Abstract
Background
Whether iron intake can affect cardiovascular disease (CVD) and dyslipidemia is controversial. However, few studies have focused on reducing the risk of CVD in people at risk for dyslipidemia. This study explored the linear relationship and possible nonlinear relationship between CVD and dyslipidemia.
Methods
Dietary data were obtained from the China Health and Nutrition Survey between 2004 and 2015. The survey included 8173 participants older than 18 years. CVD risk was estimated by the Framingham risk score (FRS). Logistic regression analysis was used to determine whether iron intake affects CVD incidence and lipid profiles. The nonlinear association was tested with restricted cubic splines (RCSs).
Results
For males, higher total iron intake [the fifth quintile (Q) vs. Q1 odds ratio (OR): 0.335, 95% confidence interval (CI): 0.248–0.453], heme iron intake (OR: 0.679, 95% CI: 0.492–0.937) and non-heme iron intake (OR: 0.362, 95% CI: 0.266–0.492) reduced CVD incidence. Heme iron intake increased high low-density lipoprotein cholesterol (LDL-C) (OR: 1.786, 95% CI: 1.226–2.602), high total cholesterol (TC) (OR: 2.404, 95% CI: 1.575–3.669), high triglyceride (TG) (OR: 1.895, 95% CI: 1.423–2.523), and low apolipoprotein A1/apolipoprotein B (ApoA-1/ApoB) risk (OR: 1.514, 95% CI: 1.178–1.945). Moderate non-heme iron intake reduced high-density lipoprotein cholesterol (HDL-C) incidence (Q5 vs. Q1 OR: 0.704, 95% CI: 0.507–0.979). For females, higher total iron intake (Q5 vs. Q1 OR: 0.362, 95% CI: 0.266–0.492) and non-heme iron intake (OR: 0.347, 95% CI: 0.154–0.781) reduced CVD incidence. Heme iron intake increased high LDL-C (OR: 1.587, 95% CI: 1.160–2.170) and high TC incidence (OR: 1.655, 95% CI: 1.187–2.309).
Conclusions
Men, especially those at risk of developing dyslipidemia, should consume non-heme rather than heme iron to reduce CVD incidence. For women, increased heme iron intake did not reduce CVD incidence. Therefore, women should minimize their heme iron intake to prevent dyslipidemia.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12944-024-02058-4.
Keywords: Iron intake, Cardiovascular disease risk, Restricted cubic spline, Dyslipidemia
Background
Iron is involved in the formation of hemoglobin and many other functions. However, excessive intake of iron may pose some health risks. Free iron can destroy cellular macromolecules through the Fenton and Haber-Weiss reactions and promote cell death and tissue damage [1]. However, whether iron intake can affect cardiovascular disease (CVD) and dyslipidemia remains inconclusive. A Swedish study showed that consuming more heme iron promoted fatal acute myocardial infarction [2]. Women who consumed more heme iron were eager to develop coronary heart disease [3]. Chen et al. [4] showed that moderate dietary iron intake can prevent nonfatal CVD. However, these studies did not reach a uniform conclusion. In addition, these studies examined only individual cardiovascular diseases and end events and did not cover other related cardiovascular diseases. Therefore, further exploration is warranted.
Dyslipidemia is essential for the development of CVD [5]. However, whether iron intake can influence dyslipidemia is still controversial. A study from Brazil showed that eating more heme iron increased triglyceride levels [6]. A study assessing health and nutritional status showed that iron intake increased triglycerides in women [7]. In addition, few studies have examined whether iron intake influences lipoprotein (a) (Lp(a)) levels and apolipoprotein A1 (ApoA-1)/ apolipoprotein B (ApoB) levels. Therefore, it is necessary to explore whether iron intake influences different lipid markers, including ApoA-1/ApoB and Lp(a), and to explore possible nonlinear relationships.
People in Western countries tend to have a diet centered around meat consumption, while people in China tend to have a predominantly plant diet. Research has indicated that heme iron contributes to 4% of total iron intake in China [8], and 10–15% in Western countries [9]. However, how iron intake affects blood lipids in China remains unknown. In addition, how to take into account the risk of other diseases while reducing the risk of CVD is also a problem that needs to be solved, but few studies have explored this topic. This study is the first to investigate whether iron intake can affect CVD risk and dyslipidemia in China and explore possible nonlinear relationships to prevent the development of CVD and dyslipidemia.
Methods
Study population
Data from the China Health and Nutrition Examination Survey (CHNS) were collected. The CHNS is a cohort study in China aimed at investigating nutritional status [10]. The study population came from across China. A multistage random cluster sampling method was used to extract samples.
Due to changes in Chinese food coding before and after 2004, data from only 2004 to 2015 were analyzed in this study. This study involved 27,780 participants. After excluding 5218 participants under 18 years old, 13,687 participants without blood test data, 16 participants without dietary intake data, 649 participants without blood pressure test data, and 38 participants lacking lipid data, a total of 8173 participants (3780 males and 4393 females) were ultimately analyzed. (Fig. 1). A comparative analysis of the general information between the included and excluded adult participants was also conducted to uncover and minimize selection bias (Supplementary Materials, Table S1).
Fig. 1.
Flowchart of the participant selection process
Laboratory data
Total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) were determined via enzymatic methods. ApoB, ApoA-1, and Lp(a) were determined by immunoturbidimetry (Randox, UK). The soluble transferrin receptor (TRFR) and transferrin (TRF) were determined via nephelometry (Siemens, Germany). ferritin (FER) levels were determined by radioimmunology (Bio-Tech, China). The details of the laboratory analysis are reported in CHNS [11].
Assessment of dietary intake
Diet assessments were conducted using weighed food stocks and three consecutive 24-hour meal recalls. A study assessed the accuracy of 24-hour meal recalls compared to food inventory weights and the difference was 1% (74 kcal/day) [12]. Another study evaluated 24-hour dietary recall and revealed no significant differences with weighed dietary records, indicating good consistency [13]. Heme iron accounts for approximately 40% of the iron in meat [14], including fish, poultry and livestock. After logarithmic conversion, they were adjusted for energy by residual method [15].
Assessment of CVD risk
The Framingham risk score (FRS) was used to estimate the risk of developing CVD [16]. The FRS was calculated based on sex, age, smoking status, TC, HDL-C, systolic blood pressure (hypertension treatment and systolic blood pressure values), and diabetes status, and ≥ 20% was considered high risk [17, 18].
Definitions of dyslipidemia
High TC levels were defined as TC ≥ 6.2 mmol/L (240 mg/dL). High TG levels were defined as TG ≥ 2.3 mmol/mol (200 mg/dL). Low HDL-C levels were defined as HDL-C < 1.0 mmol/L (40 mg/dL); high LDL-C levels were defined as LDL-C ≥ 4.1 mmol/L (160 mg/dL). Low ApoA-1/ApoB levels were defined as ApoA-1/ApoB < 1, and high Lp (a) levels were defined as Lp (a) > 300 mg/L. The presence of any abnormality in lipid biomarkers is considered to indicate dyslipidemia [19].
Assessment of covariates
General and lifestyle information, including age, sex, residential area, educational level, smoking status, alcohol consumption, and occupation (farmer, worker, other), was collected using a structured questionnaire. Hypertension and diabetes were self-reported diagnoses. Blood pressure was averaged over three measurements. Body mass index (BMI) was calculated as weight (kg) divided by the square of height (m).
