Abstract
Aim: Observational studies have reported that elevated homocysteine (Hcy) levels are associated with the risk of cardiovascular disease (CVD). However, interventions that lower Hcy do not provide a corresponding risk reduction. Therefore, the causal role of Hcy in CVD remains unclear. This 5-year prospective study investigated the associations of Hcy levels, folate intake, and host factors with arterial stiffness among the general Japanese population.
Methods: We prospectively recruited 658 participants (40–69 years old) from the general population during regular health checkup examinations. Arterial stiffness was evaluated using the cardio-ankle vascular index (CAVI) at baseline and the 5-year follow-up. Folate intake was estimated using a structured questionnaire. Genotyping was used to evaluate the MTHFR C677T and MS A2756G gene polymorphisms. Ultrafast liquid chromatography was used to measure total plasma Hcy levels. Association between these variables and CAVI values was evaluated using general linear regression and logistic regression models that were adjusted for atherosclerosis-related factors.
Results: Men had higher Hcy levels and CAVI values and lower folate intake than women (all, p < 0.001). At baseline, Hcy, folate intake, and the two genotypes were not associated with CAVI values for both sexes. Among men, Hcy levels were positively associated with CAVI values at the 5-year follow-up (p = 0.033). Folate intake and the two genotypes were not associated with the 5-year CAVI values.
Conclusion: Plasma Hcy may be involved in arterial stiffness progression, as monitored using CAVI, among men.
Keywords: Homocysteine, Arterial stiffness, Gene polymorphism, Cardio-ankle vascular index
See editorial vol. 23: 668–670
Introduction
McCully reported the vascular pathological mechanism of severe inherited homocysteinemia in 19691) and suggested the “homocysteine theory of arteriosclerosis” in 19752). Since that time, various observational studies have reported that elevated plasma/serum homocysteine (Hcy) concentrations were related to an increased risk of cardiovascular disease (CVD) and stroke3–7). The epidemiological factors for hyperhomocysteinemia are multifactorial and include age, sex, dietary intake of B-complex vitamins (folate, vitamin B6, and vitamin B12), smoking, alcohol intake, exercise, and genetic factors8). In addition, Hcy is biologically involved in atherosclerosis development through abnormal collagen cross-linking, oxidative stress, nitric oxide inhibition, and platelet agglutination5, 9). However, recent Hcy-lowering interventions via the intake of B-complex vitamins have failed to reduce the risk of CVD and stroke10, 11). Therefore, the causal effects of Hcy on the development of CVD and stoke remain debatable12, 13).
Pulse wave velocity (PWV) is a useful method for assessing arterial stiffness in individuals who are at the early stage of atherosclerosis not only in patients but also in the general population. Among patients with hypertension, most studies have investigated the association between high Hcy levels and arterial stiffness by measuring PWV14, 15). In addition, studies among the general population include seven cross-sectional studies and one cohort study16–23). However, the findings of these studies were inconsistent. Furthermore, cardio-ankle vascular index (CAVI) is a more proper marker for evaluating arterial stiffness than PWV because CAVI is independent from blood pressure and PWV is partially dependent on blood pressure24–27). Nevertheless, few studies have examined the association between Hcy levels and arterial stiffness using CAVI22). Moreover, no prospective study has evaluated the associations of Hcy levels and early-stage arterial stiffness with folate intake and host factors among the general population.
Aim
This study aimed to investigate the associations of Hcy levels, folate intake, and host factors with arterial stiffness during early-stage atherosclerosis as part of a 5-year follow-up study among the general Japanese population. In this study, we used CAVI to evaluate arterial stiffness and defined MTHFR and MS gene polymorphisms as potential host factors that may influence Hcy levels28, 29).
Methods
Population
We prospectively recruited study subjects from among participants in the Japan Multi-institutional Collaborative Cohort (J-MICC) Study, which has been described previously30–32). The baseline survey was conducted in the Amami island region of southern Japan during 2005. This baseline survey included 1,300 individuals (40–69 years old) who were undergoing a routine health checkup that was conducted by the local government or private companies after receiving the individual's written informed consent (response rate: 62.7%). The survey consisted of a questionnaire, blood collection, and examination for arterial stiffness using CAVI. Five years later, a follow-up survey with the same components and participants was performed during their voluntary routine health checkup; 856 individuals participated in that follow-up.
For our analyses, we only included patients with complete data at the baseline and 5-year surveys. Our exclusion criteria were the presence of dysrhythmia (e.g., bradycardia, atrial fibrillation, and frequent arterial or ventricular premature contractions), a history of arteriosclerosis obliterans surgery, Parkinson's disease, and an ankle brachial pressure index of < 0.9, as these conditions can cause CAVI measurement errors. We also excluded 14 subjects with outlier values for Hcy levels or folate intake (≥ 3 standard deviations). After these exclusion criteria were applied, we analyzed data from 658 individuals.
The present study was approved by the ethics review committee for human genome/gene analysis research at our institution.
Follow-Up
Using local government records, we followed all subjects to account for movement out of the study region or death. This follow-up revealed 37 cases of movement out of the study region, 16 deaths, and three withdrawals from the study during the 5-year follow-up.
Questionnaire Survey
We used a standardized structured questionnaire that was used in the J-MICC study30). The questionnaire includes questions regarding smoking, alcohol drinking, exercise, dietary habits, medical history, intake of prescription medicines and supplements, reproductive history, and stress. The questionnaire also includes a food frequency questionnaire (FFQ) regarding consumption frequencies and amounts for three staple foods, 43 other food items, and six types of alcoholic drinks33). The data regarding food and alcoholic drink consumption were collected as self-reported averages during the last year at baseline.
Health Checkup
We used health checkup data from the Kagoshima Kouseiren Health Center at the baseline and follow-up surveys, which included the participants' systolic and diastolic blood pressures, and levels of total cholesterol, triglycerides (TGs), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), fasting blood glucose (FBG), blood urea nitrogen, creatinine, and uric acid. The subjects were asked to fast for > 10 h before all testing, and LDL-C levels were calculated according to the Friedewald formula, using a TG level of < 400 mg/dl34). Cases with TG levels of ≥ 400 mg/dl were treated as having missing values for LDL-C.
