypertension by a doctor" or “Are you currently using any antihypertensive medication.” The diagnostic criteria for diabetes is "Have you ever been diagnosed with diabetes by a doctor" or “Are you currently using insulin, or any other antidiabetic medication.” The above information can be found in the Questionnaire Data or Examination Date.
R software (version 4.1.2) was used for data statistics and analysis. R is a language and environment for statistical computing and graphics. It was developed at Bell Laboratories (formerly AT&T, now Lucent Technologies). T-test was used to compare the measurement data of normal distribution, X2 test was used to compare the count data. Serum folate concentration was divided into three grades according to the quartile (low grade is Q1, middle grade is Q2-Q3, high grade is Q4). Taking the lower grade as the reference, logistic regression analysis was used to analyze the relationship between folate and osteoporosis, stratified analysis was conducted for age, gender, and race, and the results were expressed by Odds Ratios (ORs) and corresponding 95% confidence intervals (CIs). The restricted cubic spline (RCS) model can be used to analyze the nonlinear relationship between folate and osteoporosis. Due to the large sample size, four points (5th, 35th, 65th, and 95th ) of folate concentration were used as knots, and the mean was used as the reference points [18–19]. The NHANES adopted complex multi-stage sampling, our data were analyzed after matching with appropriate weights. Unless otherwise specified, all data were considered statistically significant with a two-tailed P value < 0.05.
In this study, a total of 8429 participants were selected (Table 1). The proportion of males and females was almost equivalent with 52.19% and 47.81%, respectively. Females with osteoporosis accounted for 81.58% of the data, far more compared to the males (p < 0.001). The incidence of osteoporosis was 27.82% in participants who performed moderate physical activity each day and 72.18% in participants who performed no physical activity (p < 0.001). The proportion of participants with hypertension and diabetes who acquired osteoporosis was 58.08% and 34.59% (p < 0.001), respectively. The serum folate content in participants with osteoporosis was 44.15 ± 0.58 (nmol/L), which was lower than that of participants without osteoporosis (p < 0.001).
| Variables | Osteoporosis (n = 532) | Non-osteoporosis (n = 7897) | P |
|---|---|---|---|
| Gender | < 0.001 | ||
| Female (n, %) | 434(81.58) | 3596(45.54) | |
| Male(n, %) | 98(18.42) | 4301(54.46) | |
| Age(years) | 49.11 ± 0.32 | 65.45 ± 0.54 | < 0.001 |
| Race (n, %) | < 0.001 | ||
| Mexican American | 49(9.21) | 1212(15.35) | |
| Non-Hispanic Black | 66(12.41) | 1492(18.89) | |
| Non-Hispanic White | 332(62.41) | 3947(49.98) | |
| Other Hispanic | 46(8.65) | 773(9.79) | |
| Other race | 39(7.33) | 473(5.99) | |
| BMI(kg/m2) | 28.86 ± 0.1 | 28.28 ± 0.38 | 0.1345 |
| Moderate physical activity(n, %) | < 0.001 | ||
| Yes | 148(27.82) | 3098(39.23) | |
| No | 384(72.18) | 4799(60.77) | |
| Sedentary time(min) | 370.72 ± 4.82 | 367.9 ± 8.57 | 0.7747 |
| Hypertension(n,%) | < 0.001 | ||
| Yes | 309(58.08) | 2924(37.03) | |
| No | 223(41.92) | 4973(62.97) | |
| Diabetes(n,%) | 0.001 | ||
| Yes | 184(34.59) | 2010(25.45) | |
| No | 348(65.41) | 5887(74.55) | |
| Cigarette(n,%) | 0.001 | ||
| Never | 262(49.25) | 3896(49.34) | |
| Former | 183(34.40) | 2196(27.81) | |
| Now | 87(16.35) | 1805(22.86) | |
| Alcohol(n,%) | 0.031 | ||
| Never | 210(39.46) | 2580(32.67) | |
| Light | 138(25.94) | 2046(25.91) | |
| Moderate to heavy | 184(34.59) | 3271(41.42) | |
| Folate(nmol/L) | 44.15 ± 0.58 | 57.9 ± 1.78 | < 0.001 |
