Abstract
The authors investigated whether plasma total homocysteine (tHcy) is a predictive factor for arterial stiffness (carotid‐femoral pulse wave velocity [cf‐PWV] and carotid‐radial PWV) in 1447 patients from a 4.8‐year prospective study in Beijing, People's Republic of China. Baseline tHcy showed a significant relationship with follow‐up cf‐PWV (β=0.817, P=.015) in a multivariable linear regression analysis. A stepwise logistic regression model showed that baseline levels of tHcy were significantly associated with follow‐up cf‐PWV in the adjusted models. Furthermore, the baseline tHcy levels showed a significant association with increases in cf‐PWV. There was no association between the change in tHcy and increase in PWV. The present study clearly demonstrated an association between tHcy levels and arterial stiffness, indicating that tHcy is an independent predictive factor for arterial stiffness in a community‐based population.
In recent years, great emphasis has been placed on the role of arterial stiffness in the development of cardiovascular (CV) diseases. Arterial stiffness has been increasingly recognized as a strong predictor of CV disease (CVD) and atherosclerotic disease.1, 2, 3, 4, 5 There are several conventional techniques for evaluating arterial stiffness, but these techniques are inconvenient (particularly in large clinical trials).6 Arterial stiffness can be assessed noninvasively via the measurement of pulse wave velocity (PWV),7, 8 and carotid‐femoral PWV (cf‐PWV) is the “gold standard” for arterial stiffness, with the largest amount of epidemiological evidence for its predictive value for CV events, but requires little technical expertise.1
An elevated plasma total homocysteine (tHcy) level is considered another predictive risk factor for cardiovascular diseases and subsequent mortality. In one study, patients with a tHcy level of 15 mmol/L to 19.9 mmol/L had approximately 2.5 times higher mortality, and those with tHcy >20 mmol/L had approximately six times higher mortality than patients with tHcy<9 mmol/L.9 The relationship of tHcy with these diseases may be mediated by its adverse effects on vascular endothelium and smooth muscle with resultant alterations in subclinical arterial structure and function.
Arterial stiffness and tHcy are both powerful predictors of cardiovascular disease. Previous studies have investigated the association of tHcy with arterial stiffness in humans with conflicting results.9, 10, 11 These discrepancies may be attributed to different study populations. Plasma total tHcy levels vary by age and have significant ethnic‐ and sex‐dependent differences.12, 13, 14 Moreover, previous studies used a cross‐sectional design and could not determine whether tHcy is a predictive factor for arterial stiffness.
We hypothesize that tHcy is a predictive factor for arterial stiffness. Therefore, the current study examined the relationship of tHcy with arterial stiffness (cf‐PWV and carotid‐radial PWV) by investigating: (1) the predictive relationship between levels of tHcy with arterial stiffness and the change of arterial stiffness (cf‐PWVδ); and (2) the predictive relationship between change of tHcy (tHcyδ) with arterial stiffness and the change of arterial stiffness (cf‐PWVδ) in a large community‐based longitudinal sample from China.
Methods
Study Population
This paper reports the results of a community‐based cohort study of people living in the Pingguoyuan area of Beijing, China. A total of 1680 participants were initially eligible for cross‐sectional analysis12 between September 2007 and January 2009. We prospectively followed this community‐based population. The follow‐up visits were conducted from February 1 to September 30, 2013. During these visits, all participants came and received a questionnaire survey in a mobile examination center. Demographic information, medical history, blood pressure measurements, and anthropometric measurements were obtained. Fasting blood and urine samples were also collected. The median follow‐up interval for the original 1680 participants was 4.8 years. During the period between the initiation of the study and the follow‐up, 181 participants were lost to follow‐up and were excluded from the analysis. Complete follow‐up data were obtained from 1499 participants (follow‐up rate, 89.2%). Of these, 52 were excluded from analyses because of death, leaving 1447 participants available for analysis. The study was approved by the ethics committee of the People's Liberation Army General Hospital, and each participant provided informed written consent.
Clinical Data Collection
Information about lifestyle factors, prevalent diseases, family history, and medication use was obtained through self‐reporting standardized questionnaires. Anthropometrics were evaluated by trained medical doctors. Height (cm) was measured using a wall‐mounted measuring tape, and weight (kg), without shoes, was measured using a digital scale. Systolic and diastolic blood pressures (SBP and DBP) were measured two times on the right arm after 5 minutes of rest in a sitting position. The average SBP and DBP measurements were used for further analysis.
Biomarker Variable Determination
Blood samples were collected from participants between 8 am and 10 am after a fast of at least 12 hours. Plasma aliquots were frozen at −80°C until the assays were performed. Concentrations of fasting glucose, total cholesterol (TC), triglycerides (TGs), high‐density lipoprotein cholesterol (HDL‐C), low‐density lipoprotein cholesterol (LDL‐C), and uric acid (UA) were determined using Roche enzymatic assays (Roche Diagnostics GmbH, Mannheim, Germany) on a Roche autoanalyzer (Roche Diagnostics, Indianapolis, IN). Concentrations of homocysteine were assayed by high‐performance chromatography with fluorometric detection. Concentrations of plasma creatinine (Cr) were measured by enzymatic assay (Roche Diagnostics GmbH) on a Hitachi 7600 autoanalyzer (Hitachi, Tokyo, Japan). All blood specimens were measured by the same laboratory, following the criteria of the World Health Organization Lipid Reference Laboratories.
