Traditional risk factors are estimated to account for only part of the CVD risk(Reference Tolonen, Mähönen and Asplund1). Non-traditional risk factors, such as increased fasting total plasma homocysteine (tHcy), have received great attention. Abundant studies have established elevated tHcy as a potent independent risk factor for coronary artery disease, stroke and deep vein thrombosis(Reference Wald, Law and Morris2–Reference de Ruijter, Westendorp and Assendelft4). Furthermore, our previous meta-analysis(Reference Wang, Qin and Demirtas5) suggested that tHcy-lowering therapy (mainly folic acid supplementation) could reduce the risk of stroke by 18–23 %. Our recent meta-analysis(Reference Huo, Qin and Wang6), which included fifteen randomised trials with prevention of stroke as one of the study endpoints and represented the largest number of subjects included in previously published papers from meta-analysis, also proved that folic acid supplementation significantly reduced the risk of stroke by 8 % (n 55 764; relative risk (RR) 0·92; 95 % CI 0·86, 1·00; P= 0·038). In the ten trials with no or partial folic acid fortification (n 43 426), the risk of stroke was reduced by 11 % (RR 0·89; 95 % CI 0·82, 0·97; P= 0·010). Folic acid supplementation(Reference Qin, Huo and Langman7) also was found to reduce CVD risk in patients with end-stage renal disease or advanced chronic kidney disease (creatinine clearance < 30 ml/min) by 15 and by 17 % in trials that showed a larger decrease in the concentration of tHcy. Furthermore, folic acid supplementation(Reference Qin, Xu and Zhang8) could significantly reduce the progression of carotid intima-media thickness (weighted mean difference − 0·04; 95 % CI − 0·07, − 0·02; P< 0·001) and the percentage reduction of tHcy was positively related to the effect size. Similarly, a recent meta-analysis proved that the methylenetetrahydrofolate reductase (MTHFR) 677C>T variant was associated with a larger effect on tHcy concentration in regions of low folate consumption than in regions with high dietary folate intake or with established programmes of folic acid fortification of flour for the prevention of neural tube defects. A similar pattern was noted for the genetic association with stroke risk(Reference Holmes, Newcombe and Hubacek9). However, a previous report(Reference Clarke, Halsey and Lewington10) from meta-analysis with eight randomised trials found that dietary supplementation with folic acid had no significant effects within 5 years on cardiovascular events. Therefore, additional large, randomised studies, particularly in regions with insufficient folic acid supplementation and high tHcy levels, should provide further important evidence for confirming the effect of tHcy-lowering therapy on stroke prevention.
There were significant concentration–response associations between plasma tHcy and mortality, and CVD with no apparent threshold concentration(Reference Refsum, Nurk and Smith11), which means that it is difficult to define a range of safe tHcy concentrations. However, in a community-based prospective cohort study of 2009 participants in China, participants with tHcy >9·47 μmol/l had a 2·3-fold higher risk for cardiovascular events (95 % CI 1·24, 4·18; P= 0·008)(Reference Sun, Chien and Hsu12). So, to be consistent with previous reports(Reference Stanger, Herrmann and Pietrzik13, Reference McCully14), we defined hyperhomocysteinaemia as a tHcy concentration ≥ 10 μmol/l in the present study.
Stroke has been the leading cause of death in China, and its morbidity and mortality have been rapidly rising, particularly in rural areas(Reference Zhao, Liu and Wang15). Previous studies have reported that hypertension (HTN) and hyperhomocysteinaemia are the two most important modifiable risk factors for stroke. More importantly, it has been reported that an increased tHcy concentration showed a more than multiplicative effect on CVD risk in hypertensive subjects(Reference Pezzini, Grassi and Del Zotto16, Reference Graham, Daly and Refsum17). In a recent study(Reference Towfighi, Markovic and Ovbiagele18), individuals with a combination of elevated tHcy ( ≥ 10 μmol/l) and HTN were substantially more likely to have prevalent stroke compared with individuals without either condition (men: OR 12·02, 95 % CI 6·36, 22·73; women: OR 17·34, 95 % CI 10·49, 28·64). However, to our knowledge, no previous publication has studied the prevalence of hyperhomocysteinaemia in Chinese hypertensive adults, particularly in rural areas. The present study was conducted to investigate the prevalence of hyperhomocysteinaemia and its major determinants in rural Chinese hypertensive patients.
