1 Medical Insurance Department, The Affiliated Changsha Central Hospital, Hengyang Medical School, University of South China, 410004 Changsha, Hunan, China
2 Department of Emergency Medicine, The Affiliated Changsha Central Hospital, Hengyang Medical School, University of South China, 410004 Changsha, Hunan, China
3 Department of Respiratory and Critical Care Medicine, The Affiliated Changsha Central Hospital, Hengyang Medical School, University of South China, 410004 Changsha, Hunan, China
†These authors contributed equally.
Abstract
Remnant cholesterol (RC) is a risk factor for the development of atherosclerosis. Vitamin E has antioxidant properties, making it a potentially effective management tool for preventing cardiovascular disease (CVD). However, the relationship between vitamin E intake and RC remains unclear.
We conducted a cross-sectional study using data from the National Health and Nutrition Examination Survey (NHANES) Survey 2007–2020. 11,585 participants (aged ≥20, 48% male) were included. Information on vitamin E intake (dietary vitamin E intake and total vitamin E intake) was collected. RC was defined as serum total cholesterol minus high-density lipoprotein and low-density lipoprotein cholesterol. Survey-weighted linear regression models and a restricted cubic spline (RCS) were used to test the relationship between vitamin E intake and RC. Subgroup analyses and interaction tests were also performed to verify the robustness of the results.
After adjusting for all potential confounders (demographics, socioeconomic status, lifestyle, diet, and comorbidities), dietary vitamin E intake was negatively associated with RC (β = –0.21, 95% CI: (–0.29, –0.12), p < 0.0001), and this negative association was also present between total vitamin E intake and RC (β = –0.12, 95% CI: (–0.18, –0.06), p < 0.0001). The RCS analysis revealed a nonlinear negative association between vitamin E intake and RC. The negative correlation existed in different subgroups, with no interaction except for the “use of vitamin E supplements” subgroup.
Vitamin E intake showed a protective association with RC. The results suggest that increasing dietary vitamin E intake may help reduce RC levels and CVD risk.
Keywords
- remnant cholesterol
- cardiovascular disease
- vitamin E
- cross-sectional study
- NHANES
Cardiovascular disease (CVD) is the leading cause of death globally, causing approximately 19.8 million deaths worldwide in 2022 [1]. According to a report from the American Heart Association in 2019, 48% (about 121.5 million) of adults in the United States (US) suffer from cardiovascular disease (including coronary heart disease, heart failure, stroke or hypertension) [2]. In recent years, remnant cholesterol (RC) has been recognized as a direct factor in the development of atherosclerosis. Elshazly et al. [3] reported that RC, compared with serum low-density lipoprotein cholesterol (LDL-C), correlated more significantly with the progression of coronary atherosclerosis in statin-treated patients. Moreover, high levels of RC, but not serum LDL-C, were associated with increased all-cause mortality in patients with ischaemic heart disease [4]. RC is a triglyceride-rich lipoprotein cholesterol that consists of serum very low-density lipoprotein cholesterol in the fasting state, intermediate-density lipoprotein cholesterol, and cholesterol in the celiac residue in the postprandial state [5]. As an independent lipid risk marker, RC has been suggested as a therapeutic target for the clinical development of novel anti-atherosclerotic drugs and CVD prevention [3, 6].
To date, no guidelines or consensus exists on reducing RC levels. Despite achieving desirable serum LDL-C levels, intensive lipid-lowering therapy does not eliminate the residual risk of recurrent atherosclerotic cardiovascular events [7, 8]. Several international lipid management and treatment guidelines recognize the importance of diet and lifestyle in CVD prevention [9, 10]. Thus, early management of CVD by preventing and treating risk factors through behavioral changes (i.e., weight management, healthy diet, physical activity, and smoking cessation) is essential to reduce the risk of developing CVD [11, 12]. Lipid peroxidation is central to the pathogenesis of CVD [13], and vitamin E has anti-inflammatory and antioxidant properties. Therefore, dietary vitamin E may be an effective management strategy for preventing CVD.
Vitamin E includes tocopherols and tocotrienols, both of which have four isomers:
The National Health and Nutrition Examination Survey (NHANES) is a large epidemiological survey database administered by the National Center for Health Statistics (NCHS) dedicated to investigating the nutritional and health status of US citizens. The survey utilized a multistage sampling design and collected data every two years.