Assessment of Arterial Stiffness
Arterial stiffness was measured in the morning, in a quiet environment, and at a stable temperature. Before the assessment was performed, all participants were asked to avoid caffeine, smoking, and alcohol for at least 12 hours. Arterial stiffness was determined by automatic cf‐and carotid‐radial PWV measurements using a Complior SP device (Createch Industrie, Massy, France) after the patients had rested in a supine position for 5 to 10 minutes. PWV along the artery was measured with two strain‐gauge transducers. The procedure is noninvasive and uses a TY‐306 Fukuda pressure‐sensitive transducer (Fukuda Denshi Co, Tokyo, Japan) that is fixed transcutaneously over the course of a pair of arteries separated by a known distance; the carotid, femoral, and radial arteries (all on the right side) were used. Two transducers were used: one positioned at the base of the neck over the common carotid artery and the other over the femoral artery. Two different pulse waves were obtained simultaneously at two sites. The measurement was repeated over 10 different cardiac cycles. PWV was calculated from the measurement of the pulse transit time and the distance traveled by the pulse between the two recording sites (measured on the surface of the body in meters) according to the following formula: PWV (m/s)=distance (m)/transit time (s).4
All baseline and follow‐up measurements were performed by the same technicians.
Definition of Variables
Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared (kg/m2). Hypertension was defined as a mean SBP ≥140 mm Hg, mean DBP ≥90 mm Hg, or use of antihypertensive medication. Diabetes mellitus (DM) was defined as a fasting glucose ≥7.0 mmol/L, glucose ≥11.1 mmol/L at two hours after an oral 75 g glucose challenge, or both, or use of antihyperglycemic medication.15 The estimated glomerular filtration rate (eGFR) was calculated using the following Chronic Kidney Disease Epidemiology Collaboration equations: GFR=141×min (Scr/κ, 1)α×max (Scr/κ, 1)−1.209×0.993Age×1.018 [if female]×1.159 [if black], where Scr is plasma creatinine (mg/dL), κ is 0.7 for females and 0.9 for males, α is −0.329 for females and −0.411 for males, min indicates the minimum of Scr/κ or 1, and max indicates the maximum of Scr/κ or 1.
Statistical Analyses
Results were expressed as percentages for dichotomous variables and mean±standard deviation (SD) or median (interquartile range) for continuous variables. tHcy levels and other biomarkers were normalized by natural logarithm transformation as necessary. All analyses were performed at a mean follow‐up interval of 4.8 years.
Plasma tHcy levels at baseline were categorized as quartile 1 (≤13.6 mmol/L, n=370), quartile 2 (13.7–16.7 mmol/L, n=352), quartile 3 (16.8–20.9 mmol/L, n=372), and quartile 4 (≥21.0 mmol/L, n=353). The participatns were classified into four groups according to baseline tHcy levels to analyze tHcy as a predictor of arterial stiffness. Furthermore, we investigated the association of baseline tHcy levels with the change in arterial stiffness from baseline to follow‐up and the association between the change in tHcy and the change in arterial stiffness. The change in tHcy levels was expressed as tHcyδ (tHcyfollow‐up−tHcybaseline). tHcyδ was categorized as tHcyδI (tHcyfollow‐up−tHcybaseline<0) and tHcyδII (tHcyfollow‐up−tHcybaseline≥0). The change in arterial stiffness was expressed as PWVδ (PWVfollow‐up−PWVbaseline). PWVδ was categorized as PWVδI (PWVfollow‐up−PWVbaseline<0) and PWVδII (PWVfollow‐up−PWVbaseline≥0).
A Pearson regression analysis and a stepwise multivariate linear regression analysis were performed to evaluate the associations between baseline tHcy and follow‐up arterial stiffness indices (cf‐PWV and carotid‐radial PWV) adjusting for age, sex, current smoking, hypertension, antihypertensive agents, DM, BMI, eGFR, and levels of plasma TG, TC, HDL‐C, LDL‐C, UA, and Cr. tHcy and TG levels were expressed as the natural logarithm.
In addition, a forward stepwise multivariate logistic regression analysis was performed to obtain odds ratios (ORs) and 95% confidence intervals (CIs). We investigated the association of baseline tHcy levels with follow‐up cf‐PWV (PWV ≥12 m/s vs PWV <12 m/s), baseline tHcy levels with the change in arterial stiffness (cf‐PWVδI vs cf‐PWVδII), and the change in tHcy level (tHcyδI vs tHcyδII) with the change in arterial stiffness (cf‐PWVδI vs cf‐PWVδII) with logistic regression. Regression models were adjusted for age and sex (model 1) and for hypertension, antihypertensive agents, DM, current smoking, BMI, eGFR, and levels of plasma TG, TC, HDL‐C, LDL‐C, UA, and Cr (model 2).