| *BMI:Body Mass Index | |||
We used Logistics regression to analyze the relationship between folate and osteoporosis among different subgroups (Table 2), the incidence of osteoporosis was significantly increased in men with middle and high levels of folate compared with low levels (OR = 3.51, 95% CI = 1.55, 7.91 and OR = 8.02, 95% CI = 3.63, 17.66). After adjusting for the covariates, the incidence of osteoporosis was still significantly increased (OR = 3.14, 95% CI = 1.46, 6.73 and OR = 4.26, 95% CI = 1.81, 9.95). In women, before adjustment, participants with high levels of folate were 1.76 times more than those with low levels (OR = 1.76, 95% CI = 1.29, 2.4). After adjustment, there was no statistically significant difference between the two levels of folate. In participants aged 50 to 70 years, middle and high levels of folate were associated with an increased incidence of osteoporosis compared with low levels (OR = 1.71, 95% CI = 1.03, 2.82 and OR = 1.75, 95% CI = 1.14, 2.67), and no differences in folate levels were observed after adjusting for covariates. In different ethnic subgroup analyses, high folate levels in some ethnic groups increased the incidence of osteoporosis (all p < 0.05).
| Serum folate quartiles | Osteoporosis/total | OR(95% CI) | P | Adjusted OR(95% CI) | P | |
|---|---|---|---|---|---|---|
| Male | ||||||
| Low(Q1) | 41/852 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 11/1319 | 3.51(1.55, 7.91) | 0.004 | 3.14(1.46, 6.73) | 0.005 | |
| High(Q4) | 41/852 | 8.02(3.63, 17.66) | < 0.001 | 4.26(1.81, 9.95) | 0.002 | |
| Female | ||||||
| Low(Q1) | 173/1121 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 72/953 | 1.24(0.85, 1.81) | 0.272 | 1.08(0.71, 1.63) | 0.720 | |
| High(Q4) | 189/1956 | 1.76(1.29, 2.4) | < 0.001 | 0.93(0.65, 1.34) | 0.715 | |
| Age < 50 years | ||||||
| Low(Q1) | 16/1241 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 19/1991 | 0.74(0.3, 1.81) | 0.519 | 0.79(0.29, 2.17) | 0.659 | |
| High(Q4) | 7/461 | 1.14(0.37, 3.49) | 0.816 | 0.98(0.26, 3.77) | 0.982 | |
| Age 50–70 years | ||||||
| Low(Q1) | 41/855 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 139/1732 | 1.71(1.03, 2.82) | 0.041 | 1.63(0.93, 2.84) | 0.094 | |
| High(Q4) | 91/915 | 1.75(1.14, 2.67) | 0.012 | 1.37(0.8, 2.33) | 0.255 | |
| Age > 70 years | ||||||
| Low(Q1) | 26/176 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 77/461 | 1.01(0.55, 1.87) | 0.967 | 0.98(0.53, 1.81) | 0.942 | |
| High(Q4) | 116/597 | 1.21(0.71, 2.06) | 0.481 | 1.01(0.59, 1.73) | 0.974 | |
| Mexican American | ||||||
| Low(Q1) | 6/367 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 31/710 | 2.41(0.97, 5.92) | 0.065 | 1.73(0.65, 4.58) | 0.283 | |
| High(Q4) | 12/184 | 3.84(1.4, 10.52) | 0.012 | 1.35(0.42, 4.34) | 0.618 | |
| Non-Hispanic Black | ||||||
| Low(Q1) | 20/633 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 35/740 | 1.82(1.04, 3.2) | 0.04 | 1.54(0.88, 2.69) | 0.142 | |
| High(Q4) | 11/185 | 2.65(1.3, 5.38) | 0.009 | 1.17(0.55, 2.48) | 0.693 | |
| Non-Hispanic White | ||||||
| Low(Q1) | 46/941 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 133/2041 | 1.33(0.86, 2.04) | 0.206 | 1.15(0.72, 1.83) | 0.565 | |
| High(Q4) | 153/1297 | 2.31(1.6, 3.32) | < 0.001 | 1.07(0.71, 1.59) | 0.754 | |
| Other Hispanic | ||||||
| Low(Q1) | 5/210 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 18/426 | 1.94(0.63, 5.96) | 0.256 | 1.19(0.35, 4.07) | 0.785 | |
| High(Q4) | 23/183 | 6.27(1.85, 21.06) | 0.005 | 2.52(0.59, 10.69) | 0.221 | |
| Other race | ||||||
| Low(Q1) | 6/121 | Reference | - | Reference | - | |
| Middle (Q2-Q3) | 18/267 | 2.16(0.51, 9.11) | 0.301 | 2.03(0.55, 7.46) | 0.293 | |