CAVI Measurement
The CAVI measurements were performed using the VaSera VS-1000 and VS-1500 vascular screening systems (Fukuda Denshi Co., Ltd., Tokyo, Japan), as described previously24, 25). In brief, cuffs were applied to the participant's upper arms and ankles; electrocardiography leads were attached to both wrists; and a phonocardiography lead was placed on the sternum border at the second intercostal space. The subjects were instructed to lie in the supine position and to hold their heads at the midline position. The CAVI values were calculated using the following formula: CAVI = a [(2ρ/ΔP) ln (Ps/Pd) PWV2] + b, where ρ is blood density, Ps is systolic blood pressure, Pd is diastolic blood pressure, ΔP is Ps–Pd, and a and b are constants that match the aortic PWV. The equation was derived from Bramwell–Hill's equation and the stiffness parameter β24).
Hcy Measurement
Total plasma Hcy levels were determined using ultrafast liquid chromatography (Prominence UFLC, Shimadzu, Kyoto, Japan), as described previously35, 36). The mobile phase was 0.05 mol/L potassium dihydrogenphosphate (Nacalai, Kyoto, Japan) at pH 1.9, which was combined with 30 ml/L acetonitrile (Merck, Damstadt, Germany) at a flowrate of 1.0 ml/min. The stationary phase (Chromolith Column No. UM8125/004, Merck) was held at 40°C; the excitation wavelength was 385 nm; and the fluorescence wavelength was 515 nm. N-Acetyl-L-cysteine (Wako, Osaka, Japan) was added to all samples as an internal standard. The accuracy of the measured Hcy values was evaluated using a recovery test. In this test, standard samples of seven Hcy concentrations (0, 10, 20, 30, 40, 50, and 60 µmol/L) were added to select study samples, and the measured concentrations were compared with the expected concentrations.
DNA Extraction
DNA was extracted from the buffy coat fraction using the standard method and the QIAmp Blood Mini Kit (Qiagen, Valencia, CA, USA), the GenElute Blood Genomic DNA Kit (Sigma-Aldrich, St. Louis, MO, USA), or the Blood-Animal-Plant DNA Preparation Kit (Jena Bioscience, Jena, Germany).
Genotyping Single-Nucleotide Polymorphisms
We genotyped two single-nucleotide polymorphisms (MTHFR C677T [rs1801133] and MS A2756G [rs1805087]) using TaqMan allelic discrimination kits (Applied Biosystems, Foster City, CA, USA) and real-time polymerase chain reaction (StepOne, Applied Biosystems). Each reaction was performed in a 10-µl volume that contained 1 µl of genomic DNA (10 mg/µl), 5 µl of TaqMan Universal PCR Master Mix II, 0.45 µl of 40 × genotyping Assay Mix, and 3.75 µl of dH2O. The cycling was initiated by heating at 95°C for 10 min, which was followed by 40 cycles of 95°C for 15 s, 60°C for 1 min, and 25°C for 30 s.
Statistical Analysis
Hypertension was defined based on systolic blood pressures of ≥ 140 mmHg, diastolic blood pressures of ≥ 90 mmHg, or the use of antihypertensive medication. Dyslipidemia was defined based on LDL-C levels of ≥ 140 mg/dl, HDL-C levels of < 40 mg/dl, TG levels of ≥ 150 mm/dl, or the use of antihyperlipidemic medication. Glucose intolerance was defined based on FBG values of ≥ 110 mg/dl or treatment for diabetes mellitus. The participants' frequencies and intensities of habitual exercise were used to estimate their metabolic equivalents (METs), with light exercise indicating an MET of 3.3, moderate exercise indicating an MET of 4.0, and heavy exercise indicating an MET of 8.037). We defined arterial stiffness as CAVI values of ≥ 9.038). Nutrient intakes for total energy, three major nutrients, and 22 minor nutrients (including folate) were estimated using the FFQ. The validity of the estimated values has been evaluated previously39, 40). The estimated intakes of vitamins B6 and B12 were not included, because of the low correlation with the gold standard assay39).
We compared the participants' background characteristics according to sex and evaluated differences in the mean and prevalence values using the t test and chi-square test, respectively. We also compared the participants' characteristics according to arterial stiffness. Differences in mean and prevalence values were examined using two-way analysis of variance and logistic models, respectively, after adjusting for age dummy variables (40–49, 50–59, and 60–69 years).
The associations of folate intake, Hcy levels, and the MTHFR and MS genotypes with continuous CAVI values among men and women were evaluated at baseline and the 5-year follow-up, using multivariate general linear models. The score variable (1–3) was used for all genotype analyses in the general linear model. The coefficients in the multivariate analysis were estimated via two models that were adjusted for age (a continuous variable) or atherosclerosis-related factors (age, hypertension, dyslipidemia, glucose intolerance, current smoking, current drinking, family history of coronary heart disease, body mass index (BMI), and habitual exercise)41). Menopause was included for women in the model that was adjusted for atherosclerosis-related factors. The various associations were also analyzed using logistic regression models, and the odds ratios (ORs) for arterial stiffness and their 95% confidence intervals (CIs) were estimated. We used dummy variables (0 and 1) for various groups that generally included similar numbers of subjects (except for CAVI and the two genotypes): CAVI (< 9.0 and ≥ 9.0), 5-year change in CAVI (< 0.48 and ≥ 0.48), folate intake (men: < 153.4 and ≥ 153.4 µg/day and women: < 193.3 and ≥ 193.3 µg/day), Hcy levels (men: < 19.7 and ≥ 19.7 µmol/L and women: < 15.1 and ≥ 15.1 µmol/L), the MTHFR genotype [(CC and CT) and TT)], and the MS genotype [(AA and AG) and GG)].
The differences in genotype distribution from the Hardy–Weinberg equilibrium were evaluated using Pearson's chi-square test. All statistical analyses were performed using Stata software (version 12; Stata Corp., Collage Station, TX, USA), and differences with a p-value of < 0.05 were considered statistically significant.