All analyses were conducted using SPSS software for Windows, version 13.0 (SPSS, Chicago, IL). P values <.05 were considered statistically significant.
Results
Clinical Characteristics of the Participants
A total of 1447 participants (mean age, 61.40 years; 59.98% women) were eligible for analysis (Table 1). The mean cf‐PWV gain was 0.74±3.83 m/s during the 4.8‐year period, and there was a significant difference in mean cf‐PWV gain across the quartiles of tHcy (quartile 1 vs quartile 4: 0.29±2.06 m/s vs 1.09±2.41 m/s, respectively; P<.001).
Table 1.
Characteristics of the Patients Categorized by tHcy Levels at Baseline
| Variable | Overall | Quartile 1 ≤13.6 | Quartile 2 13.7–16.7 | Quartile 3 16.8–20.9 | Quartile 4 ≥21.0 |
|---|---|---|---|---|---|
| Patients, No. | 1447 | 370 | 352 | 372 | 353 |
| Age, y | 61.40±11.4 | 53.13±10.24 | 57.09±11.59a | 61.69±11.86a | 62.27±12.88a |
| Women, No. (%) | 868 (59.98) | 310 (83.78)a | 252 (71.59)a | 223 (59.94)a | 83 (23.51)a |
| Current smoking, No. (%) | 380 (26.26) | 81 (21.89) | 55 (15.62)b | 96 (25.80) | 148 (42.92)a |
| Current alcohol use, No. (%) | 274 (18.93) | 63 (17.02) | 49 (13.92) | 75 (20.16) | 87 (24.64)b |
| BMI | |||||
| Baseline | 25.41±3.32 | 23.25±7.91 | 23.18±7.50 | 24.06±6.81 | 25.26±5.97a |
| Follow‐up | 25.72±6.98 | 25.34±3.57 | 26.04±12.44 | 25.53±4.18 | 25.99±3.27b |
| TG, mmol/L | |||||
| Baseline | 1.90±1.24 | 1.68±1.22 | 1.85±1.20 | 1.74±1.05 | 1.87±1.27 |
| Follow‐up | 1.51±0.98 | 1.50±1.01 | 1.58±1.03 | 1.48±1.05 | 1.47±0.82 |
| TC, mmol/L | |||||
| Baseline | 5.03±0.93 | 4.91±1.09 | 5.15±0.98a | 4.99±0.97 | 4.92±0.96 |
| Follow‐up | 5.17±1.08 | 5.32±1.08 | 5.16±1.04b | 5.17±1.13 | 5.01±1.04a |
| HDL‐C, mmol/L | |||||
| Baseline | 1.38±0.36 | 1.42±0.44 | 1.39±0.37 | 1.40±0.33 | 1.28±0.33a |
| Follow‐up | 1.43±0.59 | 1.45±0.38 | 1.45±0.75 | 1.45±0.38 | 1.36±0.75 |
| LDL‐C, mmol/L | |||||
| Baseline | 2.91±0.71 | 2.79±0.77 | 3.00±0.74a | 2.92±0.76b | 2.88±0.70 |
| Follow‐up | 3.18±1.07 | 3.25±0.90 | 3.18±1.19 | 3.14±0.95 | 3.16±1.19 |
| tHcy, μmol/L | |||||
| Baseline | 18.41±8.23 | 10.98±1.99 | 15.13±0.86a | 18.56±1.21a | 28.72±9.35 |
| Follow‐up | 16.29±6.57 | 11.98±3.07 | 13.93±3.09a | 16.17±3.54a | 22.93±8.65a |
| SBP, mm Hg | |||||
| Baseline | 128.74±17.71 | 116.07±17.98 | 132.23±18.60 | 119.66±16.95b | 122.06±19.74 |
| Follow‐up | 131.11±17.43 | 126.87±16.42 | 130.51±16.38 | 132.76±17.44a | 134.20±18.59 |
| DBP, mm Hg | |||||
| Baseline | 76.92±10.23 | 76.53±12.00 | 75.97±12.18 | 78.72±12.70 | 77.05±13.24 |
| Follow‐up | 75.29±10.99 | 75.15±10.61 | 74.79±10.64 | 74.09±10.36 | 77.12±10.89 |
| FBG, mmol/L | |||||
| Baseline | 5.39±1.65 | 5.35±1.66 | 5.46±1.95 | 5.29±1.44 | 5.30±1.46 |
| Follow‐up | 5.59±1.72 | 5.59±1.71 | 6.55±1.66 | 5.79±1.83 | 5.59±1.34 |
| UA, mmol/L | |||||
| Baseline | 292.34±69.62 | 258.26±77.19 | 296.35±77.64a | 297.43±71.77a | 312.57±91.91a |
| Follow‐up | 310.52±79.81 | 281.12±69.86 | 307.50±76.77a | 315.03±73.60a | 337.62±87.97a |
| Cr | |||||
| Baseline | 66.14±18.16 | 59.99±14.37 | 63.08±16.78 | 66.96±17.19a | 74.39±20.03a |
| Follow‐up | 74.32±17.54 | 65.40±12.39 | 71.09±14.67 | 75.42±15.02a | 85.15±20.74a |
| eGFR, mL/min/1.73 m2 | |||||
| Baseline | 94.2±14.3 | 98.25±14.98 | 98.70±13.14 | 94.71±14.49 | 89.47±32.07a |
| Follow‐up | 92.82±15.50 | 100.70±9.51 | 94.65±16.06 | 89.89±13.68a | 86.32±17.66a |
| cf‐PWV | |||||
| Baseline | 11.20±2.79 | 10.33±2.59 | 10.81±2.61b | 11.65±3.04a | 12.02±3.01a |
| Follow‐up | 11.93±4.10 | 10.50±2.21 | 11.57±2.70a | 12.51±2.94a | 13.14±6.53a |