Statistical analyses
The differences between groups were compared by ANOVA, the Mann‒Whitney U test, or the chi-square test. A logistic regression model was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs). Model 1 was unadjusted. Model 2 was adjusted for BMI, alcohol status, energy intake, urban residence, and education level. Model 3 was further adjusted for serum ferritin, transferrin, and transferrin receptor levels. For the dyslipidemia risk model, Model 1 was unadjusted, and the adjusted model accounted for age, BMI, alcohol status, smoking status, energy intake, urban residence, education level, and serum ferritin, transferrin, and transferrin receptor levels. All covariates were known potential risk factors for CVD incidence, dyslipidemia or diet, and biochemical confounders. There is no apparent multicollinearity. The rms packet was used to fit the restricted cubic spline (RCS). Harrell suggested that the model fit well with four knots, striking a balance between the smoothness of the curve and avoiding overfitting that may result in reduced accuracy [20]. The 5th, 35th, 65th, and 95th percentiles were used as knots. Sensitivity analysis was performed by excluding patients with hypertension and diabetes, patients diagnosed with myocardial infarction, or patients with a BMI < 18 kg/m2 at baseline. The information on missing values is presented in the Supplementary Materials, Table S2. Missing values were imputed using the median. R software (version 4.1.0) was used for statistical analysis. P < 0.05 was considered significant.
Results
Baseline characteristics
This study included 8173 participants (Table 1, Supplementary Materials, Table S3 and Table S4). The mean age was 47.3 years at baseline. A total of 46.2% were males. The average BMI was 22.1 kg/m2. People who consumed more heme iron were mostly men, were younger, had lower blood pressure, had higher education levels, were more urban residents, had a greater history of alcohol consumption, and were fewer farmers and more workers at baseline. People who consumed more non-heme iron were younger, had lower blood pressure, had a greater history of alcohol consumption; were fewer urban residents, had lower education levels, and had a greater proportion of farmers. As heme iron intake increased, fat intake increased and carbohydrate intake decreased while serum ferritin increased, transferrin decreased, and TC and TG levels increased. As non-heme iron intake increased, energy, carbohydrate, and protein intake increased, while serum ferritin increased, transferrin decreased and LDL-C decreased.
Table 1.
Population characteristics by quintiles of dietary heme and non-heme iron intake
| Variables* | Quintiles of heme iron intake,mg/day | Quintiles of non-heme iron intake,mg/day | ||||||
|---|---|---|---|---|---|---|---|---|
| Q1 | Q3 | Q5 | P | Q1 | Q3 | Q5 | P | |
| Male, n(%) | 658(40.2) | 709(43.4) | 904(55.3) | < 0.001 | 469(28.7) | 784(48.0) | 1003(61.3) | < 0.001 |
| Age, years | 49.0(38.0,60.0) | 47(37.0,58.0) | 46(35.5,54.0) | < 0.001 | 52.0(39.0,65.0) | 47.0(37.0,56.0) | 45.0(37.0,54.0) | < 0.001 |
| Body mass index, kg/m2 | 21.8(20.1,24.2) | 21.7(20.0,24.0) | 21.5(19.8,24.0) | < 0.001 | 21.9(20.0,24.4) | 21.8(20.0,23.9) | 21.6(19.9,23.7) | 0.103 |
| Systolic BP, mmHg | 116.7(105.0,124,7) | 112.0(104.7,122.0) | 110.7(104.0,120.7) | < 0.001 | 118.0(106.7,129.3) | 112.0(104.5,121.7) | 111.3(104.7,120.7) | < 0.001 |
| Diastolic BP, mmHg | 76.0(70.0,80.5) | 75.0(68.7,80.0) | 74.0(69.3,80.0) | < 0.001 | 76.7(70.0,82.0) | 75.0(69.3,80.0) | 75.0(70.0,80.0) | < 0.001 |
| Smoke, n(%) | 355(21.7) | 320(19.6) | 356(21.8) | 0.440 | 298(18.2) | 381(23.3) | 361(22.1) | 0.003 |
| Drinking Alcohol, n(%) | 191(11.7) | 256(15.7) | 301(18.4) | < 0.001 | 192(11.7) | 271(16.6) | 280(17.1) | < 0.001 |
| Diabetes, n(%) | 10(0.6) | 18(1.1) | 16(1.0) | 0.244 | 20(1.2) | 16(1.0) | 12(0.7) | 0.143 |
| Hypertension, n(%) | 73(4.5) | 72(4.4) | 64(3.9) | 0.690 | 95(5.8) | 46(2.8) | 52(3.2) | < 0.001 |
| Urban Residence, n(%) | 260(15.9) | 551(33.7) | 775(47.4) | < 0.001 | 577(35.3) | 547(33.5) | 482(29.5) | 0.005 |
| Education level, n(%) | < 0.001 | < 0.001 | ||||||
| Primary school or lower | 1087(66.5) | 829(50.7) | 635(38.8) | 889(54.4) | 828(50.6) | 763(46.7) | ||
| Middle school | 511(31.3) | 719(44.0) | 904(55.3) | 661(40.4) | 721(44.1) | 802(49.1) | ||
| college or above | 9(0.6) | 55(3.4) | 77(4.7) | 55(3.4) | 61(3.7) | 47(2.9) | ||
| Occupation, n(%) | < 0.001 | < 0.001 | ||||||
| Farmer | 982(60.1) | 704(43.1) | 463(28.3) | 561(34.3) | 718(43.9) | 791(48.4) | ||
| Worker | 181(11.1) | 245(15.0) | 380(23.2) | 292(17.9) | 264(16.1) | 275(16.8) | ||
| Other | 472(28.9) | 686(42.0) | 792(48.4) | 782(47.8) | 653(39.9) | 569(34.8) | ||
| Dietary intake | ||||||||
| Heme iron, mg/day | 0.1(0.0,0.2) | 0.8(0.7,0.9) | 2.0(1.7,2.6) | < 0.001 | 0.6(0.3,0.9) | 0.8(0.4,1.4) | 1.0(0.4,1.6) | < 0.001 |