Results
The mean ages of the male and female participants were similar (54.8 and 54.3 years, respectively; Table 1). Men had significantly higher baseline values for systolic blood pressure, diastolic blood pressure, TG, FBG, BMI, CAVI, and Hcy. Men also had significantly higher CAVI values at the 5-year follow-up. Women had significantly higher baseline levels of LDL-C, HDL-C, and folate intake. Menopause was observed in approximately 67.6% of women. The prevalence of the MTHFR and MS genotypes were similar for both sexes. None of the participants reported consuming folate as a single supplement, and the prevalence of supplement intake for plant extracts and vitamin complexes were < 4.0%. The mean folate intake amounts were similar for participants who consumed plant extracts or vitamin complexes and participants who did not consume these supplements (data not shown).
Table 1. The participants' clinical and lifestyle characteristics according to sex.
| Men (n = 285) | Women (n = 373) | P-value | |
|---|---|---|---|
| Mean ± SD | |||
| Age (years) | 54.8 ± 7.9 | 54.3 ± 7.1 | 0.346 |
| Systolic blood pressure (mmHg) | 132.3 ± 15.9 | 129.2 ± 17.9 | 0.008 |
| Diastolic blood pressure (mmHg) | 82.8 ± 8.9 | 78.4 ± 9.6 | < 0.001 |
| Triglycerides (mg/dL) | 176.1 ± 144.3 | 112.9 ± 74.8 | < 0.001 |
| LDL-cholesterol (mg/dL) | 122.3 ± 32.2 | 133.2 ± 31.9 | < 0.001 |
| HDL-cholesterol (mg/dL) | 58.0 ± 13.7 | 64.9 ± 14.1 | < 0.001 |
| Fasting blood glucose (mg/dL) | 105.3 ± 28.9 | 98.4 ± 21.7 | < 0.001 |
| BMI (kg/m2) | 25.2 ± 3.1 | 24.3 ± 3.4 | < 0.001 |
| Habitual exercise (METs/day) | 3.76 ± 4.08 | 3.18 ± 3.52 | 0.062 |
| CAVI at baseline | 8.08 ± 0.90 | 7.79 ± 0.92 | < 0.001 |
| CAVI at the 5-year follow-up | 8.63 ± 0.98 | 8.26 ± 1.00 | < 0.001 |
| Folate intake (µg/day) | 164.2 ± 89.9 | 212.8 ± 96.2 | < 0.001 |
| Homocysteine levels (µmol/L) | 20.1 ± 4.2 | 15.8 ± 3.8 | < 0.001 |
| N (%) | |||
| Current smoking | 75 (26.3) | 14 (3.8) | < 0.001 |
| Current drinking | 145 (50.9) | 7 (1.9) | < 0.001 |
| Family history of CHD | 75 (30.9) | 115 (35.2) | 0.345 |
| Supplement intake (plant extracts) | 6 (2.1) | 9 (2.4) | 0.793 |
| Supplement intake (vitamin complexes) | 10 (3.5) | 14 (3.8) | 0.868 |
| Menopause | – | 250 (67.6) | – |
| MTHFR polymorphisms | |||
| CC | 123 (43.2) | 177 (47.5) | |
| CT | 129 (45.3) | 152 (40.8) | |
| TT | 33 (11.6) | 44 (11.8) | 0.518 |
| MS polymorphisms | |||
| AA | 147 (51.6) | 182 (48.8) | |
| AG | 110 (38.6) | 148 (39.7) | |
| GG | 28 (9.8) | 43 (11.5) | 0.704 |
SD, standard deviation; LDL, low-density lipoprotein; HDL, high-density lipoprotein; BMI, body mass index; MET, metabolic equivalent; CAVI, cardio-ankle vascular index; CHD, coronary heart disease
The minor allele frequency of the MTHFR and MS polymorphisms was 0.33 and 0.30, respectively, and both genotype frequencies were within the Hardy–Weinberg equilibrium (p = 0.317 and p = 0.092, respectively; data not shown). The TT genotype of the MTHFR polymorphism was significantly associated with increased Hcy levels in both sexes after adjusting for age (both, p < 0.001). However, no clear association was observed for the MS genotype (data not shown).
As age strongly affects arterial stiffness, we evaluated each variable's association with arterial stiffness after adjusting for age. The values for systolic blood pressure in both sexes and TG in women were higher in participants with arterial stiffness (Table 2). The prevalence of current drinker status among men and the MTHFR TT genotype among women were also higher in participants with arterial stiffness. Folate intake and Hcy levels were not significantly different between participants with and without arterial stiffness. Menopause was more common among women with a CAVI value of ≥ 9.0 than among women with a CAVI value of < 9.0 (p = 0.052).
Table 2. Baseline clinical and lifestyle characteristics according to CAVI values and sex.
| Men |
Women |
|||||
|---|---|---|---|---|---|---|
| CAVI |
P-values§ | CAVI |
P-values§ | |||
| < 9.0 | ≥ 9.0 | < 9.0 | ≥ 9.0 | |||
| (n = 236) | (n = 49) | (n = 331) | n = 42) | |||
| Mean ± SD | Mean ± SD | |||||
| Age (years) | 53.7 ± 7.7 | 60.2 ± 6.6 | < 0.001 | 53.5 ± 6.9 | 60.4 ± 5.2 | < 0.001 |
| Systolic blood pressure (mmHg) | 130.0 ± 14.5 | 143.0 ± 17.7 | < 0.001 | 127.7 ± 17.7 | 140.6 ± 15.0 | 0.004 |
| Diastolic blood pressure (mmHg) | 82.4 ± 8.8 | 84.4 ± 9.5 | 0.191 | 78.1 ± 9.7 | 80.7 ± 8.9 | 0.358 |
| Triglycerides (mg/dL) | 169.4 ± 138.0 | 208.2 ± 169.4 | 0.069 | 109.3 ± 69.0 | 141.3 ± 107.6 | 0.020 |
| LDL-cholesterol (mg/dL) | 123.4 ± 32.9 | 117.1 ± 28.8 | 0.255 | 132.2 ± 32.0 | 141.0 ± 30.7 | 0.516 |
| HDL-cholesterol (mg/dL) | 57.8 ± 14.0 | 58.5 ± 12.3 | 0.819 | 65.1 ± 14.2 | 63.7 ± 13.1 | 0.701 |
| Fasting blood glucose (mg/dL) | 104.3 ± 29.5 | 109.9 ± 25.3 | 0.505 | 97.3 ± 20.1 | 106.8 ± 30.1 | 0.075 |
| BMI (kg/m2) | 25.2 ± 3.2 | 25.1 ± 2.3 | 0.505 | 24.3 ± 3.4 | 23.7 ± 2.8 | 0.096 |
| Habitual exercise (METs/day) | 3.49 ± 3.94 | 5.04 ± 4.53 | 0.118 | 3.13 ± 3.59 | 3.56 ± 2.93 | 0.792 |
| Folate intake (µg/day) | 160.7 ± 88.8 | 181.1 ± 94.1 | 0.523 | 212.7 ± 97.0 | 214.0 ± 91.2 | 0.236 |
| Homocysteine levels (µmol/L) | 20.0 ± 4.1 | 20.9 ± 4.5 | 0.115 | 15.7 ± 3.8 | 16.9 ± 3.6 | 0.418 |
| N (%) | N (%) | |||||
| Current smoking | 62 (26.3) | 13 (26.5) | 0.425 | 14 (4.2) | 0 (0.0) | – |
| Current drinking | 113 (47.9) | 32 (65.3) | 0.022 | 6 (1.8) | 1 (2.4) | 0.577 |
| Family history of CHD | 57 (27.8) | 18 (47.4) | 0.053 | 103 (35.4) | 12 (33.3) | 0.752 |
| Menopause | – | – | – | 209 (63.7) | 41 (97.6) | 0.052 |
| MTHFR (TT) | 27 (11.4) | 6 (12.2) | 0.907 | 37 (11.2) | 7 (16.7) | 0.015 |
| MS (GG) | 26 (11.0) | 2 (4.1) | 0.896 | 38 (11.5) | 5 (11.9) | 0.376 |
P-values were tested using analysis of variance or a logistic model, after adjusting for age (40–49 years, 50–59 years, and 60–69 years).