| Carotid‐radial PWV | |||||
| Baseline | 9.38±1.47 | 9.14±1.41 | 9.36±1.46 | 9.38±1.40b | 9.64±1.54a |
| Follow‐up | 9.02±1.39 | 9.24±1.48 | 9.04±1.50 | 9.30±1.43 | 9.21±1.39 |
| Hypertension, No. (%) | 755 (52.17) | 139 (37.57) | 185 (52.55) | 199 (53.49) | 232 (65.72) |
| Antihypertensive agents | |||||
| CCB, No. (%) | 244 (32.32) | 35 (25.18) | 69 (37.29) | 74 (37.19) | 65 (28.02) |
| Diuretics, No. (%) | 94 (12.45) | 13 (9.35) | 24 (12.97) | 28 (14.07) | 29 (8.73) |
| β‐Blockers, No. (%) | 172 (22.78) | 40 (16.74) | 28 (15.14) | 34 (17.09) | 70 (30.17) |
| ACE inhibitor, No. (%) | 111 (14.70) | 20 (14.38) | 27 (14.59) | 26 (13.07) | 38 (16.37) |
| ARB, No. (%) | 74 (9.80) | 13 (9.35) | 15 (8.11) | 23 (11.56) | 23 (9.91) |
Abbreviations: ACE, angiotensin‐converting enzyme; ARB, angiotensin II receptor blocker; BMI, body mass index; CCB, calcium channel blocker; Cr, creatinine; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; FBG, fasting blood glucose; HDL‐C, high‐density lipoprotein cholesterol; 2‐HpBG, 2‐hour postprandial blood sugar; LDL‐C, low‐density lipoprotein cholesterol; PWV, pulse‐wave velocity; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; tHcy, total homocysteine; UA, uric acid. Characteristics are reported as percentages for categorical variables and means (±standard deviation) or medians (interquartile range) for continuous variables. The study participants were divided into four groups based on the baseline levels of the quartile of tHcy (≤13.6, 13.7–16.7, 16.8–20.9, ≥21.0 mmol/L). Categorical variables are presented as counts and percentages. The values outside the parentheses are the number of patients and the values inside the parentheses are prevalence. The quartile 1 level of tHcy was used as the reference and quartiles 2, 3, and 4 evaluated vs quartile 1. a P<.05 vs quartile 1. b<P.01 vs quartile 1.
Association of Baseline tHcy With Follow‐Up Arterial Stiffness
The association between baseline tHcy and arterial stiffness assessed by cf‐PWV and carotid‐radial PWV at follow‐up as a continuous variable (natural logarithm transformed) was analyzed. Pearson correlation showed that tHcy had a significant relationship with cf‐PWV (r=0.274; P<.001) and a weak relationship with follow‐up carotid‐radial PWV (r=0.079; P=.006). In multivariable linear regression analysis, tHcy was independently associated with cf‐PWV (β=0.817, P=.015) but not carotid‐radial PWV (β=0.385, P=.152) (Table 2).
Table 2.
Univariate and Stepwise Multiple Linear Regression Analysis of Baseline Parameters and Follow‐Up Arterial Stiffness
| Pearson Correlation | Multiple Linear Correlation | |||
|---|---|---|---|---|
| r | P Value | β | P Value | |
| cf‐PWV | ||||
| tHcya | 0.274 | <.001 | 0.817 | .015 |
| Age | 0.474 | <.001 | 0.094 | <.001 |
| Mean arterial pressure | 0.082 | .005 | 0.036 | .002 |
| Carotid‐radial PWV | ||||
| tHcya | 0.079 | .006 | 0.385 | .152 |
| Mean arterial pressure | 0.120 | <.001 | 0.028 | .002 |
| HDL | −0.077 | .008 | −1.091 | .043 |
Abbreviations: PWV, pulse wave velocity; tHcy, total homocysteine. r, Pearson correlation coefficient; β, stepwise multiple linear correlation coefficient. Covariates in the multiple‐adjusted models included age, sex, height, weight, current smoking, hypertension, mean arterial pressure, diabetes mellitus, body mass index, estimated glomerular filtration rate, total cholesterol, high‐density lipoprotein (HDL) cholesterol, low‐density lipoprotein cholesterol, and creatinine. aNatural logarithm‐transformed.