| Nonheme iron, mg/day | 17.3(14.3,21.5) | 17.9(14.9,21.3) | 20.2(17.4,24.1) | < 0.001 | 12.8(11.5,13.7) | 18.2(17.6,18.9) | 26.8(24.7,30.7) | < 0.001 |
| Iron, mg/day | 17.5(14.4,21.6) | 18.7(15.7,22.1) | 22.5(19.5,26.6) | < 0.001 | 13.5(12.0,14.4) | 19.2(18.4,19.9) | 28.2(25.8,32.2) | < 0.001 |
| Energy, Kcal/day | 2473.2(1925.1,3019.0) | 2347.2(1918.7,2810.2) | 2387.3(2012.0,2864.4) | < 0.001 | 2097.1(1648.5,2544.9) | 2406.3(2019.2,2866.7) | 2677.3(2199.9,3184.6) | < 0.001 |
| Fat, g/day | 41.5(28.1,60.7) | 61.8(42.7,88.5) | 74.9(51.4,100.6) | < 0.001 | 56.9(38.1,81.6) | 60.1(39.4,88.6) | 59.7(39.4,90.0) | 0.001 |
| Carbohydrate, g/day | 427.9(324.1,558.1) | 361.0(273.9,446.1) | 337.2(269.7,421.1) | < 0.001 | 316.5(232.9,404.6) | 370.4(294.4,465.3) | 420.8(327.5,535.3) | < 0.001 |
| Protein, g/day | 69.2(53.0,88.1) | 67.4(54.3,83.3) | 74.7(62.2,91.1) | < 0.001 | 60.2(47.4,75.3) | 71.4(57.7,86.1) | 77.9(64.1,96.4) | < 0.001 |
| Biomarkers | ||||||||
| Ferritin, ng/ml | 74.0(38.0,133.6) | 75.1(37.7,143.0) | 94.9(46.4,171.1) | < 0.001 | 69.3(32.5,128.0) | 81.0(43.3,146.9) | 89.1(45.9,155.8) | < 0.001 |
| Transferrin, mg/dl | 287.5(257.0,324.0) | 282.0(248.0,318.0) | 280.0(250.8,314.0) | < 0.001 | 279.0(250.0,316.0) | 282.0(252.0,316.0) | 283.0(252.0,320.0) | 0.152 |
| Transferrin receptor, mg/L | 1.3(1.1,1.6) | 1.4(1.1,1.7) | 1.3(1.1,1.6) | 0.082 | 1.4(1.1,1.7) | 1.4(1.1,1.6) | 1.3(1.1,1.6) | < 0.001 |
| LDL-C, mmol/L | 2.9(2.3,3.5) | 2.9(2.3,3.5) | 3.0(2.4,3.6) | 0.059 | 3.0(2.3,3.7) | 2.9(2.3,3.5) | 2.9(2.4,3.5) | < 0.001 |
| HDL-C, mmol/L | 1.4(1.2,1.6) | 1.4(1.2,1.6) | 1.4(1.1,1.6) | < 0.001 | 1.4(1.2,1.7) | 1.4(1.1,1.6) | 1.4(1.2,1.7) | 0.006 |
| Total cholesterol, mmol/L | 4.7(4.1,5.4) | 4.8(4.2,5.4) | 4.9(4.3,5.6) | < 0.001 | 4.8(4.2,5.5) | 4.8(4.2,5.5) | 4.8(4.2,5.4) | 0.025 |
| Triglycerides, mmol/L | 1.2(0.8,1.9) | 1.2(0.8,2.0) | 1.3(0.9,2.1) | < 0.001 | 1.3(0.9,1.9) | 1.3(0.9,2.0) | 1.2(0.8,1.9) | 0.109 |
| Apolipoprotein A1, g/L | 1.07(0.93,1.24) | 1.09(0.94,1.28) | 1.12(0.96,1.35) | < 0.001 | 1.10(0.95,1.31) | 1.10(0.95,1.29) | 1.09(0.93,1.28) | 0.009 |
| Apolipoprotein B, g/L | 0.85(0.71,1.03) | 0.88(0.73,1.06) | 0.92(0.76,1.12) | < 0.001 | 0.90(0.74,1.09) | 0.88(0.72,1.07) | 0.88(0.73,1.07) | 0.021 |
| Lipoprotein (a), mg/L | 92.0(47.0,190.0) | 76.0(41.0,170.0) | 68.0(33.0,147.0) | < 0.001 | 85.0(45.0,180.5) | 83.0(41.0,174.5) | 72.0(38.0,151.0) | < 0.001 |
*Variables are presented as the mean (SD), median (IQR) or n (%)
Associations between iron intake and high CVD risk
For males, higher total iron intake (the fifth quintile(Q) vs. Q1 OR: 0.335, 95% CI: 0.248–0.453, Table 2), heme iron intake (OR: 0.679, 95% CI: 0.492–0.937) and non-heme iron intake (OR: 0.362, 95% CI: 0.266–0.492) reduced the risk of CVD. The RCS analysis revealed an approximately inverse J-shaped relationship between total iron intake (P for nonlinearity < 0.001, Fig. 2A) and non-heme iron intake (P for nonlinearity < 0.001) and CVD risk, while a U-shaped relationship was shown for heme iron intake (P for nonlinearity = 0.031).
Table 2.
Association between iron intake and high cardiovascular disease risk in men
| Model 1 | Model 2 | Model 3 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Quintiles of total iron intake | OR | 95%CI | P | OR | 95%CI | P | OR | 95%CI | P |
| Q1 | Ref | Ref | Ref | ||||||
| Q2 | 0.541 | 0.426–0.688 | < 0.001 | 0.564 | 0.438–0.728 | < 0.001 | 0.569 | 0.441–0.736 | < 0.001 |
| Q3 | 0.347 | 0.367–0.451 | < 0.001 | 0.381 | 0.289–0.503 | < 0.001 | 0.382 | 0.288–0.505 | < 0.001 |
| Q4 | 0.250 | 0.188–0.332 | < 0.001 | 0.251 | 0.185–0.340 | < 0.001 | 0.259 | 0.190–0.352 | < 0.001 |
| Q5 | 0.284 | 0.215–0.374 | < 0.001 | 0.324 | 0.241–0.436 | < 0.001 | 0.335 | 0.248–0.453 | < 0.001 |
| Quintiles of heme iron intake | |||||||||
| Q1 | Ref | Ref | Ref | ||||||
| Q2 | 1.139 | 0.880–1.474 | 0.324 | 1.057 | 0.800-1.395 | 0.698 | 1.001 | 0.756–1.327 | 0.993 |
| Q3 | 0.955 | 0.733–1.245 | 0.735 | 0.918 | 0.690–1.221 | 0.558 | 0.854 | 0.640–1.141 | 0.286 |
| Q4 | 0.877 | 0.670–1.147 | 0.337 | 0.921 | 0.687–1.234 | 0.580 | 0.846 | 0.628–1.139 | 0.270 |
| Q5 | 0.645 | 0.485–0.859 | 0.003 | 0.732 | 0.534–1.004 | 0.053 | 0.679 | 0.492–0.937 | 0.018 |
| Quintiles of non-heme iron intake | |||||||||
| Q1 | Ref | Ref | Ref | ||||||
| Q2 | 0.541 | 0.425–0.688 | < 0.001 | 0.549 | 0.460–0.768 | < 0.001 | 0.612 | 0.472–0.793 | < 0.001 |
| Q3 | 0.393 | 0.304–0.508 | < 0.001 | 0.420 | 0.319–0.552 | < 0.001 | 0.445 | 0.337–0.588 | < 0.001 |
| Q4 | 0.260 | 0.195–0.346 | < 0.001 | 0.275 | 0.203–0.374 | < 0.001 | 0.297 | 0.218–0.404 | < 0.001 |
| Q5 | 0.279 | 0.211–0.369 | < 0.001 | 0.334 | 0.246–0.453 | < 0.001 | 0.362 | 0.266–0.492 | < 0.001 |
Model 1: Unadjusted. Model 2: Adjusted for BMI, alcohol status, energy intake, urban residence, and education level. Model 3: Adjusted for Model 2 + serum ferritin, transferrin, and transferrin receptor levels. Heme and non-heme iron intakes were adjusted mutually
Fig. 2.