SD, standard deviation; LDL, low-density lipoprotein; HDL, high-density lipoprotein; BMI, body mass index; MET, metabolic equivalent; CAVI, cardio-ankle vascular index; CHD, coronary heart disease
After adjusting for age and other atherosclerosis-related variables, the baseline CAVI values were not associated with folate intake, Hcy levels, or the MTHFR and MS polymorphisms in both sexes (Table 3). However, the 5-year CAVI values were positively associated with Hcy levels among men after adjusting for age and other atherosclerosis-related variables (p < 0.033; Table 4). The difference between the baseline and 5-year CAVI values was also significant related to the elevated Hcy levels among men. However, these associations were not observed for folate intake or the MTHFR and MS polymorphisms in both sexes, and Hcy levels were not associated with CAVI values among women. Men with higher Hcy levels exhibited an elevated OR of arterial stiffness (OR: 1.94, 95% CI: 1.03–3.67; Table 5). Men with the MS GG genotype exhibited a decreased OR of arterial stiffness, although this analysis only included a small number of patients with the GG genotype (26 patients without arterial stiffness and two patients with arterial stiffness).
Table 3. Associations of folate intake, homocysteine levels, and host factors with CAVI at baseline according to sex.
| Men |
Women |
|||||||
|---|---|---|---|---|---|---|---|---|
| Coef.§ | P-value | Coef.¶ | P-value | Coef.§ | P-value | Coef.¶ | P-value | |
| Folate intake (µg/day) | 0.001 | 0.244 | 0.0003 | 0.954 | −0.001 | 0.123 | −0.0009 | 0.064 |
| Homocysteine levels (µmol/L) | 0.010 | 0.412 | 0.001 | 0.962 | −0.014 | 0.229 | −0.01 | 0.437 |
| MTHFR (CC, CT, TT) | −0.022 | 0.765 | 0.012 | 0.876 | 0.058 | 0.342 | 0.072 | 0.278 |
| MS (AA, AG, GG) | 0.003 | 0.073 | 0.038 | 0.618 | 0.050 | 0.417 | 0.041 | 0.541 |
Univariate analysis that was adjusted for age.
Multivariate general linear model that was adjusted for age, hypertension, dyslipidemia, glucose intolerance, current smoking, current drinking, family history of coronary heart disease, body mass index, habitual exercise, and menopause (among women).
CAVI, cardio-ankle vascular index; Coef., coefficient
Table 4. Associations of folate intake, homocysteine levels, and host factors with 5-year CAVI values, and the 5-year change in these variables, according to sex.
| Men |
Women |
|||||||
|---|---|---|---|---|---|---|---|---|
| Coef.§ | P-value | Coef.¶ | P-value | Coef.§ | P-value | Coef.¶ | P-value | |
| 5-year follow-up | ||||||||
| Folate intake (µg/day) | −0.0002 | 0.688 | −0.0002 | 0.760 | −0.0003 | 0.480 | −0.0006 | 0.228 |
| Homocysteine levels (µmol/L) | 0.031 | 0.012 | 0.031 | 0.033 | −0.009 | 0.459 | −0.008 | 0.536 |
| MTHFR (CC, CT, TT) | −0.029 | 0.714 | −0.001 | 0.987 | 0.002 | 0.975 | 0.012 | 0.867 |
| MS (AA, AG, GG) | −0.056 | 0.480 | −0.050 | 0.551 | −0.015 | 0.817 | 0.028 | 0.687 |
| Difference between the baseline and 5-year values | ||||||||
| Folate intake (µg/day) | −0.001 | 0.110 | −0.0002 | 0.713 | 0.0003 | 0.385 | 0.0003 | 0.552 |
| Homocysteine levels (µmol/L) | 0.022 | 0.063 | 0.030 | 0.030 | 0.005 | 0.651 | 0.002 | 0.899 |
| MTHFR (CC, CT, TT) | −0.007 | 0.924 | −0.013 | 0.875 | −0.056 | 0.310 | −0.061 | 0.343 |
| MS (AA, AG, GG) | −0.059 | 0.426 | −0.092 | 0.286 | −0.065 | 0.241 | −0.012 | 0.847 |
Univariate analysis that was adjusted for age.