Subjects and methods
Study population
The study subjects were participants of an ongoing China Stroke Primary Prevention Trial (CSPPT, clinicaltrials.gov identifier: NCT00794885). The CSPPT is a multi-centre randomised controlled trial designed to confirm that enalapril maleate and folic acid tablets combined is more effective in preventing stroke among patients with HTN when compared with enalapril maleate alone. Details regarding inclusion/exclusion criteria, treatment assignment and outcome measures of the trial have been described elsewhere (http://clinicaltrials.gov/ct2/show/NCT00794885). In the present study, we included subjects from Lianyungang who participated in the double-blinded treatment phase of the trial in which tHcy measurements were performed.
Briefly, we conducted a community-based screening in twenty townships within two counties (Ganyu, which is coastal, and Donghai, which is inland) in Lianyungang of Jiangsu province, China, from October 2008 to September 2009. The inclusion criteria were as follows: (1) aged 45–75 years and (2) seated systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg at both of the two screening visits (with at least 24 h between the visits) or currently under antihypertensive treatment. Participants were excluded if they reported a history of myocardial infarction, stroke, heart failure, cancer or serious mental disorders; or they were unwilling to participate in the survey. The present study was approved by the Ethics Committee of the Institute of Biomedicine, Anhui Medical University, Hefei, China. Written informed consent was obtained from each participant before data collection.
Data collection procedures
In the screening phase, researchers and village doctors travelled to participating communities to screen local residents for hypertensive patients. Candidate hypertensive patients were then invited to the local study centres for a formal recruitment visit. At the beginning of the recruitment visit, each participant was asked to provide written informed consent in compliance with the Declaration of Helsinki and the requirements of the ethics committee.
Baseline data were collected by trained research staff according to the standard operating procedure. Each participant was interviewed using a standardised questionnaire designed specifically for the present study. The question about standard of living was phrased as follows: ‘How does your standard of living compare to others?’; and a choice of three responses was given as follows: poor, fair and good. The question about physical activity was phrased as follows: ‘How do you describe your daily physical activity level?’; and a choice of three responses was provided as follows: low, moderate and high. Finally, the question about fruit and green vegetable consumption was phrased as follows: ‘How much fruit and green vegetables do you eat (count the annually averaged weekly intake of fruits and green vegetables)?’; and a choice of three responses regarding weekly intake was given as follows: < 1 jin ( < 500 g), 1–3 jin (500–1500 g) and ≥ 3 jin ( ≥ 1500 g).
Anthropometric measurements, including height, weight and waist circumference (WC), were taken using the standard operating procedure. Height was measured without shoes to the nearest 0·1 cm on a portable stadiometer. Weight was measured in light indoor clothing without shoes to the nearest 0·1 kg. BMI was calculated as weight (kg)/height (m2). WC was measured as the minimum circumference between the inferior margin of the ribcage and the crest of the ileum.
Seated blood pressure measurements were obtained by trained research staff after subjects had been seated for 10 min using a mercury manometer, and using the standard method of calibration and appropriately sized cuffs, according to the standard operating procedure. Triplicate measurements on the same arm were taken, with at least 2 min between readings. Each patient's SBP and DBP were calculated as the mean of three independent measures. Blood pressure measured at visit 2 was used for analysis.
Blood sample collection and laboratory methods
After 12–15 h of fasting, a venous blood sample was obtained from each subject. Serum or plasma samples were separated within 30 min of collection and were stored at − 70°C. Plasma tHcy was measured by an enzyme cycling method using a Hitachi 7020 Automatic Analyzer (Hitachi). Serum creatinine was measured by a modified kinetic rate Jaffe reaction method using a Dade Dimension Chemistry Analyzer (Siemens). The MTHFR 677C>T genotype was determined by the Taqman assay designed and manufactured by Applied Biosystems.