High CVD risk according to iron intake in men (A) and women (B) in Model 3
For females, higher total iron intake (Q5 vs. Q1 OR: 0.362, 95% CI: 0.266–0.492, Table 3) and non-heme iron intake (OR: 0.347, 95% CI: 0.154–0.781) reduced the risk of CVD. The RCS curves demonstrated an approximate L-shaped relationship (P for nonlinearity = 0.028, Fig. 2B).
Table 3.
Association between iron intake and high CVD risk in women
| Model 1 | Model 2 | Model 3 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Quintiles of total iron intake | |||||||||
| Q1 | Ref | Ref | Ref | ||||||
| Q2 | 0.574 | 0.344–0.959 | 0.034 | 0.514 | 0.294–0.899 | 0.020 | 0.541 | 0.308–0.950 | 0.032 |
| Q3 | 0.235 | 0.117–0.473 | < 0.001 | 0.244 | 0.115–0.518 | < 0.001 | 0.244 | 0.115–0.520 | < 0.001 |
| Q4 | 0.283 | 0.148–0.543 | < 0.001 | 0.329 | 0.164–0.663 | 0.002 | 0.355 | 0.176–0.715 | 0.004 |
| Q5 | 0.211 | 0.102–0.438 | < 0.001 | 0.274 | 0.124–0.609 | 0.001 | 0.282 | 0.126–0.629 | 0.002 |
| Quintiles of heme iron intake | |||||||||
| Q1 | Ref | Ref | Ref | ||||||
| Q2 | 0.912 | 0.501-1,660 | 0.763 | 0.787 | 0.410–1.513 | 0.473 | 0.777 | 0.403–1.499 | 0.452 |
| Q3 | 1.090 | 0.614–1.935 | 0.770 | 1.213 | 0.652–2.257 | 0.542 | 1.117 | 0.596–2.092 | 0.730 |
| Q4 | 0.778 | 0.417–1.454 | 0.433 | 0.870 | 0.439–1.726 | 0.690 | 0.823 | 0.413–1.642 | 0.581 |
| Q5 | 0.385 | 0.177–0.837 | 0.016 | 0.537 | 0.232–1.244 | 0.147 | 0.498 | 0.214–1.159 | 0.106 |
| Quintiles of non-heme iron intake | |||||||||
| Q1 | Ref | Ref | Ref | ||||||
| Q2 | 0.702 | 0.425–1.161 | 0.168 | 0.687 | 0.396–1.191 | 0.181 | 0.716 | 0.411–1.247 | 0.237 |
| Q3 | 0.127 | 0.050–0.323 | < 0.001 | 0.130 | 0.045–0.374 | < 0.001 | 0.134 | 0.047–0.388 | < 0.001 |
| Q4 | 0.410 | 0.227–0.742 | 0.003 | 0.578 | 0.305–1.093 | 0.092 | 0.641 | 0.337–1.219 | 0.175 |
| Q5 | 0.255 | 0.126–0.514 | < 0.001 | 0.333 | 0.149–0.746 | 0.008 | 0.347 | 0.154–0.781 | 0.011 |
Model 1: Unadjusted. Model 2: Adjusted for BMI, alcohol status, energy intake, urban residence, and education level. Model 3: Adjusted for Model 2 + serum ferritin, transferrin, and transferrin receptor levels. Heme and non-heme iron intakes were adjusted mutually
Association between iron intake and risk of developing dyslipidemia in men
More heme iron intake increased the high LDL-C levels (Q5 vs. Q1 OR: 1.786, 95% CI: 1.226–2.602, Table 4), high TC levels (OR: 2.404, 95% CI: 1.575–3.669), high TG levels (OR: 1.895, 95% CI: 1.423–2.523), and low ApoA-1/ApoB levels risk (OR: 1.514, 95% CI: 1.178–1.945). Moderate non-heme iron intake reduced low HDL-C levels (Q5 vs. Q1 OR: 0.704, 95% CI: 0.507–0.979), low ApoA-1/ApoB levels (Q4 vs. Q1 OR: 0.742, 95% CI: 0.742) and high Lp (a) levels (Q3 vs. Q1 OR: 0.657, 95% CI: 0.470–0.917). Heme iron intake nonlinearly influenced high LDL-C levels (P for nonlinearity = 0.012, Fig. 3A), high TC levels (P for nonlinearity < 0.001), high TG levels (P for nonlinearity < 0.001), and low apoA-1/apoB levels (P for nonlinearity = 0.040), resembling an inverted L-shape. Non-heme iron intake nonlinearly influenced high LDL-C levels (P for nonlinearity = 0.007, Fig. 3B) and high TG levels (P for nonlinearity = 0.005).
Table 4.
Association between iron intake and risk of developing dyslipidemia in men
| Dyslipidemia Type | heme iron intake | non-heme iron intake | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Model 1 | Adjusted model | Model 1 | Adjusted model | |||||||||
| High LDL-C | OR | 95%CI | P | OR | 95%CI | P | OR | 95%CI | P | OR | 95%CI | P |
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.223 | 0.842–1.775 | 0.290 | 1.194 | 0.815–1.750 | 0.362 | 1.028 | 0.743–1.421 | 0.869 | 1.115 | 0.797–1.560 | 0.526 |
| Q3 | 1.280 | 0.885–1.952 | 0.190 | 1.234 | 0.841–1.810 | 0.283 | 0.946 | 0.681–1.316 | 0.743 | 1.020 | 0.721–1.443 | 0.909 |
| Q4 | 1.593 | 1.116–2.275 | 0.010 | 1.663 | 1.145–2.416 | 0.008 | 0.680 | 0.476–0.969 | 0.033 | 0.702 | 0.482–1.024 | 0.067 |
| Q5 | 1.722 | 1.212–2.449 | 0.002 | 1.786 | 1.226–2.602 | 0.003 | 0.798 | 0.567–1.124 | 0.196 | 0.815 | 0.559–1.189 | 0.288 |
| Low HDL-C | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.114 | 0.822–1.509 | 0.486 | 1.030 | 0.749–1.416 | 0.855 | 1.000 | 0.754–1.325 | 1.000 | 0.977 | 0.728–1.312 | 0.877 |
| Q3 | 1.140 | 0.842–1.542 | 0.397 | 1.078 | 0.785–1.481 | 0.641 | 0.990 | 0.746–1.312 | 0.943 | 0.914 | 0.676–1.234 | 0.556 |
| Q4 | 1.284 | 0.955–1.728 | 0.098 | 1.183 | 0.862–1.623 | 0.298 | 0.878 | 0.658–1.172 | 0.378 | 0.795 | 0.582–1.085 | 0.148 |
| Q5 | 1.392 | 1.039–1.866 | 0.027 | 1.295 | 0.942–1.779 | 0.112 | 0.704 | 0.521–0.952 | 0.023 | 0.704 | 0.507–0.979 | 0.037 |
| High TC | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.395 | 0.920–2.115 | 0.117 | 1.538 | 0.999–2.366 | 0.050 | 0.966 | 0.671–1.391 | 0.852 | 0.966 | 0.659–1.416 | 0.860 |
| Q3 | 1.696 | 1.134–2.536 | 0.010 | 1.904 | 1.248–2.905 | 0.003 | 0.983 | 0.683–1.414 | 0.926 | 0.968 | 0.659–1.422 | 0.869 |
| Q4 | 1.696 | 1.134–2.536 | 0.010 | 2.106 | 1.375–3.226 | < 0.001 | 0.915 | 0.633–1.324 | 0.638 | 0.806 | 0.541–1.201 | 0.290 |
| Q5 | 2.035 | 1.376–3.011 | < 0.001 | 2.404 | 1.575–3.669 | < 0.001 | 0.966 | 0.671–1.391 | 0.852 | 0.869 | 0.579–1.304 | 0.497 |
| High TG | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.047 | 0.803–1.365 | 0.735 | 1.072 | 0.800−1.435 | 0.642 | 1.256 | 0.982–1.606 | 0.070 | 1.142 | 0.871–1.498 | 0.338 |