Multivariate general linear model that was adjusted for age, hypertension, dyslipidemia, glucose intolerance, current smoking, current drinking, family history of coronary heart disease, body mass index, and habitual exercise in men and women, and additional variable of menopause in women. CAVI, cardio-ankle vascular index; Coef., coefficient
Table 5. Odds ratios for the effects of folate intake, homocysteine levels, and host factors on 5-year CAVI values, and the 5-year change, according to sex†.
| Men |
Women |
|||||||
|---|---|---|---|---|---|---|---|---|
| OR‡ | 95% CI | OR§ | 95% CI | OR‡ | 95% CI | OR§ | 95% CI | |
| 5-year follow-up | ||||||||
| Folate intake‖ | 0.74 | 0.44–1.26 | 0.77 | 0.41–1.44 | 0.90 | 0.52–1.55 | 0.81 | 0.42–1.54 |
| Homocysteine levels# | 1.39 | 0.82–2.35 | 1.94 | 1.03–3.67 | 0.98 | 0.57–1.68 | 0.98 | 0.52–1.86 |
| MTHFR (CC, CT, TT) | 0.74 | 0.32–1.70 | 0.81 | 0.29–2.22 | 0.85 | 0.36–1.99 | 1.02 | 0.39–2.63 |
| MS (AA, AG, GG) | 0.20 | 0.06–0.70 | 0.19 | 0.04–0.91 | 0.28 | 0.09–0.84 | 0.32 | 0.10–1.04 |
| Difference between the baseline and 5-year values | ||||||||
| Folate intake‖ | 0.79 | 0.50–1.27 | 0.89 | 0.52–1.54 | 1.37 | 0.90–2.07 | 1.41 | 0.88–2.26 |
| Homocysteine levels# | 1.16 | 0.73–1.85 | 1.28 | 0.75–2.20 | 1.06 | 0.70–1.61 | 0.97 | 0.60–1.56 |
| MTHFR (CC, CT, TT) | 0.98 | 0.47–2.04 | 1.37 | 0.58–3.24 | 1.08 | 0.57–2.02 | 1.18 | 0.58–2.38 |
| MS (AA, AG, GG) | 0.94 | 0.43–2.06 | 0.90 | 0.36–2.26 | 0.58 | 0.30–1.12 | 0.68 | 0.33–1.38 |
Dummy variables (0 and 1) were used for the 5-year CAVI values (< 9.0 and ≥ 9.0) and for the 5-year change in the CAVI values (< 0.48 and ≥ 0.48).
Univariate analysis that was adjusted for age.
Logistic model that was adjusted for age, hypertension, dyslipidemia, glucose intolerance, current smoking, current drinking, family history of coronary heart disease, body mass index, habitual exercise, and menopause (among women).
Dummy variables (0 and 1) were used for folate intake (men: < 153.4 µg/day and ≥ 153.4 µg/day, women: < 193.3 µg/day and ≥ 193.3 µg/day).
Dummy variables (0 and 1) were used for homocysteine levels (men: < 19.7 µmol/L and ≥ 19.7 µmol/L, women: < 15.1 µmol/L and ≥ 15.1 µmol/L). OR, odd ratio; CI, confidence interval; CAVI, cardio-ankle vascular index
Discussion
This prospective study was designed to investigate the associations of CAVI values with Hcy levels, folate intake, and host factors among the general Japanese population who were in the early stage of atherosclerosis. Our results indicate that Hcy levels are significantly associated with 5-year CAVI values among men.
Previous observational studies have consistently reported that high Hcy levels and low folate levels were associated with a risk of vascular disease3–6). We reviewed the findings of seven cross-sectional studies and one cohort study to investigate the association between Hcy levels and arterial stiffness using PWV or CAVI as a quantitative marker for the degree of early arteriosclerosis (Supplemental Table 1). The cross-sectional studies reported inconsistent results, with no association found in three studies16–18) and a positive association found in four studies19–22). Furthermore, one of the cross-sectional studies reported a positive association between Hcy levels and CAVI values, although the number of participants was small, and the authors did not adjust for age or other factors21). Our findings of a positive association between Hcy levels and 5-year CAVI values among men is partially in agreement with the cohort study's findings, which revealed a positive association after a 4.8-year follow-up23). In this context, PWV is potentially over-estimated by hypertension, but CAVI is independent from blood pressure at the measuring time24, 25), which allows for a more accurate measurement of arterial stiffness. Furthermore, a prospective observational study has less bias than a cross-sectional study, which includes the potential for reverse causality. Thus, the present study's strengths include its prospective design and use of CAVI to measure arterial stiffness.
Supplemental Table 1. Previous cross-sectional and cohort studies regarding the association between homocysteine levels and arterial stiffness among the general population.
| Authors§ | Year | Country | Design | Participants (mean age in years) | Arterial stiffness measurement | Results | Remarks |
|---|---|---|---|---|---|---|---|
| Woodside JV, et al. | 2004 | US | Cross-sectional | 251 men (22) and 238 women (23) | PWV | No association | |
| Nakhai-Pour HR, et al. | 2007 | Nether-lands | Cross-sectional | 376 men (60) | PWV, IMT | No independent association | Positive association in the univariate model |
| Bree A, et al. | 2006 | France | Cross-sectional | 556 men (60) and 559 women (60) | PWV, IMT | No independent association in both sexes | Positive association among men after adjusting for age |
| Ruan L, at al. | 2008 | US | Cross-sectional | 444 men (37) and 585 women (36) | af-PWV | Positive association | 735 white and 294 black participants |
| Yun J, et al. | 2011 | Korea | Cross-sectional | 612 men (47) | ba-PWV | Positive association | |
| Zhong J, et al. | 2013 | China | Cross-sectional | 71 men (59) and 81 women (50) | CAVI | Positive association | No adjustment |
| Zhang S, et al. | 2014 | China | Cross-sectional | 369 men and 411 women (72) | cf-PWV, cr-PWV | Positive association with cf-PWV, no association with cr-PWV | |
| Wang XN, et al. | 2015 | China | 4.8-year cohort | 579 men and 868 women (61) | cf-PWV, cr-PWV | Positive association with 4.8-year data | No association with the change |
| Present study | 2015 | Japan | 5-year cohort | 285 men (55) and 373 women (54) | CAVI | Positive association with 5-year data and the change among men | No association among women or among men at baseline |
References: 16–23
af-PWV, aorta-femoral pulse wave velocity; ba-PWV, brachial-ankle PWV; CAVI, cardio-ankle vascular index; cf-PWV, carotid-femoral PWV; cr-PWV, carotid-radial PWV.