Statistical analysis
HTN was categorised into three grades: grade 1, SBP 140–159 and/or DBP 90–99 mmHg; grade 2, SBP 160–179 and/or DBP 100–109 mmHg; grade 3, SBP ≥ 180 and/or DBP ≥ 110 mmHg. Treated HTN was defined as receiving antihypertensive medication within the past 2 weeks. Current smoker was defined as having smoked at least one cigarette per d or eighteen packs or more in the last year. Current drinking was defined as drinking alcohol at least twice per week in the last year. Glomerular filtration rate (GFR) was estimated by using the Cockcroft–Gault equation adjusted for body surface area and corrected for the bias in the Modification of Diet in Renal Disease (MDRD) Study sample(Reference Levey, Coresh and Greene19):
$$\begin{eqnarray} GFR\,(ml/min\,per\,1\cdot 73\hairsp m^{2}) = 0\cdot 8\times ((140 - age)\times weight\,(kg))\times 0\cdot 85\,(if\,female)\times 1\cdot 73/(72\times body\,surface\,area\times serum\,creatinine). \end{eqnarray}$$
Body surface area was calculated using the DuBios method(Reference DuBois and DuBois20):
$$\begin{eqnarray} Body\,surface\,area\,(m^{2}) = 0\cdot 007184\times height\,(cm)^{0\cdot 725}\times weight\,(kg)^{0\cdot 425}. \end{eqnarray}$$
Hyperhomocysteinaemia was defined as a tHcy concentration ≥ 10 μmol/l.
Means and proportions were calculated for population characteristics by sex. The difference in population characteristics was compared using Student's t tests or χ2 test. The adjusted OR and 95 % CI of having hyperhomocysteinaemia were determined from logistic regression models that included age group (45–54, 55–64 and 65–75 years), sex, BMI ( ≥ 25 v. < 25 kg/m2), cigarette smoking, alcohol drinking, GFR (>90, 60–90 and < 60 ml/min per 1·73 m2), antihypertensive treatment status (treated and untreated), HTN grades (controlled blood pressure or grade 1 HTN, grade 2 HTN and grade 3 HTN), geographic region (coastal and inland), season, standard of living (poor, fair and good), fruit and green vegetable consumption ( < 1 jin ( < 500 g), 1–3 jin (500–1500 g) and ≥ 3 jin ( ≥ 1500 g)), education level (illiterate, primary level, and elementary or higher level), physical activity level (low, moderate and high) and MTHFR 677C>T polymorphism (CC, CT and TT). Both sex-specific and sex-combined regression analyses were performed using the above model. In the sex-combined model, an interaction term between sex and the genotype was added. Similar linear models for ln-transformed tHcy concentrations were also analysed. All statistical analyses were performed using SAS 8.2 (SAS Institute).
Results
Overall, plasma tHcy was measured in 16 441 participants aged 45–75 years with HTN. In the present study, study participants with vitamin use (n 256), CVD (n 474), cancer (n 36), diabetes (n 542), dyslipidaemia (n 371), or with any missing data (n 816) on antihypertensive treatment status, age, sex, height, weight, WC, smoking status, drinking status, standard of living, fruit and green vegetable consumption, education, physical activity level and MTHFR 677C>T polymorphism were excluded. The final analysis included 13 946 participants (5421 men and 8525 women).
The population characteristics by sex are listed in Table 1. Men had significantly higher values for age, DBP, percentage of cigarette smoking and alcohol drinking, and levels of living standard and education, and had lower values for SBP, BMI and percentage of antihypertensive treatment.
Population characteristics by sex (Mean values and standard deviations; number of participants and percentages; geometric means and interquartile ranges (IQR))
SBP, systolic blood pressure; DBP, diastolic blood pressure; HTN, hypertension; BP, blood pressure, GFR, glomerular filtration rate; MTHFR, methylenetetrahydrofolate reductase.
* A total of 388 subjects with antihypertensive treatment and controlled BP were included.
The geometric mean of tHcy was 12·0 (interquartile range (IQR) 9·3–14·2) μmol/l in men and 9·6 (IQR 7·8–11·6) μmol/l in women. The prevalence of hyperhomocysteinaemia ( ≥ 10 μmol/l) was 51·6 % (women 42·7 % and men 65·6 %) (Table 2).
Adjusted* OR of having hyperhomocysteinaemia (total plasma homocysteine ≥10 μmol/l) in different subgroups (Adjusted odds ratios and 95 % confidence intervals)
MTHFR, methylenetetrahydrofolate reductase; GFR, glomerular filtration rate; HTN, hypertension; BP, blood pressure.
* All variables were included in the same model.
† A total of 388 subjects with antihypertensive treatment and controlled BP were included.
The results from the logistic regression analyses for hyperhomocysteinaemia are presented in Table 2. In general, age, male sex, higher GFR, T-allele of MTHFR 677C>T, BMI, inlanders and HTN treatment (but not HTN grade) were associated with an increased risk of hyperhomocysteinaemia, whereas the physical activity level was associated with decreased risk. Similar association patterns were observed in men and women. Similar findings were also observed in the regression analysis of ln-transformed tHcy (Table 3). In addition, smokers in both men and women, men with grade 3 HTN and male non-drinkers had higher tHcy. Most importantly, a significant interaction between sex and the MTHFR 677C>T genotype was observed for both tHcy and hyperhomocysteinaemia. While the TT genotype was a significant independent risk factor in both sexes, TT in men was associated with an additional 22 % increase in tHcy and 34 % increase in OR for hyperhomocysteinaemia (Tables 2 and 3; Fig. 1).