| Q3 | 1.365 | 1.058–1.763 | 0.017 | 1.570 | 1.182–2.087 | 0.002 | 1.312 | 1.027–1.676 | 0.030 | 1.101 | 0.837–1.448 | 0.490 |
| Q4 | 1.472 | 1.143–1.896 | 0.003 | 1.746 | 1.314–2.322 | < 0.001 | 1.068 | 0.831–1.374 | 0.608 | 0.794 | 0.597–1.056 | 0.113 |
| Q5 | 1.780 | 1.389–2.280 | < 0.001 | 1.895 | 1.423–2.523 | < 0.001 | 1.008 | 0.782−1.300 | 0.948 | 0.787 | 0.587–1.055 | 0.109 |
| Low ApoA-1/ApoB | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.200 | 0.951–1.514 | 0.124 | 1.219 | 0.954–1.557 | 0.113 | 1.006 | 0.806–1.256 | 0.955 | 0.993 | 0.786–1.254 | 0.952 |
| Q3 | 1.257 | 0.998–1.584 | 0.052 | 1.337 | 1.046–1.709 | 0.020 | 0.962 | 0.770–1.202 | 0.733 | 0.915 | 0.721–1.160 | 0.462 |
| Q4 | 1.429 | 1.137–1.795 | 0.002 | 1.563 | 1.222–1.999 | < 0.001 | 0.807 | 0.643–1.013 | 0.064 | 0.742 | 0.580–0.948 | 0.017 |
| Q5 | 1.411 | 1.123–1.774 | 0.003 | 1.514 | 1.178–1.945 | 0.001 | 0.907 | 0.725–1.134 | 0.392 | 0.847 | 0.660–1.085 | 0.189 |
| High L(a) | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 0.886 | 0.652–1.202 | 0.436 | 0.908 | 0.664–1.240 | 0.542 | 1.134 | 0.854–1.507 | 0.385 | 1.204 | 0.899–1.611 | 0.213 |
| Q3 | 0.874 | 0.643–1.188 | 0.390 | 0.869 | 0.633–1.192 | 0.382 | 0.616 | 0.447–0.849 | 0.003 | 0.657 | 0.470–0.917 | 0.013 |
| Q4 | 1.094 | 0.815–1.468 | 0.549 | 1.105 | 0.811–1.505 | 0.527 | 0.818 | 0.605–1.106 | 0.193 | 0.900 | 0.656–1.236 | 0.516 |
| Q5 | 0.818 | 0.600−1.117 | 0.207 | 0.900 | 0.647–1.251 | 0.529 | 0.767 | 0.564–1.041 | 0.089 | 0.810 | 0.584–1.122 | 0.205 |
Model 1: Unadjusted. Adjusted model: Adjusted for age, BMI, alcohol status, smoking status, energy intake, urban residence, education level, serum ferritin, transferrin, and transferrin receptor level. Heme and non-heme iron intakes were adjusted mutually
Fig. 3.
The risk of developing dyslipidemia according to heme (A) and non-heme iron intake (B) in men in the adjusted model
Association between iron intake and the risk of developing dyslipidemia in women
A higher heme iron intake increased both high LDL-C levels (Q5 vs. Q1 OR: 1.587, 95% CI: 1.160–2.170, Table 5) and high TC level risk (OR: 1.655, 95% CI: 1.187–2.309) after adjustment. Heme iron intake nonlinearly influenced high LDL-C levels (P for nonlinearity = 0.003, Fig. 4A), low HDL-C levels (P for nonlinearity = 0.020), high TC levels(P for nonlinearity < 0.001) and high Lp(a) levels (P for nonlinearity < 0.022), and the relationships were similar to an inverted L-shaped curve. Non-heme iron intake was not effective (Fig. 4B).
Table 5.
Association between iron intake and the risk of developing dyslipidemia in women
| Dyslipidemia Type | heme iron intake | non-heme iron intake | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Model 1 | Adjusted model | Model 1 | Adjusted model | |||||||||
| High LDL-C | OR | 95%CI | P | OR | 95%CI | P | OR | 95%CI | P | OR | 95%CI | P |
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.149 | 0.864–1.530 | 0.340 | 1.240 | 0.918–1.675 | 0.160 | 0.901 | 0.686–1.182 | 0.451 | 1.121 | 0.839–1.497 | 0.440 |
| Q3 | 0.978 | 0.728–1.312 | 0.881 | 1.120 | 0.821–1.529 | 0.474 | 0.785 | 0.593–1.038 | 0.089 | 0.970 | 0.719–1.308 | 0.840 |
| Q4 | 1.327 | 1.004–1.754 | 0.047 | 1.664 | 1.231–2.250 | < 0.001 | 0.803 | 0.608–1.061 | 0.122 | 1.113 | 0.825–1.503 | 0.483 |
| Q5 | 1.206 | 0.908–1.601 | 0.195 | 1.587 | 1.160–2.170 | 0.004 | 0.900 | 0.685–1.181 | 0.446 | 1.228 | 0.910–1.658 | 0.179 |
| Low HDL-C | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 0.942 | 0.670–1.325 | 0.733 | 0.846 | 0.594–1.205 | 0.354 | 1.067 | 0.752–1.515 | 0.716 | 1.030 | 0.717–1.481 | 0.871 |
| Q3 | 0.769 | 0.538–1.098 | 0.149 | 0.713 | 0.492–1.034 | 0.075 | 0.935 | 0.652–1.340 | 0.714 | 0.946 | 0.648–1.381 | 0.774 |
| Q4 | 0.812 | 0.571–1.156 | 0.248 | 0.747 | 0.515–1.085 | 0.126 | 0.969 | 0.678–1.384 | 0.861 | 0.998 | 0.684–1.456 | 0.993 |
| Q5 | 0.670 | 0.463–0.970 | 0.034 | 0.674 | 0.454–1.001 | 0.051 | 0.775 | 0.532–1.128 | 0.183 | 0.782 | 0.522–1.170 | 0.232 |
| High TC | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.090 | 0.803–1.478 | 0.581 | 1.235 | 0.896–1.702 | 0.198 | 0.940 | 0.707–1.250 | 0.669 | 1.156 | 0.854–1.566 | 0.347 |
| Q3 | 1.050 | 0.773–1.428 | 0.754 | 1.264 | 0.911–1.752 | 0.160 | 0.770 | 0.572–1.036 | 0.084 | 0.946 | 0.690–1.299 | 0.733 |
| Q4 | 1.365 | 1.018–1.829 | 0.038 | 1.880 | 1.369–2.582 | < 0.001 | 0.839 | 0.627–1.124 | 0.239 | 1.148 | 0.838–1.573 | 0.390 |
| Q5 | 1.152 | 0.852–1.558 | 0.357 | 1.655 | 1.187–2.309 | 0.003 | 0.858 | 0.642–1.147 | 0.302 | 1.097 | 0.797–1.512 | 0.570 |
| High TG | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.049 | 0.822–1.340 | 0.700 | 1.178 | 0.904–1.534 | 0.225 | 0.903 | 0.705–1.156 | 0.418 | 0.876 | 0.670–1.145 | 0.332 |
| Q3 | 0.992 | 0.775–1.270 | 0.950 | 1.203 | 0.919–1.575 | 0.178 | 1.023 | 0.803–1.304 | 0.853 | 1.029 | 0.790–1.340 | 0.832 |
| Q4 | 0.954 | 0.744–1.223 | 0.712 | 1.233 | 0.938–1.620 | 0.134 | 0.895 | 0.699–1.147 | 0.382 | 0.939 | 0.716–1.231 | 0.648 |
| Q5 | 0.846 | 0.656–1.090 | 0.197 | 1.223 | 0.919–1.626 | 0.167 | 0.835 | 0.650–1.074 | 0.160 | 0.811 | 0.612–1.074 | 0.143 |
| Low ApoA-1/ApoB | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 1.212 | 0.968–1.519 | 0.094 | 1.287 | 1.015–1.632 | 0.037 | 0.894 | 0.712–1.121 | 0.331 | 0.987 | 0.776–1.256 | 0.916 |
| Q3 | 1.063 | 0.846–1.337 | 0.599 | 1.181 | 0.925–1.506 | 0.181 | 0.855 | 0.681–1.075 | 0.181 | 0.998 | 0.782–1.274 | 0.990 |