One cross-sectional study investigated the association between Hcy levels and PWV on the basis of sex and found no independent association, although a positive association was observed among men after adjusting for age18). Another cross-sectional study among men found a positive association between Hcy levels and PWV20). Sex-specific differences are generally observed in some background factors for atherosclerosis, and our findings indicate that men have higher Hcy levels and CAVI values than women. It is possible that the association between Hcy levels and arterial stiffness in men is more detectable than that in women, which may require a greater number of subjects to provide sufficient statistical power. Nevertheless, a previous cohort study did not observe any sex-specific differences in the association between Hcy levels and the risk of stroke that may be caused by advanced atherosclerosis 5). Therefore, sex-specific analyses in a large-scale prospective study are needed to confirm the sex-specific effect(s) of Hcy levels on early-stage arterial stiffness.
MTHFR, MS, and cystathionine beta-synthase enzymes play important roles in the metabolism of Hcy28, 29). For example, MTHFR and MS regulate remethylation, which is a key step in the methylene metabolic pathway. In addition, MTHFR is the most important enzyme in Hcy metabolism and is responsible for the synthesis of the circulating form of folate (5-methyltetrahydrofolate). Therefore, genetic polymorphisms in these enzymes may be related to altered plasma levels of Hcy. The homozygous TT variant of the MTHFR C677T polymorphism is associated with reduced MTHFR activity and mild hyperhomocysteinemia42). A previous meta-analysis has also reported similar highly significant results regarding the association between the MTHFR C677T polymorphism and serum Hcy levels in prospective studies of ischemic heart disease6). However, another meta-analysis regarding the association between the MTHFR C677T polymorphism and coronary heart disease found no significant association among European, North American, or Australian studies; these findings conflict with those of Middle Eastern and Asian studies43). The latter report concluded that there was no clear role for folic acid in preventing CVD by lowering Hcy levels. Our findings also revealed a positive association between the MTHFR C677T polymorphism and Hcy levels, although there was no association with arterial stiffness. Therefore, these findings suggest that measured Hcy levels may be more relevant as an indicator for the development of arterial stiffness, rather than as a marker for genetic susceptibility to this condition.
It is also unclear whether the MS A2756G polymorphism affects the enzyme's function. Because the MS A2756G polymorphism is located in a potentially functional domain of the enzyme44), it has been hypothesized that this polymorphism may increase Hcy levels. However, previous studies have demonstrated that the MS A2756G polymorphism was not associated with Hcy levels29, 45). Furthermore, although we observed that the GG genotype was negatively associated with arterial stiffness in our logistic analysis, this finding may be due to random error, given the limited number of subjects with the GG genotype.
We also did not observe an association between folate intake and CAVI values, and this finding is in agreement with the previous studies that evaluated interventions that lowered Hcy levels10, 11). In the present study, which was performed in the Amami island region, we estimated folate consumption using a structured FFQ that has been validated in Japan's Tokai and Amami island regions39, 40). Therefore, it is unlikely that our negative finding was related to incorrectly estimated folate consumption.
Several limitations must be considered when interpreting our findings. First, this study did not evaluate a large number of participants, and it is possible that a random error was introduced. However, the effect of this error on the conventional risk factors for CAVI is possibly small because our findings regarding CAVI risk factors are in agreement with those of a larger study (n = 4,523)32). Second, we selected MTHFR and MS polymorphisms, which may be involved in the metabolism of Hcy. Although several genes are involved in this pathway, MTHFR is the thermolabile enzyme in this pathway, and the C677T variant is functional42). Therefore, it is likely that the MTHFR C677T polymorphism plays the most important role in modulating Hcy levels, and other enzymes possibly play less significant roles. Third, we did not evaluate plasma/serum folate levels in the present study, although most previous studies have reported that Hcy levels and plasma/serum folate levels are related8). Although there may be some discrepancy between folate intake and plasma/serum folate levels, because of individuals' variations in their absorption and metabolism, our study reflects the dietary folate intake of the general Japanese population. Fourth, we did not use the estimated intakes of vitamin B6 and B12, because the questionnaire that we used has low validity for these parameters39).
In conclusion, this prospective study revealed that plasma Hcy levels were associated with arterial stiffness, as monitored using CAVI, among men. However, further large-scale prospective studies are needed to confirm whether Hcy levels are involved in early-stage arterial stiffness for each sex.
Acknowledgments
We thank the local governments and staff of Wadomari Town, China Town, and JA Kagoshima Kouseiren Medical Health Care Center for collaborating in the data collection. We also thank Prof. Kenji Wakai (Nagoya University Graduate School of Medicine), Ms. Nahomi Imaeda (Nagoya Women's University), and Ms. Chiho Goto (Nagoya Bunri University) for estimating the participants' folate intake. Furthermore, we thank Editage (www.editage.jp) for English language editing. This study was supported in part by a grant-in-aid for Scientific Research on Priority Areas of Cancer (No. 17015018) and Innovative Areas (No. 221S0001), and a grant-in-aid for Scientific Research (C) (No. 22590551) from the Japanese Ministry of Education, Culture, Sports, Science and Technology.
Conflicts of Interest
None.