In this cross-sectional study, we use publicly available data for seven cycles: 2007–2008, 2009–2010, 2011–2012, 2013–2014, 2015–2016, 2017–2018, and 2019–2020. The study included 116,876 potential participants. Participants under 20 years of age (n = 52,563), lacking RC data (n = 37,615), and deficient in vitamin E intake (n = 11,455) were excluded. Participants with missing weights were excluded (n = 693). Participants with incomplete data were excluded (n = 2965). A total of 11,585 participants aged
Fig. 1. Flowchart of the standard for participants enrolled in the study.
Participants’ dietary consumption, including food composition, nutrient content, and caloric intake, was calculated according to the study protocol. The diet interview was conducted with the United States Department of Agriculture and the United States Department of Health and Human Services (DHHS). DHHS’s National Center for Health Statistics is responsible for sample design and data collection; and the US Department of Agriculture’s Dietary Research Food and Nutrition Database is used to encode individual foods and portion sizes reported by participants, process data, and calculate the nutritional value of nutritional intake. The staff is professionally trained and the data is reviewed to ensure quality. Data is collected every 2 years. The dietary interviews were conducted in two sessions. The first interview was conducted face-to-face, and the second was conducted by phone 3-10 days later. Dietary vitamin E intake could only be derived from food intake data and did not include supplemental use. According to previously published literature [15, 19], 24-hour dietary vitamin E intake and vitamin E supplements were calculated by averaging two 24-hour dietary recall interviews, and total vitamin E intake equaled the sum of dietary vitamin E and vitamin E supplements. The detailed data processing and calculation methods are publicly available at https://wwwn.cdc.gov/Nchs/Data/Nhanes/Public/2007/DataFiles/dr1tot_e.htm#DR1TATOC.
According to previous studies [21, 22, 23], RC levels were calculated as serum total cholesterol (TC) minus high-density lipoprotein cholesterol (HDL-C) and LDL-C. The Friedewald equation was used to calculate serum LDL-C as recommended by the guidelines [11].
Based on previously published literature [15, 16, 19], we collected potential covariates which included demographics, socioeconomic status, lifestyle, diet, and comorbidities. Demographics included sex, age, and race. The socioeconomic factors included education level, marriage status, body mass index (BMI), and family income-poverty ratio (PIR). PIR was categorized as
| Variables | Total | Q1 [2, 15] | Q2 (15, 25] | Q3 (25, 80] | p-value | |
| Age (years), n (%) | ||||||
| 3701 (35.61) | 1661 (46.14) | 1116 (32.52) | 924 (27.63) | |||
| 40–60 | 3949 (37.37) | 1202 (32.35) | 1373 (38.01) | 1374 (42.07) | ||
| 3935 (27.02) | 1036 (21.50) | 1541 (29.47) | 1358 (30.30) | |||
| Gender, n (%) | ||||||
| Female | 6021 (51.86) | 2283 (58.38) | 2055 (51.63) | 1683 (45.13) | ||
| Male | 5564 (48.14) | 1616 (41.62) | 1975 (48.37) | 1973 (54.87) | ||
| Race, n (%) | ||||||
| Non-Hispanic Black | 2370 (10.59) | 1192 (16.18) | 796 (9.92) | 382 (5.34) | ||
| Non-Hispanic White | 5148 (69.71) | 1544 (65.95) | 1810 (70.70) | 1794 (72.68) | ||
| Mexican American | 1616 (7.61) | 368 (5.82) | 588 (7.72) | 660 (9.41) | ||
| Other Hispanic | 1137 (5.11) | 304 (4.72) | 414 (5.27) | 419 (5.35) | ||
| Other Race | 1314 (6.98) | 491 (7.33) | 422 (6.39) | 401 (7.23) | ||
| Education level, n (%) | ||||||