| High(Q4) | 15/124 | 2.92(0.71, 11.96) | 0.145 | 1.41(0.37, 5.36) | 0.620 | |
*OR: odds ratio; CI: confidence interval; Q: quartile
**Adjusted for age, race, sex, BMI, physical activity, sedentary time, smoking, alcohol consumption, hypertension, and diabetes.
There is a nonlinear relationship between folate concentration and osteoporosis. In the total population, the incidence of osteoporosis showed a downward trend with the increase of folate concentration at first, until it reached the lowest point around 25nmol/L, and then with the increase of folate levels, the incidence of osteoporosis began to increase, but still below the normal level. After 50 nmol/L, the incidence of osteoporosis increased significantly. In men and women groups, the relationship between osteoporosis and folate was similar to that in the total population. In age stratification, less than 50 years and more than 70 years were insignificant, age between 50–70 years old shows a continuous increase in osteoporosis with the increase of folate. (Fig. 1, 2)
Our study investigated the relationship between folate levels and osteoporosis, in the US population, over the last decade. Due to folate’s unique properties and functions, it has always played an irreplaceable role in organisms. Recent studies have shown that folate may regulate lipid metabolism and oxidative stress reaction, activate AMP-activated protein kinase (AMPK) pathway, and reduce high-fat diet induced osteoporosis. Additionally, it has also been found that B vitamins and folate supplements can improve BMD [13, 20]. All these suggest that folate may affect bone formation.
In this study, age, sex, hypertension and physical activity can influence the occurrence of osteoporosis, which is consistent with existing studies [21]. However, the incidence of osteoporosis was not higher in diabetic than in non-diabetic (P < 0.001). Studies have found that BMD increases in diabetic patients, although the risk of fracture is increased eventually, this is related to the duration of diabetes and complications [22–23], and more research needs to be conducted.
Folate concentration was significantly lower in the osteoporosis group than in the non-osteoporosis group (P < 0.001). For further study, we divided folate into low, medium, and high grades. In men, the risk of osteoporosis with moderate and high folate levels was significantly higher than the low folate levels, and the higher the folate level, the higher the risk, suggesting that folate may affect the formation of bone, and high folate levels may have an inhibitory effect. In age stratification, between the ages of 50 to 70, medium and high levels of folate showed to have a higher risk of osteoporosis than lower levels. After adjusting for covariates, such as gender, no significant difference was found in the results, which may be related to female hormones. Women will experience menopause at about 50 years old, and the loss of estrogen after menopause will lead to bone loss [24], increasing the occurrence of overall osteoporosis. No association between folate and osteoporosis was found in the group older than 70 years. It was considered that the decreased secretion of hormones in older individuals can stimulate osteoclasts and inhibit osteoblasts [25], accompanied by the decline of organ function and physical activity, resulting in bone decline, which will affect the occurrence of osteoporosis. In racial stratification, except for the multi-ethnic minorities, other races have an increased risk of osteoporosis with the increase of folate levels. This shows that folate is associated with osteoporosis, however the relationship between different races needs further investigation.