A stepwise logistic regression model was fitted, and the quartile 1 level of tHcy was used as the reference. cf‐PWV was significantly associated with the quartile 4 levels of tHcy (OR, 4.935; 95% CI, 3.437–7.086; P<.001). This association was generally stronger than those observed for quartile 2 (OR, 2.126; 95% CI, 1.473–3.068; P<.001) and quartile 3 (OR, 3.922; 95% CI, 2.744–5.606; P<.001). In the adjusted models (model 2), carotid‐femoral PWV was significantly associated with the quartile 4 level of tHcy (OR, 2.304; 95% CI, 1.261–4.209; P=.007) and quartile 3 level of tHcy (OR, 1.833; 95% CI, 1.069–3.142; P=.028) but not quartile 2 (OR, 1.481; 95% CI, 0.873–2.511; P=.145) (Table 3).
Table 3.
Logistic Regression Analysis for the Association Between Baseline Levels of Homocysteine and Follow‐Up cf‐PWV and cf‐PWVδ
| Quartile 2 vs Quartile 1 13.7–16.7 vs ≤13.6 | Quartile 3 vs Quartile 1 16.8–20.9 vs ≤13.6 | Quartile 4 vs Quartile 1 ≥21.0 vs ≤13.6 | ||||
|---|---|---|---|---|---|---|
| OR (95% CI) | P Value | OR (95% CI) | P Value | OR (95% CI) | P Value | |
| cf‐PWV | ||||||
| Unadjusted | 2.126 (1.473–3.068) | <.001 | 3.922 (2.744–5.606) | <.001 | 4.935 (3.437–7.086) | <.001 |
| Model 1 | 1.557 (1.046–2.317) | .029 | 1.769 (1.157–2.705) | .009 | 2.060 (1.307–3.246) | .002 |
| Model 2 | 1.481 (0.873–2.511) | .145 | 1.833 (1.069–3.142) | .028 | 2.304 (1.261–4.209) | .007 |
| cf‐PWVδ | ||||||
| Unadjusted | 1.300 (0.893–1.892) | .172 | 1.453 (1.050–2.011) | .024 | 1.588 (1.143–2.205) | .006 |
| Model 1 | 1.327 (0.906–1.945) | .146 | 1.700 (1.183–2.442) | .004 | 1.670 (1.119–2.492) | .012 |
| Model 2 | 1.656 (1.011–2.711) | .045 | 1.757 (1.097–2.813) | .019 | 1.756 (1.058–2.915) | .029 |
Abbreviations: cf‐PWV, carotid‐femoral pulse wave velocity; cf‐PWVδ, change in carotid‐femoral pulse wave velocity; tHcy, total homocysteine. Levels of tHcy were Ln‐transformed to normalize their distributions. Data are presented as odds ratios (ORs; per standard deviation increase in LntHcy level) and corresponding 95% confidence intervals (CIs). In the logistic regression model, cf‐PWV (PWV ≥12 m/s vs PWV<12 m/s) and PWVδ (PWVδ >0 vs PWVδ ≤0) were treated as the dependent variables. Model 1: adjusted for age, sex; model 2: adjusted for age, sex, hypertension, antihypertensive drugs, diabetes mellitus, current smoking, body mass index, estimated glomerular filtration rate, and levels of serum triglycerides, total cholesterol, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, uric acid, and creatinine.
The relationship between quartile levels of tHcy and carotid‐femoral PWVδ is shown in Table 3. A stepwise logistic regression showed a significant association between quartile 4 tHcy level and PWVδ (OR, 1.756; 95% CI, 1.058–2.915; P=.029), generally stronger than those observed for quartile 3 (OR, 1.757; 95% CI, 1.097–2.8131; P=.019) and quartile 2 (OR, 1.656; 95% CI, 1.011–2.711; P=.045) in model 2.
Association of the Change in tHcy and the Change in Arterial Stiffness Between Baseline and Follow‐Up
The relationship between tHcyδ and carotid‐femoral PWVδ is shown in Table 4. No significant association between them was shown after stepwise logistic regression.
Table 4.
Logistic Regression Analysis for the Association Between tHcyδ and cf‐PWVδ
| cf‐PWVδ | tHcyδ | |
|---|---|---|
| OR (95%CI) | P Value | |
| Unadjusted | 0.846 (0.660–1.129) | .284 |
| Model 1 | 0.260 (0.653–1.122) | .260 |
| Model 2 | 0.818 (0.611–1.095) | .177 |
Abbreviations: cf‐PWVδ, change in carotid‐femoral pulse wave velocity; tHcyδ, change in total homocysteine. Data are presented as odds ratios (ORs; per standard deviation increase in Ln tHcy level) and corresponding 95% confidence intervals (CIs). In the logistic regression model, PWVδ (PWVδ >0 vs PWVδ ≤0) was treated as the dependent variable. Model 1: adjusted for age, sex; model 2: adjusted for age, sex, hypertension, antihypertensive drugs, diabetes mellitus, current smoking, body mass index, estimated glomerular filtration rate, and levels of serum triglycerides, total cholesterol, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, uric acid, and creatinine.