References
- 1). McCully KS: Vascular pathology of homocysteinemia; implications for the pathogenesis of arteriosclerosis. Am J Pathol, 1969; 56: 111-128 [PMC free article] [PubMed] [Google Scholar]
- 2). McCully KS, Wilson RB: Homocysteine theory of arteriosclerosis. Atherosclerosis, 1975; 22: 215-227 [DOI] [PubMed] [Google Scholar]
- 3). Stampfer MJ, Malinow MR, Willett WC, Newcomer LM, Upson B, Ullmann D, Tishler PV, Hennekens CH: A prospective study of plasma homocyst(e)ine and risk of myocardial infarction in US physicians. JAMA, 1992; 268: 887-881 [PubMed] [Google Scholar]
- 4). Verhoef P, Hennekens CH, Malinow MR, Kok FJ, Willet WC: A prospective study of plasma homocyst(e)ine and risk of ischemic stroke. Stroke, 1994; 25: 1924-1930 [DOI] [PubMed] [Google Scholar]
- 5). Iso H, Moriyama Y, Sato S, Kitamura A, Tanigawa T, Yamagishi K, Imano H, Ohira T, Okamura T, Naito Y, Shimamoto T: Serum total homocysteine concentrations and risk of stroke and its subtypes in Japanese. Circulation, 2004; 109: 2766-2772 [DOI] [PubMed] [Google Scholar]
- 6). Wald DS, Law M, Morris JK: Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis. BMJ, 2002; 325: 1202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7). Kumakura H, Fujita K, Kanai H, Araki Y, Hojo Y, Kasama S, Iwasaki T, Ichikawa S, Nakashima K, Minami K: High-sensitivity C-reactive protein, lipoprotein(a) and homocysteine are risk factors for coronary artery disease in Japanese patients with peripheral arterial disease. J Atheroscler Thromb, 2015; 22: 344-354 [DOI] [PubMed] [Google Scholar]
- 8). Refsum H, Ueland PM, Nygard O, Vollset SE: Homocysteine and cardiovascular disease. Annu Rev Med, 1998; 49: 31-62 [DOI] [PubMed] [Google Scholar]
- 9). Arensen E, Refsum H, Bonaa KH, Ueland PM, Forde OH, Nordrehaugh JE: Serum total homocysteine and coronary artery disease. Int J Epidemiol, 1995; 24: 704-709 [DOI] [PubMed] [Google Scholar]
- 10). Marti-Carvajal AJ, Sola I, Lathyris D, Salanti G: Homocysteine lowering interventions for preventing cardiovascular events. Cochrane Database Syst Rev, 2009; 10.1002/14651858.CD006612.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11). Bazzano LA, Reynolds K, Holder KN, He J: Effect of folic acid supplementation on risk of cardiovascular diseases: a meta-analysis of randomized controlled trials. JAMA, 2006; 296: 2720-2726 [DOI] [PubMed] [Google Scholar]
- 12). Moat SJ: Plasma total homocysteine: instigator or indicator of cardiovascular disease? Ann Clin Biochem, 2008; 45: 345-348 [DOI] [PubMed] [Google Scholar]
- 13). Potter K: Homocysteine and cardiovascular disease: should we treat? Clin Biochem Rev, 2008; 29: 27-30 [PMC free article] [PubMed] [Google Scholar]
- 14). Xiao W, Bai Y, Ye P, Luo L, Liu D, Wu H, Bai J: Plasma homocysteine is associated with aortic arterial stiffness but not wave reflection in Chinese hypertensive subjects. PLoS One, 2014; 9: e85938. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15). Tayama J, Munakata M, Yoshinaga K, Toyota T: Higher plasma homocysteine concentration is associated with more advanced systemic arterial stiffness and greater blood pressure response to stress in hypertensive patients. Hypertens Res, 2006; 29: 403-409 [DOI] [PubMed] [Google Scholar]
- 16). Woodside JV, McMahon R, Gallagher AM, Cran GW, Boreham CA, Murray LJ, Strain JJ, McNulty H, Robson PJ, Brown KS, Whitehead AS, Savage M, Young IS: Total homocysteine is not a determinant of arterial pulse wave velocity in young healthy adults. Atherosclerosis, 2004; 177: 337-344 [DOI] [PubMed] [Google Scholar]
- 17). Nakhai-Pour HR, Grobbee DE, Bots ML, Muller M, van der Schouw YT: Circulating homocysteine and large arterial stiffness and thickness in a population-based sample of middle-aged and elderly men. J Hum Hypertens, 2007; 21: 942-948 [DOI] [PubMed] [Google Scholar]
- 18). de Bree A, Mennen LI, Zureik M, Ducros V, Guilland JC, Nicolas JP, Emery-Fillon N, Blacher J, Hercberg S, Galan P: Homocysteine is not associated with arterial thickness and stiffness in healthy middle-aged French volunteers. Int J Cardiol, 2006; 113: 332-340 [DOI] [PubMed] [Google Scholar]
- 19). Ruan L, Chen W, Srinivasan SR, Xu J, Sun M, Toprak A, Berenson GS: Relation of plasma homocysteine to arterial stiffness in black and white young adults (from the Bogalusa Heart Study). Am J Cardiol, 2009; 103: 985-988 [DOI] [PubMed] [Google Scholar]
- 20). Yun J, Kim JY, Kim OY, Jang Y, Chae JS, Kwak JH, Lim HH, Park HY, Lee SH, Lee JH: Associations of plasma homocysteine level with brachial-ankle pulse wave velocity, LDL atherogenicity, and inflammation profile in healthy men. Nutr Metab Cardiovasc Dis, 2011; 21: 136-143 [DOI] [PubMed] [Google Scholar]
- 21). Zhong J, Wang Y, Wang X, Li F, Hou Y, Luo H, Chen H: Significance of CAVI, hs-CRP and homocysteine in subclinical arteriosclerosis among a healthy population in China. Clin Invest Med, 2013; 36: E81-E86 [DOI] [PubMed] [Google Scholar]
- 22). Zhang S, Bai YY, Luo LM, Xiao WK, Wu HM, Ye P: Association between serum homocysteine and arterial stiffness in elderly: a community-based study. J Geriatr Cardiol, 2014; 11: 32-38 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23). Wang XN, Ye P, Cao RH, Yang X, Xiao WK, Zhang Y, Bai YY, Wu HM: Plasma Homocysteine is a Predictive Factor for Arterial Stiffness: A Community-Based 4.8-Year Prospective Study. J Clin Hypertens (Greenwich), 2015; 17: 594-600 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24). Shirai K, Utino J, Otsuka K, Takata M: A novel blood pressure-independent arterial wall stiffness parameter; cardio- ankle vascular index (CAVI). J Atheroscler Thromb, 2006; 13: 101-107 [DOI] [PubMed] [Google Scholar]