| Q4 | 0.981 | 0.778–1.237 | 0.869 | 1.132 | 0.882–1.453 | 0.320 | 0.887 | 0.707–1.114 | 0.303 | 1.117 | 0.875–1.426 | 0.376 |
| Q5 | 0.952 | 0.754–1.201 | 0.678 | 1.153 | 0.892–1.490 | 0.276 | 0.974 | 0.779–1.219 | 0.819 | 1.225 | 0.957–1.568 | 0.107 |
| High Lp(a) | ||||||||||||
| Q1 | Ref | Ref | Ref | Ref | ||||||||
| Q2 | 0.751 | 0.576–0.978 | 0.033 | 0.760 | 0.580–0.997 | 0.047 | 1.321 | 1.011–1.727 | 0.041 | 1.405 | 1.068–1.850 | 0.015 |
| Q3 | 0.772 | 0.594–1.004 | 0.054 | 0.792 | 0.604–1.039 | 0.092 | 1.178 | 0.897–1.547 | 0.239 | 1.288 | 0.970–1.709 | 0.080 |
| Q4 | 0.850 | 0.657−1.100 | 0.216 | 0.905 | 0.691–1.187 | 0.472 | 1.158 | 0.881–1.522 | 0.292 | 1.316 | 0.989–1.752 | 0.060 |
| Q5 | 0.787 | 0.606–1.023 | 0.073 | 0.866 | 0.654–1.148 | 0.317 | 1.062 | 0.805–1.401 | 0.671 | 1.234 | 0.920–1.657 | 0.160 |
Model 1: Unadjusted. Adjusted model: Adjusted for age, BMI, alcohol status, smoking status, energy intake, urban residence, education level, serum ferritin, transferrin, and transferrin receptor level. Heme and non-heme iron intakes were adjusted mutually
Fig. 4.
The risk of developing dyslipidemia according to heme (A) and non-heme iron intake (B) in women in the adjusted model
Subgroup and sensitivity analyses
Participants were stratified by age (< 65 or ≥ 65 years), BMI (< 25 or ≥ 25 kg/m2), hypertension status (yes or no), and urban residence status (yes or no) (Fig. 5). In men, CVD risk decreased in subgroups with a BMI < 25 kg/m2, individuals without hypertension, and individuals in the rural resident subgroup. In women, non-heme iron intake reduced CVD risk in subgroups with a BMI < 25 kg/m2, no hypertension, and no urban residence, while heme iron intake did not.
Fig. 5.
Subgroup analysis, in Model 3 excluding the subgroup variable itself. The OR was calculated per 1-unit increase in heme or non-heme iron intake
Sensitivity analysis was performed by excluding candidates with hypertension and diabetes (Supplementary Materials, Table S5), patients diagnosed with myocardial infarction (Supplementary Materials, Table S6), or candidates with a BMI < 18 kg/m2 at baseline (possible digestive malabsorption affecting iron digestion and absorption) (Supplementary Materials, Table S7). Intake was categorized based on quintiles, and a comparison was made between the highest quintile and the lowest. There were no significant changes in the observed risk estimates.
Discussion
The outsized impact of CVD is primarily due to its asymptomatic nature during “silent” episodes and its progression to plaque deposits over time as blood vessels become blocked. Previous research has shown that the development of CVD is often accompanied by myocardial energy metabolism disorders [21]. Many clinical guidelines now recommend many alternative biomarkers [22]. The accuracy of the FRS, the most widely used tool [16], has been validated in a variety of populations [17, 18], and it is helpful for predicting CVD incidence and performing clinical primary prevention. For Chinese men, higher iron intake can reduce CVD risk, and the relationship was close to an inverse J-shape. These findings are similar to some previous findings [4]. Non-heme iron diet could reduce CVD risk [23]. In terms of dyslipidemia, a study in Tehran showed that total iron intake promoted high triglycerides [24]. However, other studies have not shown an influence on CVD mortality [25, 26] or lipids [6]. These study populations mostly originated from Western countries. These differences in results may be attributed to variations in ethnicity and diet.
For men, moderate heme iron intake could reduce CVD risk, and the effect was nearly U-shaped. However, higher heme iron intake increased high LDL-C, TC, TG, and low ApoA-1/ApoB risk and the relationship was close to an inverted L-shaped curve. Chen et al. [4] showed that less heme iron intake promoted nonfatal CVD incidence. However, some studies have come to different conclusions. When heme iron intake increases every 7 mg/day, CVD risk increases by 1% [24]. Heme iron intake promoted myocardial infarction but not stroke or CVD mortality [27]. This may be because Chinese individuals consume relatively little heme iron.
The relationships between total, non-heme iron intake and CVD incidence in women were close to L-shaped. Heme iron intake increased high LDL-C and high TC risk but did not reduce cardiovascular risk. This may be because men tend to retain excess iron [28], and women have less iron due to menstruation [29]. Heme iron intake promoted CVD in women. Moderate non-heme iron reduced CVD risk [30].
ApoB and ApoA-1 are the main surface proteins of atherosclerotic lipoproteins and HDL-C. A Swedish study showed that people with CVD had higher LDL-C levels and lower ApoA-1/ApoB ratios [31]. ApoA-1/ApoB is more precise than ApoB or ApoA-1 alone [32]. However, the measured levels of ApoB may not be consistent with those of LDL-C [33]. For individuals whose ApoB is highly inconsistent with LDL-C, the nutrients that contribute most significantly to the diet are total fat, saturated fatty acids, and thiamine [34]. Therefore, it is necessary to evaluate the levels of multiple lipid markers. The risk of a low ApoA-1/ApoB ratio gradually increased with increasing heme iron intake. The serum ferritin concentration increases the ApoB/ApoA-1 ratio [35], similar to this study.