Discussion
In the present longitudinal study, we observed associations of plasma tHcy with indices of arterial stiffness. We found that levels of tHcy were positively associated with central arterial stiffness (cf‐PWV) but not with peripheral arterial stiffness (carotid‐radial PWV). Importantly, the tHcy level was an independent predictor of central arterial stiffness after controlling for age and sex as well as other conventional cardiovascular risk factors. Furthermore, the tHcy levels showed a significant association with increases in cf‐PWV (cf‐PWVδ). On the other hand, we found no association between tHcyδ and cf‐PWVδ.
To the best of our knowledge, few studies have investigated the association between tHcy and arterial stiffness, and there have been no previous longitudinal community‐based studies. Bortolotto and colleagues16 found an association between higher tHcy levels and higher cf‐PWV in 236 patients with hypertension. The B‐Vitamins for the Prevention of Osteoporotic Fractures (B‐PROOF) trial17 reported that tHcy was associated with arterial stiffness, primarily in the oldest patients. Recently, the Framingham Heart Study,18 which included a large sample of nearly 2000 men and women, found associations of tHcy with central pulse and forward wave pressure, whereas another Framingham report19 did not find a significant relationship between tHcy and arterial stiffness. In the present study, with a median follow‐up interval of 4.8 years, we found that tHcy was an independent predictor of arterial stiffness. These discrepancies may be attributed to the following conditions. First, previous results were based on cross‐sectional study. Because of the cross‐sectional design and its inherent limitations, the previous study could not identify whether tHcy is a predictive factor for arterial stiffness. Accordingly, we designed this longitudinal study, which not only tracked each participant's outcome (eg, arterial stiffness) but also repeatedly measured risk factors that would change with time (eg, tHcy). Second, the study participants were different. Plasma total tHcy levels vary by age and have significant ethnic‐ and sex‐dependent differences.12, 13, 14 tHcy level is higher in China than in other countries, and it is much higher in people in northern China than in those living in southern areas because of the lower intake of high‐folate–containing green leafy vegetables among northerners.20 In the present study, the average tHcy level was 18.28±7.65 μmol/L in persons from Beijing, which is in northern China.
The mechanisms underlying the relationship between tHcy and arterial stiffness are not yet fully understood. However, it is known that elevated homocysteine‐induced oxidative inactivation of nitric oxide,21 which is a strong relaxing factor, has been obtained in cultured endothelial cells in vitro. Moreover, evidence for elevated oxidative inactivation of nitric oxide during hyperhomocysteinemia has been observed in animals using both genetic22, 23 and pharmacologic24, 25, 26, 27, 28 approaches. Hyperhomocysteinemia‐induced endothelial dysfunction may further promote proliferation of vascular smooth muscle cells29, 30 and increase collagen synthesis of vascular smooth muscle cells.31 Homocysteine also induces the expression of monocyte chemoattractant protein 1,32, 33, 34 which promotes the binding of monocytes to the endothelium and their recruitment to the subendothelial cell space, a critical step in atherosclerotic lesion development. Furthermore, a cross‐sectional study35 found that enhanced arterial stiffness in hyperhomocysteinemia might be attributed, in part, to homocysteine‐related LDL atherogenicity, such as small LDL particle size and oxidative modification of LDL.
Interestingly, there was no association between tHcyδ and PWVδ, showing that only baseline tHcy levels, but not change in tHcy levels, are associated with increase in arterial stiffness. These results are consistent with those reported by the Study of the Effectiveness of Additional Reductions in Cholesterol and Homocysteine (SEARCH)36 and the Heart Outcomes Prevention Evaluation‐2 (HOPE‐2) trial. 37
In the present study, we found that the effects of tHcy on central arteries (predominantly elastic arteries) and peripheral arteries (predominantly muscular arteries) are different. This finding may be attributed to the composition of arterial wall material, namely, smooth muscle cells and extracellular matrix, which is a strong determinant of PWV. An experimental study found that mild hyperhomocysteinemia causes an arterial site–dependent deterioration of the elastic structure involving metalloproteinase‐related elastolysis.38
Study Strengths
Our study has several strengths. First, it is the first analysis on the association of tHcy with arterial stiffness based on a longitudinal study that used baseline and follow‐up data; thus, it differs from cross‐sectional studies or cohort studies that describe only baseline tHcy measurements. Second, unlike the previous studies that selected participants based on either sex or homocysteine level differences, the present study was based on a sample in which selection bias was inherently low.
Study Limitations
The present study has limitations. First, it was performed on Chinese residents from two communities in Beijing; thus, the results may not represent Chinese individuals from other areas. Second, 181 patients (10.7%) were lost to follow‐up. This loss is an unavoidable limitation of epidemiological studies that may be biased toward the null hypothesis because of the loss of cases that presumably had more extreme values for the analyzed variables. Third, folate and vitamin B12 levels were not measured in our study. Information on these levels may be helpful for explaining the relationship between tHcy and arterial stiffness. Fourth, the lack of assessment of atherosclerosis such as stress tests and carotid intimal‐medial thickness is a limitation of this study. Atherosclerosis data may be helpful in identifying the relationship between tHcy and coronary artery atherosclerosis in this community‐based population.