Relationships between homocysteine concentrations* and related factors in different subgroups† (Geometric means, interquartile ranges (IQR, first to third quartiles) and 95 % confidence intervals)
tHcy, total plasma homocysteine; MTHFR, methylenetetrahydrofolate reductase; GFR, glomerular filtration rate; HTN, hypertension; BP, blood pressure.
* Ln-transformed homocysteine concentrations were used in the regression model.
† All variables were included in the same models.
‡ Calculated as (antilog β coefficient − 1) × 100.
§ A total of 388 subjects with antihypertensive treatment and controlled BP were included.
Prevalence of hyperhomocysteinaemia (total plasma homocysteine ≥ 10 μmol/l) by methylenetetrahydrofolate reductase 677C>T polymorphism and sex (men (
), n 5421; women (□), n 8525). * There was a significant interaction effect between sex and 677CT genotype (P< 0·05). † There was a significant interaction effect between sex and 677TT genotype (P< 0·05).
Discussion
In the present study, the geometric mean tHcy concentration was 10·5 μmol/l (women 9·6 and men 12·0) and the median tHcy concentration was 10·1 μmol/l (women 9·4 and men 11·2). The prevalence of hyperhomocysteinaemia was 51·7 % (women 42·8 % and men 65·8 %). These estimates are lower than those in a previous report of hypertensive adults in different Chinese populations (Ha'erbin, Shanghai, Shenyang, Beijing, Xi'an and Nanjing; median tHcy: 12·2 μmol/l)(Reference Qin, Li and Cui21, Reference Qin, Li and Cui22)), and lie in the middle between the southerners and the northerners with respect to another study of 2471 Chinese adults aged 35–64 years(Reference Hao, Ma and Zhu23). The differences in tHcy levels described in various reports could be due to different population inclusion criteria, genetic backgrounds, as well as differences in risk factor profiles across regional areas.
Consistent with previous studies(Reference Refsum, Nurk and Smith11, Reference Jacques, Bostom and Wilson24), the present study found that a higher prevalence of hyperhomocysteinaemia or tHcy was associated with men, older age, overweight (BMI ≥ 25 kg/m2) and current smoking. However, the effect size of current smoking and current drinking on tHcy, about 4 % increase and 3 % decrease in tHcy, respectively, is relatively small. The relationship between alcohol consumption and tHcy concentration from previous reports had been inconclusive(Reference Ganji and Kafai25, Reference de Bree, Verschuren and Blom26). In the present study, current alcohol drinkers were significantly associated with decreased tHcy only in men. This is partly due to the percentage of current drinkers in women being very low (approximately 3 %), hence a larger sample size is needed to detect such a small effect size. Since BMI and WC were highly correlated (Pearson's r 0·79), we did not include BMI and WC in the same regression model.
The present study also observed a 3 % decrease in tHcy and an approximately 20 % decrease in OR for hyperhomocysteinaemia in subjects with high physical activities. A recent meta-analysis reported that moderately or highly physically active people had a lower risk of stroke incidence or mortality than those with a low level of activity(Reference Lee, Folsom and Blair27). Given the fact that elevated tHcy was more strongly associated with stroke than with IHD(Reference Wald, Law and Morris2), it would be interesting to test whether physical activity is an independent risk factor after taking into consideration of tHcy.
It is noteworthy that, in the present study, the inlanders had a higher tHcy and prevalence of hyperhomocysteinaemia, which is contrary to the results from a small study (n 208) in coastal West Africa(Reference Amouzou, Chabi and Adjalla28). A possible explanation for the observation in the present study could be a higher betaine(Reference Craig29) and vitamin B12(Reference Vogiatzoglou, Smith and Nurk30) (rich in seafood) intake in coastal areas. Unfortunately, we did not have such data in the present study.
In the present study, although participants with antihypertensive treatment had slightly lower SBP (167·9 v. 169·1 mmHg, P= 0·001) and DBP (95·1 v. 95·9 mmHg, P< 0·001) than those without antihypertensive treatment, participants with antihypertensive treatment had a higher prevalence of hyperhomocysteinaemia. Whether this was a treatment side effect is still inconclusive(Reference Poduri, Kaur and Thakur31, Reference Westphal, Rading and Luley32).