| 2317 (13.22) | 563 (9.47) | 843 (14.19) | 911 (16.20) | |||
| High School | 2623 (22.85) | 814 (20.02) | 935 (23.66) | 874 (25.01) | ||
| 6645 (63.93) | 2522 (70.51) | 2252 (62.15) | 1871 (58.79) | |||
| Marital status, n (%) | ||||||
| Never married | 2066 (17.87) | 953 (22.92) | 634 (16.42) | 479 (14.00) | ||
| Married/Living with a partner | 7008 (64.27) | 2208 (61.74) | 2463 (64.75) | 2337 (66.46) | ||
| Widowed/Divorced/Separated | 2511 (17.86) | 738 (15.34) | 933 (18.83) | 840 (19.53) | ||
| PIR, n (%) | 0.01 | |||||
| 2186 (12.76) | 684 (12.14) | 775 (13.21) | 727 (12.96) | |||
| 1–3 | 4860 (35.50) | 1571 (33.17) | 1679 (36.09) | 1610 (37.38) | ||
| 4539 (51.73) | 1644 (54.70) | 1576 (50.70) | 1319 (49.66) | |||
| BMI (kg/m2), n (%) | ||||||
| 3217 (29.38) | 1619 (45.32) | 1028 (26.58) | 570 (15.28) | |||
| 25–30 | 3783 (32.63) | 1166 (29.71) | 1406 (35.87) | 1211 (32.28) | ||
| 4585 (37.99) | 1114 (24.97) | 1596 (37.54) | 1875 (52.44) | |||
| Smoke status, n (%) | ||||||
| Former | 2884 (25.84) | 846 (23.35) | 1008 (25.57) | 1030 (28.80) | ||
| Never | 6531 (56.61) | 2455 (62.65) | 2262 (56.52) | 1814 (50.23) | ||
| Now | 2170 (17.55) | 598 (13.99) | 760 (17.90) | 812 (20.98) | ||
| Alcohol consumption, n (%) | ||||||
| No | 2952 (20.98) | 869 (18.21) | 1047 (22.13) | 1036 (22.69) | ||
| Yes | 8633 (79.02) | 3030 (81.79) | 2983 (77.87) | 2620 (77.31) | ||
| Lipid-lowering drugs, n (%) | ||||||
| No | 9036 (79.99) | 3295 (86.26) | 3086 (79.36) | 2655 (73.96) | ||
| Yes | 2549 (20.01) | 604 (13.74) | 944 (20.64) | 1001 (26.04) | ||
| Recreational activity, n (%) | ||||||
| No | 5757 (43.93) | 1651 (36.22) | 2072 (45.75) | 2034 (50.24) | ||
| Yes | 5828 (56.07) | 2248 (63.78) | 1958 (54.25) | 1622 (49.76) | ||
| Hypertension, n (%) | ||||||
| No | 6735 (62.87) | 2640 (74.32) | 2284 (61.49) | 1811 (52.08) | ||
| Yes | 4850 (37.13) | 1259 (25.68) | 1746 (38.51) | 1845 (47.92) | ||
| DM, n (%) | ||||||
| No | 7120 (67.39) | 2986 (81.65) | 2441 (67.37) | 1693 (52.11) | ||
| Borderline | 2053 (17.11) | 477 (10.97) | 741 (18.04) | 835 (22.70) | ||
| Yes | 2412 (15.50) | 436 (7.38) | 848 (14.58) | 1128 (25.19) | ||
| CVD, n (%) | ||||||
| No | 10,307 (91.11) | 3580 (94.02) | 3552 (91.04) | 3175 (88.07) | ||
| Yes | 1278 (8.89) | 319 (5.98) | 478 (8.96) | 481 (11.93) | ||
| Use of vitamin E supplements, n (%) | ||||||
| No | 9565 (81.13) | 3133 (78.31) | 3362 (82.05) | 3070 (83.17) | ||
| Yes | 2020 (18.87) | 766 (21.69) | 668 (17.95) | 586 (16.83) | ||
| Dietary vitamin E intake (mg/day) | 8.72 (0.08) | 9.48 (0.14) | 8.38 (0.12) | 8.28 (0.13) | ||
| Total vitamin E intake (mg/day) | 9.49 (0.11) | 10.38 (0.20) | 9.12 (0.17) | 8.93 (0.16) | ||
| Fast triglyceride (mg/dL) | 113.81 (1.00) | 56.36 (0.33) | 100.28 (0.32) | 189.94 (1.64) | ||
| Fast total cholesterol (mg/dL) | 190.74 (0.59) | 176.02 (0.83) | 191.21 (0.83) | 206.02 (0.92) | ||
| HDL-C (mg/dL) | 54.53 (0.25) | 62.79 (0.41) | 54.43 (0.32) | 45.79 (0.31) | ||
| LDL-C (mg/dL) | 113.45 (0.46) | 101.98 (0.70) | 116.72 (0.68) | 122.24 (0.80) | ||
| DII | 1.42 (0.04) | 1.26 (0.05) | 1.52 (0.05) | 1.48 (0.05) | ||
| Dietary vitamin C intake (g/day) | 78.76 (1.14) | 82.82 (1.99) | 77.01 (1.47) | 76.30 (1.43) | 0.02 | |
| Dietary fiber intake (g/day) | 17.13 (0.16) | 17.72 (0.27) | 16.63 (0.19) | 17.03 (0.21) | 0.001 | |
| Energy intake (kcal/day) | 2095.02 (9.61) | 2096.27 (14.05) | 2076.10 (15.91) | 2114.00 (19.88) | 0.32 | |
Continuous variables were expressed as weighted mean (standard error), categorical variables as counts (weighted percentage).