Study strengths and limitations
This study explored the nonlinear relationships of iron intake with CVD incidence and dyslipidemia in the Chinese population and comprehensively analyzed multiple risk factors, providing more reasonable dietary recommendations. Second, this study was rigorous in design and had a large sample size. Third, to minimize confounding effects, the model was adjusted for various factors.
This study has several limitations. First, long-term follow-up data were lacking. The study sample included only Chinese participants. Second, rough estimates of heme iron intake may lead to discrepancies. In addition, the CHNS does not include iron supplements or medications for lipid-lowering or blood control. However, studies have shown that the overall intake of dietary supplements among the Chinese population is relatively low, at approximately 0.71%. The intake of iron supplements is approximately 0.16% [36]. Therefore, the findings may not have changed materially. Third, some socioeconomic and behavioral factors were not considered in the present study, which may have confounded the results.
Conclusions
In the process of preventing CVD risk and dyslipidemia, as well as formulating clinical management strategies, greater attention should be given to dietary iron intake. For men, moderate iron intake reduced CVD, low ApoA-1/ApoB, and high Lp (a) risk. Heme iron intake increased high LDL-C, TC, TG, and low ApoA-1/ApoB risks. Therefore, men, especially those at risk of developing dyslipidemia, should consume non-heme iron to prevent CVD. For women, moderate non-heme rather than heme iron intake can reduce CVD risk. Heme iron intake incresed high LDL-C and TC incidence. Therefore, women should minimize their heme iron intake to reduce dyslipidemia risk.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
We acknowledge CHNS and all the organizations that have contributed to this study.
Abbreviations
- ApoA-1
Apolipoprotein A1
- ApoB
Apolipoprotein B
- BMI
Body mass index
- BP
Blood pressure
- CHNS
China Health and Nutrition Survey
- CI
Confidence interval
- CVD
Cardiovascular disease
- FET
Ferritin
- FRS
Framingham risk score
- HDL-C
High-density lipoprotein cholesterol
- LDL-C
Low-density lipoprotein cholesterol
- Lp(a)
Lipoprotein (a)
- OR
Odds ratio
- RCS
Restricted cubic splines
- TC
Total cholesterol
- TG
Triglyceride
- TRF
Transferrin
- TRFR
Transferrin receptor
Author contributions
X.Q. contributed to the conception and design of the study. M.C.conducted data analysis and wrote the manuscript. M.C., H.W., H.Z., L.W. and X.Q. contributed to analysis and interpretation of the data. All authors have critically reviewed the manuscript for important intellectual content, and given approval of the final version for publication.
Funding
This study was supported by the National Natural Science Foundation of China (52173285), the Tianjin Health Committee (2021028), the Tianjin Health Research Project (No.TJWJ2021QN019) and the Tianjin Union Medical Center Project (2023YJZD003).
Data availability
The data supporting the findings of this study are available from CHNS (https://www.cpc.unc.edu/projects/china/data/datasets).
Declarations
Ethics approval and consent to participate
The CHNS study was approved by the Institutional Review Board at the University of North Carolina and the National Institute of Nutrition and Health, Chinese Center for Disease Control and Prevention. All participants provided written informed consent in the survey.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Mohammadifard N, Humphries KH, Gotay C, Mena-Sánchez G, Salas-Salvadó J, Esmaillzadeh A, et al. Trace minerals intake: risks and benefits for cardiovascular health. Crit Rev Food Sci Nutr. 2019;59(8):1334–46. doi: 10.1080/10408398.2017.1406332. [DOI] [PubMed] [Google Scholar]
- 2.Kaluza J, Larsson SC, Håkansson N, Wolk A. Heme iron intake and acute myocardial infarction: a prospective study of men. Int J Cardiol. 2014;172(1):155–60. doi: 10.1016/j.ijcard.2013.12.176. [DOI] [PubMed] [Google Scholar]
- 3.van der Peeters ADL, Grobbee PH, Marx DE, van der Schouw JJ. Dietary haem iron and coronary heart disease in women. Eur Heart J. 2005;26(3):257–62. doi: 10.1093/eurheartj/ehi027. [DOI] [PubMed] [Google Scholar]
- 4.Chen P, Wu S, He J, Sui Y, Li K, Fang A. Long-term dietary iron intake and risk of non-fatal cardiovascular diseases in the China Health and Nutrition Survey. Eur J Prev Cardiol. 2023;30(18):2032–43. doi: 10.1093/eurjpc/zwad244. [DOI] [PubMed] [Google Scholar]
- 5.Rabizadeh S, Rajab A, Mechanick JI, Moosaie F, Rahimi Y, Nakhjavani M, et al. LDL/Apo B ratio predict coronary heart disease in type 2 diabetes independent of ASCVD risk score: a case-cohort study. Nutr Metab Cardiovasc Dis. 2021;31(5):1477–85. doi: 10.1016/j.numecd.2021.01.013. [DOI] [PubMed] [Google Scholar]
- 6.Dos Santos Vieira DA, Hermes Sales C, Galvão Cesar CL, Marchioni DM, Fisberg RM. Influence of Haem, Non-haem, and Total Iron Intake on metabolic syndrome and its components: a Population-based study. Nutrients. 2018;10(3):314. doi: 10.3390/nu10030314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Li G, Yu W, Yang H, Wang X, Ma T, Luo X. Relationship between Serum Ferritin Level and Dyslipidemia in US adults based on data from the National Health and Nutrition Examination surveys 2017 to 2020. Nutrients. 2023;15(8):1878. doi: 10.3390/nu15081878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.He J, Shen X, Fang A, Song J, Li H, Guo M, et al. Association between predominantly plant-based diets and iron status in Chinese adults: a cross-sectional analysis. Br J Nutr. 2016;116(9):1621–32. doi: 10.1017/S0007114516003639. [DOI] [PubMed] [Google Scholar]
- 9.Hurrell R, Egli I. Iron bioavailability and dietary reference values. Am J Clin Nutr. 2010;91(5):1461S–7. doi: 10.3945/ajcn.2010.28674F. [DOI] [PubMed] [Google Scholar]
- 10.Zhang B, Zhai FY, Du SF, Popkin BM. The China Health and Nutrition Survey, 1989–2011. Obes Rev. 2014;15(Suppl 1):2–7. doi: 10.1111/obr.12119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yan S, Li J, Li S, Zhang B, Du S, Gordon-Larsen P, et al. The expanding burden of cardiometabolic risk in China: the China Health and Nutrition Survey. Obes Rev. 2012;13(9):810–21. doi: 10.1111/j.1467-789X.2012.01016.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zhai F, Guo X, Popkin BM, et al. Evaluation of the 24-Hour Individual Recall Method in China. FoodNutr Bull. 1996;17(2):1–7. [Google Scholar]