The MTHFR 677C>T polymorphism is the most important genetic determinant of plasma tHcy(Reference Hustad, Midttun and Schneede33). Furthermore, previous studies(Reference Papoutsakis, Yiannakouris and Manios34, Reference Russo, Friso and Jacques35) suggested that age and sex may modify the contribution of the MTHFR 677C>T polymorphism to tHcy concentrations under conditions of lower folate status. For example, Papoutsakis et al. (Reference Papoutsakis, Yiannakouris and Manios34) reported that only males with the MTHFR 677TT genotype had tHcy that was significantly higher than tHcy levels of C-allele carriers (P= 0·001). Russo et al. (Reference Russo, Friso and Jacques35) observed that only in younger men ( < 55 years old), TT subjects had significantly higher tHcy concentrations than those with the CT or CC genotype (P< 0·05 for either of these genotypes). In the present study, subjects with MTHFR 677TT and CT genotypes (v. CC genotype) had higher tHcy concentrations in men or women. Also, there were significant interaction effects between the MTHFR C677T polymorphism and sex on the prevalence of hyperhomocysteinaemia. Most importantly, the frequency of the MTHFR 677TT genotype was 26·4 %, similar to the result (24·8 %) found in a previous hypertensive adult study in different Chinese regions(Reference Qin, Li and Cui21, Reference Qin, Li and Cui22), but higher than that in other countries(Reference Wilcken, Bamforth and Li36). Another study has found that the MTHFR 677C>T polymorphism was one of the major determinants of stroke risk(Reference Holmes, Newcombe and Hubacek9). Thus, the high prevalence of the MTHFR 677TT genotype and its interaction effects with sex on the prevalence of hyperhomocysteinaemia may partly explain the high incidence of stroke, even with strict HTN control, in Chinese hypertensive adults, particularly in men(Reference Zhao, Liu and Wang15, Reference Kjeldsen, Julius and Hedner37).
In summary, we simultaneously evaluated the effect of a major genetic polymorphism, lifestyle, region, socio-economic status and diet on the prevalence of hyperhomocysteinaemia in Chinese hypertensive adults. The inclusion or exclusion of participants with known CVD, diabetes and dyslipidaemia in the present analyses had little effect on the results. There are limitations of the present study. The data collected on food intake and lifestyle habit are quite limited; plasma levels of folic acid and other B vitamins are not available at this time; and our sample was from hypertensive adults in one region. So, the findings from the present study may not be directly generalisable to the general population. We mainly focused on the effect of lifestyle factors (smoking, alcohol drinking, physical activity, etc.), socio-economic status and genetic background on the prevalence of hyperhomocysteinaemia.
In conclusion, there was a high prevalence of hyperhomocysteinaemia in Chinese hypertensive adults, particularly in participants in inland (v. coastal) areas. Sex, age and the MTHFR genotype were the major risk factors, while BMI, GFR, smoking, alcohol drinking and physical activity were also significant determinants. Folic acid supplementation and the concomitant lifestyle change, including smoking cessation, obesity control and improvement in physical activity levels may help to decrease tHcy and reduce the CVD risk in this population.
Acknowledgements
The study was supported by the Ministry of Science and Technology of the People's Republic of China (2012zx09101-105); the Department of Development and Reform, Shenzhen(2010)1744; the Department of Science, Industry, Trade and Information Technology, Shenzhen. The sponsors did not participate in the design or conduct of the study, collection, management, analysis, and the interpretation of the data, or preparation, review and approval of the manuscript. The authors' contributions are as follows: Y. W., X. L., X. Q., Y. C., M. H., L. S., J. L., Y. Z., G. T., B. W., N. S., Xin Xu, L. L., Xiping Xu and Y. H. participated in the design of the study; Y. W., X. L., X. Q., Y. C., M. H., L. S., J. L., Y. Z., G. T., B. W., N. S., Xin Xu, L. L., Xiping Xu and Y. H. conducted the study; Y. W., X. L., X. Q., Xin Xu, Xiping Xu and Y. H. analysed the data; Y. W., X. L., X. Q., Y. C., M. H., L. S., J. L., Y. Z., G. T., B. W., N. S., Xin Xu, L. L., Xiping Xu and Y. H. wrote the paper. All authors read and approved the final manuscript. The authors declare that they have no competing interests.