Abbreviations: n, numbers; PIR, poverty income ratio; BMI, body mass index; DM, diabetes mellitus; CVD, cardiovascular disease; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; DII, dietary inflammatory index; RC, remnant cholesterol. Q1–3 respectively represent the groups divided according to the tertiles of remnant cholesterol.
We weighted the data using sampling weights (WTSA2YR) to represent US citizens nationally. Categorical variables are presented as counts (weighted percentages) and were analyzed using survey-weighted chi-square tests. Continuous variables are displayed as weighted means (standard errors) and were analyzed using weighted one-way ANOVA. Weighted linear regression models were used to analyze the association between vitamin E intake and RC. Three models were developed. Model 1 did not adjust for covariates. Model 2 was adjusted for age, sex, and race. Model 3 was adjusted for age, sex, race, education, marriage status, BMI, PIR, smoking status, alcohol consumption, recreational activity, use of vitamin E supplements, lipid-lowering drug use, DII, total energy intake, vitamin C intake, dietary fiber intake, hypertension, CVD, and DM. We examined the dose-response relationships between dietary and total vitamin E and RC concentrations using a restricted cubic spline (RCS). In addition, we performed subgroup analyses and interaction tests. All analysis steps were performed using the R software package (version 4.2.2, http://www.R-project.org). p-values
The baseline characteristics of all participants are presented in Table 1. The percentage of participants with obesity, comorbid hypertension, DM, and CVD were 37.99%, 37.13%, 15.50%, and 8.89%, respectively. Those with higher levels of RC tended to have the following characteristics compared to those with lower RC: greater prevalence at 40–60 years of age, greater prevalence of male sex, obesity, comorbid hypertension, DM, and CVD; and lower proportions of participation in recreational activities. It showed that dietary vitamin E intake (8.72 mg/day) and total vitamin E intake (9.49 mg/day) are below the estimated average requirement (12 mg/day) [25].
Table 2 shows that dietary vitamin E intake and total vitamin E intake were all negatively associated with RC, with
| Variables | p-value | ||
| Age | |||
| ref | ref | ||
| 40–60 | 4.30 (3.50, 5.10) | ||
| 3.45 (2.62, 4.28) | |||
| Gender | |||
| Female | ref | ref | |
| Male | 2.77 (2.10, 3.45) | ||
| Race | |||
| Non-Hispanic Black | ref | ref | |
| Non-Hispanic White | 5.62 (4.91, 6.33) | ||
| Mexican American | 7.51 (6.33, 8.69) | ||
| Other Hispanic | 5.48 (4.38, 6.58) | ||
| Other Race | 5.56 (4.26, 6.86) | ||
| Education level | |||
| ref | ref | ||
| High School | –1.42 (–2.34, –0.50) | 0.003 | |
| –3.06 (–3.78, –2.33) | |||
| Marital status | |||
| Never married | ref | ref | |
| Married/Living with a partner | 2.78 (1.90, 3.65) | ||
| Widowed/Divorced/Separated | 3.25 (2.18, 4.32) | ||
| PIR | |||
| ref | ref | ||
| 1–3 | 0.10 (–0.78, 0.97) | 0.83 | |
| –0.75 (–1.64, 0.14) | 0.10 | ||
| BMI | |||
| ref | ref | ||