- 25). Ibata J, Sasaki H, Kakimoto T, Matsuno S, Nakatani M, Kobayashi M, Tatsumi K, Nakano Y, Wakasaki H, Furuta H, Nishi M, Nanjo K: Cardio-ankle vascular index measures arterial wall stiffness independent of blood pressure. Diabetes Res Clin Pract, 2008; 80: 265-270 [DOI] [PubMed] [Google Scholar]
- 26). Soska V, Frantisova M, Dobsak P, Dusek L, Jarkovsky J, Novakova M, Shirai K, Fajkusova L, Freiberger T: Cardioankle vascular index in subjects with dyslipidaemia and other cardiovascular risk factors. J Atheroscler Thromb, 2013; 20: 443-451 [DOI] [PubMed] [Google Scholar]
- 27). Dobsak P, Soska V, Sochor O, Jarkovsky J, Novakova M, Homolka M, Soucek M, Palanova P, Lopez-Jimenez F, Shirai K: Increased cardio-ankle vascular index in hyperlipidemic patients without diabetes or hypertension. J Atheroscler Thromb, 2015; 22: 272-283 [DOI] [PubMed] [Google Scholar]
- 28). Jacques PF, Bostom AG, Williams RR, Ellison RC, Eckfeldt JH, Rosenberg IH, Selhub J, Rozen R: Relation between folate status, a common mutation in methylenetetrahydrofolate reductase, and plasma homocysteine concentrations. Circulation, 1996; 93: 7-9 [DOI] [PubMed] [Google Scholar]
- 29). Chen J, Stampfer MJ, Ma J, Selhub J, Malinow MR, Hennekens CH, Hunter DJ: Influence of a methionine synthase (D919G) polymorphism on plasma homocysteine and folate levels and relation to risk of myocardial infarction. Atherosclerosis, 2001; 154: 6678-6672 [DOI] [PubMed] [Google Scholar]
- 30). Hamajima N: The Japan Multi-Institutional Collaborative Cohort Study (J-MICC Study) to detect gene-environment interactions for cancer. Asian Pac J Cancer Prev, 2007; 8: 317-323 [PubMed] [Google Scholar]
- 31). Naito M, Eguchi H, Okada R, Ishida Y, Nishio K, Hishida A, Wakai K, Tamakoshi A, Hamajima N: Controls for monitoring the deterioration of stored blood samples in the Japan Multi-Institutional Collaborative Cohort Study (J-MICC Study). Nagoya J Med Sci, 2008; 70: 107-115 [PubMed] [Google Scholar]
- 32). Hirasada K, Niimura H, Kubozono T, Nakamura A, Tatebo M, Ogawa S, Tsunematsu N, Chiba S, Matsushita T, Kusano K, Miyata M, Takezaki T: Values of cardio-ankle vascular index (CAVI) between Amami islands and Kagoshima mainland among health checkup examinees. J Atheroscler Thromb, 2012; 19: 69-80 [DOI] [PubMed] [Google Scholar]
- 33). Tokudome S, Goto C, Imaeda N, Tokudome Y, Ikeda M, Maki S: Development of a data-based short food frequency questionnaire for assessing nutrient intake by middle-aged Japanese. Asian Pac J Cancer Prev, 2004; 5: 40-43 [PubMed] [Google Scholar]
- 34). Friedewald WT, Levy RI, Fredrickson DS: Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem, 1972; 18: 499-502 [PubMed] [Google Scholar]
- 35). Refsum H, Ueland PM, Svardal AM: Fully automated fluorescence assay for determining total homocysteine in plasma. Clin Chem, 1989; 35: 1921-1927 [PubMed] [Google Scholar]
- 36). Kakimoto T, Otsuka A, Kawaguchi H, Ogata K, Tanimoto A, Kanouchi H: Plasma homocysteine concentrations in novel microminipigs. In Vivo, 2014; 28: 579-582 [PubMed] [Google Scholar]
- 37). American College of Sports Medicine : ACSM's Metabolic Calculations Handbook. Lippincott Williams & Wilkins, Philadelphia, 2006 [Google Scholar]
- 38). Suzuki K, Ishizuka N, Miyashita H, Shirai K: Standardization of CAVI as a non-invasive and blood-pressure-independent method for the examination of arterial stiffness: an epidemiological investigation on standard values and its plausibility. The Niigata Journal of Medical Technology, 2008; 48: 2-10 [in Japanese] [Google Scholar]
- 39). Tokudome Y, Goto C, Imaeda N, Hasegawa T, Kato R, Hirose K, Tajima K, Tokudome S: Relative validity of a short food frequency questionnaire for assessing nutrient intake versus three-day weighed diet records in middle-aged Japanese. J Epidemiol, 2005; 15: 135-145 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40). Nakahata NT, Takeda AN, Imaeda N, Goto C, Kuwabara KH, Niimura H, Arai Y, Yoshida K, Takezaki T: Validity of the Food Frequency Questionnaire in a population with high alcohol consumption in Japan. Asia Pac J Clin Nutr, 2016. (in press) [DOI] [PubMed] [Google Scholar]
- 41). Teramoto T, Sasaki J, Ishibashi S, Birou S, Daida H, Dohi S, Egusa G, Hiro T, Hirobe K, Iida M, Kihara S, Kinoshita M, Maruyama C, Ohta T, Okamura T, Yamashita S, Yokode M, Yokote K: Cardiovascular disease risk factors other than dyslipidemia. Executive summary of the Japan Atherosclerosis Society (JAS) guidelines for the diagnosis and prevention of atherosclerotic cardiovascular diseases in Japan - 2012 version. J Atheroscler Thromb, 2013; 20: 733-742 [DOI] [PubMed] [Google Scholar]
- 42). Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, Boers GJ, den Heijer M, Kluijtmans LA, van den Heuvel LP, Rozen R: A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet, 1995; 10: 111-113 [DOI] [PubMed] [Google Scholar]
- 43). Lewis SJ, Ebrahim S, Davey Smith G: Meta-analysis of MTHFR 677C->T polymorphism and coronary heart disease: does totality of evidence support causal role for homocysteine and preventive potential of folate? BMJ, 2005; 331: 1053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44). Matthews RG, Sheppard C, Goulding C: Methylenetetrahydrofolate reductase and methionine synthase: biochemistry and molecular biology. Eur J Pediatr, 1998; 157 (Suppl 2): S54-59 [DOI] [PubMed] [Google Scholar]
- 45). Klerk M, Lievers KJ, Kluijtmans LA, Blom HJ, den Heijer M, Schouten EG, Kok FJ, Verhoef P: The 2756A>G variant in the gene encoding methionine synthase: its relation with plasma homocysteine levels and risk of coronary heart disease in a Dutch case-control study. Thromb Res, 2003; 110: 87-91 [DOI] [PubMed] [Google Scholar]