Conclusions
The present community‐based prospective study clearly demonstrated an association between tHcy level and arterial stiffness, indicating that tHcy is an independent predictive factor for arterial stiffness.
Acknowledgments and Disclosures
We thank colleagues at the Department of Laboratory Medicine, the PLA General Hospital, for help with biochemical measurements. We are also grateful to all study participants for their participation in the study. This research is supported by a grant from the Key National Basic Research Program of China (2012CB517503, 2013CB530804) and the Key Science and Technology Foundation of China (2012ZX09303004‐002) to Dr Ping Ye.
J Clin Hypertens (Greenwich). 2015;17:594–600. DOI: 10.1111/jch.12555. © 2015 Wiley Periodicals, Inc.
References
- 1. Laurent S, Cockcroft J, Van Bortel L, et al. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J. 2006;27:2588–2605. [DOI] [PubMed] [Google Scholar]
- 2. Vlachopoulos C, Aznaouridis K, Stefanadis C. Clinical appraisal of arterial stiffness: the Argonauts in front of the Golden Fleece. Heart. 2006;92:1544–1550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. O'Rourke MF, Staessen JA, Vlachopoulos C, et al. Clinical applications of arterial stiffness; definitions and reference values. Am J Hypertens. 2002;15:426–444. [DOI] [PubMed] [Google Scholar]
- 4. McEniery C, Cockcroft JR. Does arterial stiffness predict atherosclerotic coronary events? Adv Cardiol. 2007;44:160–172. [DOI] [PubMed] [Google Scholar]
- 5. Kullo IJ, Malik AR. Arterial ultrasonography and tonometry as adjuncts to cardiovascular risk stratification. J Am Coll Cardiol. 2007;49:1413–1426. [DOI] [PubMed] [Google Scholar]
- 6. Santelices LC, Calano SJ, Erhart JC, et al. Experimental system for ex vivo measurement of murine aortic stiffness. Physiol Meas. 2007;28:N39–N49. [DOI] [PubMed] [Google Scholar]
- 7. Avolio AP, Chen S, Wang R, et al. Effects of aging on changing arterial compliance and left ventricular load in a northern Chinese urban community. Circulation. 1983;68:50–58. [DOI] [PubMed] [Google Scholar]
- 8. Asmar R, Benetos A, Topouchian J, et al. Assessment of arterial distensibility by automatic pulse wave velocity measurement: validation and clinical application study. Hypertension. 1995;26:485–490. [DOI] [PubMed] [Google Scholar]
- 9. Mayer O, Filipovský J, Dolejsová M, et al. Mild hyperhomocysteinaemia is associated with increased aortic stiffness in general population. J Hum Hypertens. 2006;20:267–271. [DOI] [PubMed] [Google Scholar]
- 10. Feng SQ, Ye P, Luo LM, et al. Associations of plasma homocysteine and high‐sensitivity C‐reactive protein levels with arterial stiffness in Chinese population: a community‐based study. Chin Med J (Engl). 2012;125:44–49. [PubMed] [Google Scholar]
- 11. Vyssoulis G, Karpanou E, Kyvelou SM, et al. Associations between plasma homocysteine levels, aortic stiffness and wave reflection in patients with arterial hypertension, isolated office hypertension and normotensive controls. J Hum Hypertens. 2009;24:183–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Must A, Jacques PF, Rogers G, et al. Serum total homocysteine concentrations in children and adolescents: results from the Third National Health and Nutrition Examination Survey (NHANES III). J Nutr. 2003;133:2643–2649. [DOI] [PubMed] [Google Scholar]
- 13. Jacques PF, Rosenberg IH, Rogers G, et al. Serum total homocysteine concentrations in adolescent and adult Americans: results from the third National Health and Nutrition Examination Survey. Am J Clin Nutr. 1999;69:482–489. [DOI] [PubMed] [Google Scholar]
- 14. Malinow MR, Ducimetiere P, Luc G, et al. Plasma homocysteine levels and graded risk for myocardial infarction: findings in two populations at contrasting risk for coronary heart disease. Atherosclerosis. 1996;126:27–34. [DOI] [PubMed] [Google Scholar]
- 15. AlbertiK G, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120:1640–1645. [DOI] [PubMed] [Google Scholar]
- 16. Bortolotto LA, Safar ME, Billaud E, et al. Plasma homocysteine, aortic stiffness, and renal function in hypertensive patients. Hypertension. 1999;34:837–842. [DOI] [PubMed] [Google Scholar]
- 17. van Dijk SC, Smulders YM, Enneman AW, et al. Homocysteine level is associated with aortic stiffness in elderly: cross‐sectional results from the B‐PROOF study. J Hypertens. 2013;31:952–959. [DOI] [PubMed] [Google Scholar]
- 18. Levy D, Hwang SJ, Kayalar A, et al. Associations of plasma natriuretic peptid, adrenomedullin, and homocysteine levels with alterations in arterial stiffness: the Framingham Heart Study. Circulation. 2007;115:3079–3085. [DOI] [PubMed] [Google Scholar]