| 25–30 | 5.85 (5.11, 6.60) | ||
| 8.70 (7.98, 9.43) | |||
| Smoke status | |||
| Former | ref | ref | |
| Never | –2.07 (–2.80, –1.34) | ||
| Now | 1.19 (0.09, 2.29) | 0.03 | |
| Alcohol consumption | |||
| No | ref | ref | |
| Yes | –1.05 (–1.68, –0.43) | 0.001 | |
| Lipid-lowering drugs | |||
| No | ref | ref | |
| Yes | 3.74 (2.88, 4.60) | ||
| Recreational activity | |||
| No | ref | ref | |
| Yes | –2.92 (–3.54, –2.31) | ||
| Hypertension | |||
| No | ref | ref | |
| Yes | 5.15 (4.43, 5.87) | ||
| DM | |||
| No | ref | ref | |
| Borderline | 5.37 (4.52, 6.23) | ||
| Yes | 8.20 (7.27, 9.13) | ||
| CVD | |||
| No | ref | ref | |
| Yes | 3.53 (2.39, 4.67) | ||
| Use of vitamin E supplements | |||
| No | ref | ref | |
| Yes | –1.90 (–2.72, –1.09) | ||
| Dietary vitamin E intake | –0.17 (–0.23, –0.12) | ||
| Total vitamin E intake | –0.12 (–0.17, –0.08) | ||
| DII | 0.18 (–0.01, 0.36) | 0.06 | |
| Dietary vitamin C intake | 0.07 (0.06, 0.08) | 0.08 | |
| Dietary fiber intake | –0.02 (–0.05, 0.02) | 0.30 | |
| Energy intake | 0.00 (0.00,0.00) | 0.14 | |
Abbreviations: RC, remnant cholesterol; PIR, poverty income ratio; BMI, body mass index; DM, diabetes mellitus; CVD, cardiovascular disease; DII, dietary inflammatory index; ref, reference.
Table 3 displays the multivariate linear regression relationship between vitamin E intake and RC. In Model 3, the
| Model 1 | Model 2 | Model 3 | |||||
| p | p | p | |||||
| Dietary vitamin E intake | –0.17 (–0.23, –0.12) | –0.24 (–0.30, –0.18) | –0.21 (–0.29, –0.12) | ||||
| Categories | |||||||
| Q1 [0.06, 4.84] | ref | ref | ref | ref | ref | ref | |
| Q2 (4.84, 7.06] | –0.41 (–1.16, 0.35) | 0.29 | –0.88 (–1.61, –0.14) | 0.02 | –0.69 (–1.40, 0.02) | 0.06 | |
| Q3 (7.06, 10.1] | –1.40 (–2.34, –0.45) | 0.004 | –2.03 (–2.93, –1.13) | –1.97 (–2.94, –1.00) | |||
| Q4 (10.1, 77.88] | –2.20 (–3.15, –1.24) | –3.39 (–4.30, –2.48) | –3.24 (–4.47, –2.02) | ||||
| p for trend | |||||||
| Total vitamin E intake | –0.12 (–0.17, –0.08) | –0.16 (–0.20, –0.12) | –0.12 (–0.18, –0.06) | ||||
| Categories | |||||||
| Q1 [0.06, 4.88] | ref | ref | ref | ref | ref | ref | |
| Q2 (4.88, 7.17] | –0.27 (–1.02, 0.47) | 0.47 | –0.76 (–1.49, –0.03) | 0.04 | –0.56 (–1.26, 0.14) | 0.12 | |
| Q3 (7.17, 10.46] | –1.53 (–2.49, –0.57) | 0.002 | –2.14 (–3.04, –1.25) | –2.02 (–2.98, –1.06) | |||
| Q4 (10.46, 136.69] | –2.12 (–3.10, –1.29) | –3.31 (–4.18, –2.45) | –3.07 (–4.23, –1.91) | ||||
| p for trend | |||||||
Model 1, variables unadjusted;
Model 2, gender, age, and race were adjusted;
Model 3, gender, age, race, education, marriage, BMI, PIR, smoking status, alcohol consumption, recreational activity, use of vitamin E supplements, lipid-lowering drug use, DII, total energy intake, dietary fiber intake, vitamin C intake, hypertension, CVD, and DM were adjusted.
95% CI, 95% confidence interval; RC, remnant cholesterol; PIR, poverty income ratio; BMI, body mass index; DII, dietary inflammatory index; CVD, cardiovascular disease; DM, diabetes mellitus; Q1–4 respectively represent the groups divided according to the quartiles of vitamin E intake.