Considering that folate levels do not show a simple linear relationship with osteoporosis, we used RCS to describe it. In the total population, with the increase of folate concentration, the risk of osteoporosis gradually decreased, reaching the lowest point at 25nmol/L, and then there was a J-shaped relationship between the two. In the stratified analysis, the images were similar among the subgroups of men and women, aged 50–70. Folate levels lower than 50nmol/L seems to have a protective factor against osteoporosis, while levels higher 50nmol/L seems to be a risk factor, which gives us a reference for the possible prevention of osteoporosis. This is similar to a study by Stanley et al., who found that both high and low folate levels increase cardiovascular mortality in hypertensive individuals [26].
The United States is a large folate supplement distributor. Since the 1990s, the government has implemented adding folate to flour and grain to improve people's folate levels. The implementation of folate policy has also played a role in reducing the occurrence of several diseases. In 2018, the United States stopped implementing the policy of adding folate, causing speculation. Some studies have found that excessive folate does not bring benefits, however increases cardiovascular events and all-cause mortality [27–28]. Professor Evans points out that high serum folate is associated with an increased risk of death in adults with diabetes in a 15- year study [29]. The current explanation is that excessive folate may increase the level of unbound folate and increase the degradation of folate [30], and also control biological methylation and nucleotide synthesis, thus damaging DNA integrity [31]. Moreover, excessive intake of natural folate will not cause poisoning, while long-term high dose of synthetic folate intake will produce a large amount of unmetabolized folate, which may reduce the cytotoxicity of its natural killer cells [32]. It may also affect the absorption of other nutrients or mask the symptoms of vitamin B12 deficiency [33].
According to our knowledge, our study is the first to reveal the relationship between folate and osteoporosis, and to determine the cut-off point of folate concentration to prevent osteoporosis. The NHANES database had an abundant amount of representative population data, which provided a large number of samples to support our research. At the same time, there were some limitations. Dietary intake is an important factor affecting folate levels. There is currently insufficient data regarding this, thus we cannot report on the statistically relevant variables. Additionally, the article is a cross-sectional study, it cannot reveal the prognostic relationship between folate and osteoporosis.
In summary, the concentration of folate can affect the occurrence of osteoporosis. Low levels of folate can play a protective role, while high levels folate can be a risk factor. The ascertain the specific folate cut-off point further studies need to be conducted.
National Health and Nutrition Examination Survey
Neonatal neural tube defects
Megaloblastic anemia
Bone mineral density
Corona Virus Disease 2019
Body mass index
National Center for Health Statistics
Dual-energy X-ray absorptiometry
World Health Organization
Odds Ratios
Corresponding 95% confidence intervals
Restricted cubic spline
Availability of data and materials
The datasets used and/or analyzed during the current study are available at https://wwwn.cdc.gov/nchs/nhanes/nhanes3/DataFiles.aspx
Ethics approval and consent to participate
The survey protocol was approved by the National Center for Health Statistics (NCHS) Research Ethics Review Board, and all participants provided informed consent.
Consent for publication
Not applicable.
Competing Interests
The authors declare no conflict of interest.
Funding
This study was supported by Beijing Medical Award Foundation of China (No.YXJL202103530603).
Authors' contributions
Senjie Li: Conceptualization, Data curation, Formal analysis, Investigation, Writing- original draft, Writing-review & editing. Dongqing Lv: Conceptualization, Investigation, Methodology, Project administration, Writing-original draft, Writing-review & editing. Hong Yang: Conceptualization, Investigation, Methodology, Project administration, Software. ShaoYi Yan: Project administration, Formal analysis, Investigation. Lei Wu: Conceptualization, Methodology, Data curation, Supervision, Writing-review & editing. Yongping Jia: Conceptualization, Methodology, Project administration, Supervision, Writing-review & editing.
Acknowledgements
We thank Professor Jia for guidance on the content of the article and Doctor Yang for his support in data analysis.