- 13.Xue H, Yang M, Liu Y, Duan R, Cheng G, Zhang X. Relative validity of a 2-day 24-hour dietary recall compared with a 2-day weighed dietary record among adults in South China. Nutr Diet. 2017;74(3):298–307. doi: 10.1111/1747-0080.12315. [DOI] [PubMed] [Google Scholar]
- 14.Monsen ER, Hallberg L, Layrisse M, Hegsted DM, Cook JD, Mertz W, et al. Estimation of available dietary iron. Am J Clin Nutr. 1978;31(1):134–41. doi: 10.1093/ajcn/31.1.134. [DOI] [PubMed] [Google Scholar]
- 15.Willett W, Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol. 1986;124(1):17–27. doi: 10.1093/oxfordjournals.aje.a114366. [DOI] [PubMed] [Google Scholar]
- 16.D’Agostino RB, Sr, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743–53. doi: 10.1161/CIRCULATIONAHA.107.699579. [DOI] [PubMed] [Google Scholar]
- 17.Bi M, Feng A, Liu Y, Tian S. U-shaped association of serum uric acid with cardiovascular disease risk scores and the modifying role of sex among Chinese adults. Nutr Metab Cardiovasc Dis. 2023;33(5):1066–76. doi: 10.1016/j.numecd.2023.02.025. [DOI] [PubMed] [Google Scholar]
- 18.Bosomworth NJ. Practical use of the Framingham risk score in primary prevention: Canadian perspective. Can Fam Physician. 2011;57(4):417–23. [PMC free article] [PubMed] [Google Scholar]
- 19.Joint Committee on the Chinese Guidelines for Lipid Management [Chinese guidelines for lipid management (2023)] Zhonghua Xin xue guan bing za zhi. 2023;51(3):221–55. doi: 10.3760/cma.j.cn112148-20230119-00038. [DOI] [PubMed] [Google Scholar]
- 20.Harrell FE. Regression modeling strategies: with applications to linear models, logistic regression, and survival analysis. New York: Springer; 2001. pp. 20–4. [Google Scholar]
- 21.Li H, Ma Z, Zhai Y, Lv C, Yuan P, Zhu F, et al. Trimetazidine ameliorates myocardial metabolic remodeling in Isoproterenol-Induced rats through regulating ketone body metabolism via activating AMPK and PPAR α. Front Pharmacol. 2020;11:1255. doi: 10.3389/fphar.2020.01255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pearson GJ, Thanassoulis G, Anderson TJ, Barry AR, Couture P, Dayan N, et al. 2021 Canadian Cardiovascular Society Guidelines for the management of Dyslipidemia for the Prevention of Cardiovascular Disease in adults. Can J Cardiol. 2021;37(8):1129–50. doi: 10.1016/j.cjca.2021.03.016. [DOI] [PubMed] [Google Scholar]
- 23.Shay CM, Stamler J, Dyer AR, Brown IJ, Chan Q, Elliott P, et al. Nutrient and food intakes of middle-aged adults at low risk of cardiovascular disease: the international study of macro-/micronutrients and blood pressure (INTERMAP) Eur J Nutr. 2012;51(8):917–26. doi: 10.1007/s00394-011-0268-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Esfandiar Z, Hosseini-Esfahani F, Mirmiran P, Habibi-Moeini AS, Azizi F. Red meat and dietary iron intakes are associated with some components of metabolic syndrome: Tehran lipid and glucose study. J Transl Med. 2019;17(1):313. doi: 10.1186/s12967-019-2059-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wang W, Gao J, Li N, Han S, Wu L, Zhang Y, et al. Dietary iron and vitamins in association with mortality. Clin Nutr. 2021;40(4):2401–9. doi: 10.1016/j.clnu.2020.10.038. [DOI] [PubMed] [Google Scholar]
- 26.Fang X, An P, Wang H, Wang X, Shen X, Li X, et al. Dietary intake of heme iron and risk of cardiovascular disease: a dose-response meta-analysis of prospective cohort studies. Nutr Metab Cardiovasc Dis. 2015;25(1):24–35. doi: 10.1016/j.numecd.2014.09.002. [DOI] [PubMed] [Google Scholar]
- 27.Quintana Pacheco DA, Sookthai D, Wittenbecher C, Graf ME, Schübel R, Johnson T, et al. Red meat consumption and risk of cardiovascular diseases-is increased iron load a possible link? Am J Clin Nutr. 2018;107(1):113–9. doi: 10.1093/ajcn/nqx014. [DOI] [PubMed] [Google Scholar]
- 28.Han M, Guan L, Ren Y, Zhao Y, Liu D, Zhang D, et al. Dietary iron intake and risk of death due to cardiovascular diseases: a systematic review and dose-response meta-analysis of prospective cohort studies. Asia Pac J Clin Nutr. 2020;29(2):309–21. doi: 10.6133/apjcn.202007_29(2).0014. [DOI] [PubMed] [Google Scholar]
- 29.Aregbesola A, Voutilainen S, Virtanen JK, Mursu J, Tuomainen TP. Gender difference in type 2 diabetes and the role of body iron stores. Ann Clin Biochem. 2017;54(1):113–20. doi: 10.1177/0004563216646397. [DOI] [PubMed] [Google Scholar]
- 30.Lee DH, Folsom AR, Jacobs DR., Jr Iron, zinc, and alcohol consumption and mortality from cardiovascular diseases: the Iowa women’s Health Study. Am J Clin Nutr. 2005;81(4):787–91. doi: 10.1093/ajcn/81.4.787. [DOI] [PubMed] [Google Scholar]
- 31.Ivert T, Malmström H, Hammar N, Carlsson AC, Wändell PE, Holzmann MJ, et al. Cardiovascular events in patients under age fifty with early findings of elevated lipid and glucose levels - the AMORIS study. PLoS ONE. 2018;13(8):e0201972. doi: 10.1371/journal.pone.0201972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Nacarelli GS, Fasolino T, Davis S. Dietary, macronutrient, micronutrient, and nutrigenetic factors impacting cardiovascular risk markers apolipoprotein B and apolipoprotein A1: a narrative review. Nutr Rev. 2023:nuad102. [DOI] [PubMed]
- 33.Sniderman AD, Thanassoulis G, Glavinovic T, Navar AM, Pencina M, Catapano A, et al. Apolipoprotein B particles and Cardiovascular Disease: a narrative review. JAMA Cardiol. 2019;4(12):1287–95. doi: 10.1001/jamacardio.2019.3780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Mazidi M, Webb RJ, George ES, Shekoohi N, Lovegrove JA, Davies IG. Nutrient patterns are associated with discordant apoB and LDL: a population-based analysis. Br J Nutr. 2022;128(4):712–20. doi: 10.1017/S000711452100369X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Zhou B, Ren H, Zhou X, Yuan G. Associations of iron status with apolipoproteins and lipid ratios: a cross-sectional study from the China Health and Nutrition Survey. Lipids Health Dis. 2020;19(1):140. doi: 10.1186/s12944-020-01312-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Gong W, Liu A, Yao Y, Ma Y, Ding C, Song C, et al. Nutrient supplement use among the Chinese Population: a cross-sectional study of the 2010–2012 China Nutrition and Health Surveillance. Nutrients. 2018;10(11):1733. doi: 10.3390/nu10111733. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are available from CHNS (https://www.cpc.unc.edu/projects/china/data/datasets).