- 19. Lieb W, Larson MG, Benjamin EJ, et al. Multimarker to evaluate correlates of vascular stifiness: the Framingham Heart Study. Circulation. 2009;119:37–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Hao L, Ma J, Zhu J, et al. High prevalence of hyperhomocysteinemia in Chinese adults is associated with low folate, vitamin B‐12, and vitamin B‐6 status. J Nutr. 2007;137:407–413. [DOI] [PubMed] [Google Scholar]
- 21. Starkebaum G, Harlan JM. Endothelial cell injury due to coppercatalyzed hydrogen peroxide generation from homocysteine. J Clin Invest. 1986;77:1370–1376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Weiss N, Zhang YY, Heydrick S, et al. Overexpression of cellular glutathione peroxidase rescues homocyst(e)ine‐induced endothelial dysfunction. Proc Natl Acad Sci USA. 2001;98:12503–12508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Dayal S, Brown KL, Weydert CJ, et al. Deficiency of glutathione peroxidase‐1 sensitizes hyperhomocysteinemic mice to endothelial dysfunction. Arterioscler Thromb Vasc Biol. 2002;22:1996–2002. [DOI] [PubMed] [Google Scholar]
- 24. Eberhardt RT, Forgione MA, Cap A, et al. Endothelial dysfunction in a murine model of mild hyperhomocyst(e)inemia. J Clin Invest. 2000;106:483–491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Kanani PM, Sinkey CA, Browning RL, et al. Role of oxidant stress in endothelial dysfunction produced by experimental hyperhomocyst(e)inemia in humans. Circulation. 1999;100:1161–1168. [DOI] [PubMed] [Google Scholar]
- 26. Chambers JC, McGregor A, Jean‐Marie J, et al. Demonstration of rapid onset vascular endothelial dysfunction after hyperhomocysteinemia. An effect reversible with vitamin C therapy. Circulation. 1999;99:1156–1160. [DOI] [PubMed] [Google Scholar]
- 27. Ungvari Z, Csiszar A, Edwards JG, et al. Increased superoxide production in coronary arteries in hyperhomocysteinemia: role of tumor necrosis factor‐alpha, NAD(P)H oxidase, and inducible nitric oxide synthase. Arterioscler Thromb Vasc Biol. 2003;23:418–424. [DOI] [PubMed] [Google Scholar]
- 28. Weiss N, Heydrick S, Zhang YY, et al. Cellular redox state and endothelial dysfunction in mildly hyperhomocysteinemic cystathionine beta‐synthase‐deficient mice. Arterioscler Thromb Vasc Biol. 2002;22:34–41. [DOI] [PubMed] [Google Scholar]
- 29. Chen C, Halkos ME, Surowiec SM, et al. Effects of homocysteine on smooth muscle cell proliferation in both cell culture and artery perfusion culture models. J Surg Res. 2000;88:26–33. [DOI] [PubMed] [Google Scholar]
- 30. Nishio E, Watanabe Y. Homocysteine as modulator of platelet‐derived growth factor action in vascular smooth muscle cell: a possible role for hydrogen preoxide. Br J Pharmacol. 1997;122:269–274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Toohey JI. Homocysteine toicity in connective tissue: theories, old and new. Connect Tissue Res. 2008;49:57–61. [DOI] [PubMed] [Google Scholar]
- 32. Poddar R, Sivasubramanian N, Dibello PM, et al. Homocysteine induces expression and secretion of monocyte chemoattractant protein‐1 and interleukin‐8 in human aortic endothelial cells: implications for vascular disease. Circulation. 2001;103:2717–2723. [DOI] [PubMed] [Google Scholar]
- 33. Wang G, O K. Homocysteine stimulates the expression of monocyte chemoattractant protein‐1 receptor (CCR2) in human monocytes: possible involvement of oxygen free radicals. Biochem J. 2001;357:233–240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Wang G, Siow YL, O K. Homocysteine induces monocyte chemoattractant protein‐1 expression by activating NF‐kappa B in THP‐1 macrophages. Am J Physiol Heart Circ Physiol. 2001;280:H2840–H2847. [DOI] [PubMed] [Google Scholar]
- 35. Yun J, Kim JY, Kim OY, et al. 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]
- 36. Lonn E, Yusuf S, Arnold MJ, et al. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med. 2006;354:1567–1577. [DOI] [PubMed] [Google Scholar]
- 37. Bowman L, Armitage J, Bulbulia R, et al. Study of the effectiveness of additional reductions in cholesterol and homocysteine (SEARCH): characteristics of a randomized trial among 12064 myocardial infarction survivors. Am Heart J. 2007;154:815–823. [DOI] [PubMed] [Google Scholar]
- 38. Charpiot P, Bescond A, Augier T, et al. Hyperhomocysteinemia induces elastolysis in minipig arteries: structural consequences, arterial site specificity and effect of captopril‐hydrochlorothiazide. Matrix Biol. 1998;17:559–574. [DOI] [PubMed] [Google Scholar]