After adjusting for all potential variables, nonlinear inverse relationships with RC were found to exist in dietary vitamin E intake (Fig. 2A), total vitamin E intake (Fig. 2B), and different subgroups (Fig. 3). There was no interaction between the subgroups except for the “use of vitamin E supplements” subgroup.
Fig. 2. Restricted cubic spline curves of both dietary vitamin E intake (A) and total vitamin E intake (B) with remnant cholesterol (n =11,585). Adjusted for gender, age, race, education, marriage status, BMI, PIR, smoking status, alcohol consumption, recreational activity, use of vitamin E supplements, lipid-lowering drug use, DII, total energy intake, dietary fiber intake, vitamin C intake, hypertension, CVD, and DM. Abbreviations: 95% CI, 95% confidence interval; BMI, body mass index; PIR, Ratio of family income to poverty; DM, diabetes mellitus; DII, dietary inflammatory index; CVD, cardiovascular disease.
Fig. 3. Stratified analyses between dietary vitamin E intake (A) and total vitamin E intake (B) and remnant cholesterol (n = 11,585). Adjusted for gender, age, race, education, marriage status, BMI, PIR, smoking status, alcohol consumption, recreational activity, use of vitamin E supplements, lipid-lowering drug use, DII, total energy intake, dietary fiber intake, vitamin C intake, hypertension, CVD, and DM except for the stratification variable. Abbreviations: 95% CI, 95% confidence interval; BMI, body mass index; PIR, Ratio of family income to poverty; DII, dietary inflammatory index; CVD, cardiovascular disease; DM, diabetes mellitus.
In this cross-sectional study, we assessed the relationship between vitamin E intake and RC. Our findings revealed that vitamin E deficiencies are prevalent among adults in the US aged 20 years or older. The average dietary (8.72 mg/day) and total vitamin E intakes (9.49 mg/day) were lower than the recommended intake (15 mg/day) for adults in the US [26]. Previous studies have yielded similar conclusions. A micronutrient intake survey of 26,282 adults (
The possible mechanisms by which
Vitamin E has a protective association with new-onset hypertension (reverse J-shaped association) and type 2 diabetes (by reducing insulin resistance) [36, 37]. Moreover, in vivo and in vitro studies have confirmed that vitamin E can reduce the risk of CVD and mortality [20, 38, 39], by lowering serum cholesterol and triglyceride levels [40]. Several studies have shown that
This study has several strengths. In this epidemiologic survey, we aimed to assess the association between vitamin E intake and RC with a large sample size, multiple adjusted variables, and weighted analyses representing the national US populations. Consuming more vitamin E may help reduce RC and cardiovascular risks. Nonetheless, this study had several limitations. First, the cross-sectional design does not yield causal relationships; therefore, further longitudinal studies are required to verify these findings. Second, dietary data were self-reported and might have been subject to recall bias. Third, the population of this study was adults aged 20 years or older of the US, and it is not possible to generalize the results to other age groups or populations outside of the US. Further studies are necessary to validate our findings and explore the potential relationships between vitamin E intake and RC using more comprehensive designs.
The 2020–2025 Dietary Guidelines in the US recommend that nutritional needs should be met primarily through foods and beverages [49]. We recommend following the current nutritional guidelines to obtain adequate vitamin E from food to reduce RC levels. Our findings can be significant for developing public health policies and treatment practices.
This cross-sectional study revealed a protective association between vitamin E intake and serum RC. We speculate that vitamin E-rich diets can be used as part of dietary management and may contribute to lowering serum RC concentrations, thereby helping to control the risk of CVD. Furthermore, longitudinal studies should be conducted to verify causality and observe the effects of dietary vitamin E intervention.
Original contributions are included in the article/supplementary material, which can be further queried by the corresponding author. This data can be found at: https://wwwn.cdc.gov/nchs/nhanes/.
SZ and HY designed the research. SZ performed the research. YX provided help and advice on the study. JC and YS analyzed the data. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.
NHANES was approved by the National Center for Health Statistics (NCHS). Ethics Review Board approved the study protocol (Continuation of Protocol #2005-06; Protocol #2011-17; Continuation of Protocol #2011-17; Protocol #2018-01). All participants provided written informed consent. All methods were carried out in accordance with relevant guidelines and regulations (Declarations of Helsinki).
Not applicable.
This research was funded by the Hunan Natural Science Foundation (2023JJ60403).
The authors declare no conflict of interest.
Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/IJVNR26882.
References
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.



