Homocysteine, Ischemic Stroke, and Coronary Heart Disease in Hypertensive Patients
A Population-Based, Prospective Cohort Study
Background and Purpose—Total homocysteine level (tHcy) is a risk factor of ischemic stroke (IS) and coronary heart disease. However, the results are conflicting and mainly focused on healthy individuals in developed countries.
Methods—A prospective, population-based cohort study was conducted among 5935 participants from 60 communities in the city of Shenzhen, China. A Cox regression analysis was applied to evaluate the contribution of tHcy to the risk of IS and coronary heart disease. The effect of folic acid supplementation on tHcy levels was also evaluated among 501 patients with essential hypertension, who received an average of 2.5 years of folic acid supplementation.
Results—After adjustment for confounding factors, the hazard ratios (95% confidence intervals) of IS caused by hyperhomocysteinemia were 2.18 (1.65–2.89), 2.40 (1.56–3.67), and 2.73 (1.83–4.08) in the total, male, and female participants, respectively. Compared with normal levels of tHcy (<15 μmol/L), the hazard ratios (95% confidence intervals) for IS in the highest tHcy category (≥30 μmol/L) were 4.96 (3.03–8.12), 6.11 (3.44–10.85), and 1.84 (0.52–6.46) in the total, males, and females participants, respectively. However, we did not observe a significant relationship between tHcy and the risk of coronary heart disease. The 2.5 years of folic acid supplementation reduced tHcy levels by 6.7 μmol/L (27.92%) in patients with essential hypertension.
Conclusions—Hyperhomocysteinemia in Chinese hypertensive patients is significantly associated with IS risk but not coronary heart disease susceptibility, and folic acid supplementation can efficiently reduce tHcy levels.
Stroke is the leading cause of permanent disability in China,1 and ischemic stroke (IS) accounts for almost 80% of all strokes,2 whereas coronary heart disease (CHD) is a major worldwide health threat. In tandem with the economic success of China, the number of individuals in China with IS and CHD has increased significantly in recent years, with stroke resulting in 301 million disability-adjusted life-years.3 CHD in Chinese adults aged 35 to 84 years is predicted to increase by 64% during the period 2020–2029.4
Although numerous epidemiology studies have researched the association between hyperhomocysteinemia and IS or CHD, the results have been conflicting. Most studies have been retrospective and focused on healthy populations in the developed world, whereas data from developing countries remains limited.
Vitamins B12 and B6 and folate are involved in the metabolism of methionine, and deficiency of these B vitamins can cause elevation of total homocysteine (tHcy) levels.5 Deficiencies in these nutrients are more prevalent in developing countries. A meta-analysis of randomized trials demonstrated that homocysteine-lowering interventions for stroke seem not to have a significant effect in geographic regions with high dietary folate intake, but may have a substantial effect in regions with low folate intake, such as Asia.6 Regarding public health guidelines, China does not emphasize or promote the importance of taking folic acid routinely; therefore, people in China potentially have low levels of folic acid and relatively high tHcy levels. Moreover, different populations have diverse genetic backgrounds and encounter different environmental risk factors. We therefore examined the impact of hyperhomocysteinemia on the risk of IS and CHD and evaluated the effect of folic acid supplementation on tHcy levels in hypertensive patients in China. Of note, no studies to date have suggested that folic acid supplementation reduces tHcy levels in hypertensive patients.
We hypothesized that hypertension combined with hyperhomocysteinemia would greatly increase the risk of IS and CHD. The objectives of this prospective cohort study were to confirm the association between tHcy levels and the future risk of IS and CHD in hypertensive patients and to explore the effect of folic acid supplementation on tHcy levels in hypertensive patients.
This prospective longitudinal study included 5488 hypertensive patients with a 2.7-year follow-up. The study relied on the hypertension management information system in community health service centers (CHSCs). Patients were enrolled from the 60 CHSCs in the Nanshan District of the city of Shenzhen, Guangdong Province in southern China. Six to eight communities were selected from each of the 8 subdistricts in Nanshan District using a simple random procedure according to a sequence of computer-generated random numbers. The baseline survey was conducted from April 2010 to September 2011.7 The specific details of the study participants, recruitment, and baseline data collection have been described previously.7 Written informed consent was obtained from all participants. The protocol of this study was approved by the ethics committees of the collaborating hospitals.
Inclusion and Exclusion Criteria
Inclusion criteria were as follows: (1) patients with essential hypertension ≥20 years of age; (2) local residents who had lived in Shenzhen for ≥6 months; (3) native Chinese who could be followed-up without difficulty for at least 3 years; and (4) patients whose health records had been established in the CHSC.
Exclusion criteria were as follows: (1) patients with secondary hypertension, cancer, severe liver and kidney disease, or pregnancy; (2) patients taking folic acid or vitamin B6 or B12 before the follow-up study; and (3) hypertensive patients with IS or CHD.
During the 2.7-year follow-up, we defined hypertensive patients as having IS if they had subarachnoid hemorrhage, intracerebral hemorrhage, nontraumatic intracranial hemorrhage, or cerebral infarction and as having CHD if they had myocardial infarction, angina pectoris, silent myocardial ischemia, or ischemic heart disease. IS patients were verified according to symptoms and examination results, including computed tomography (CT), magnetic resonance imaging, cerebral angiography, and transcranial doppler ultrasound. CHD in patients was ascertained by electrocardiography, coronary angiography, echocardiography, cardiac CT, and cardiac enzymes.
Patients who fulfilled the inclusion criteria were enrolled consecutively. Trained staff collected the data through physical examination, questionnaires, biochemical measurement, medical records, and relevant tests.
Physical measurements comprised height, weight, waist circumference, hip circumference, and blood pressure. Body mass index (BMI) was calculated as weight (kg)/height (m2) and was ranked into 4 categories: <18=1 (thin), 18 to 24=2 (normal), 24 to 28=3 (overweight), >28=4 (obese). Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured using a standard mercury sphygmomanometer on the right arm of seated participants after a 5-minute rest. Essential hypertension was defined as a rise in blood pressure of unknown cause that increased the risk of cerebral, cardiac, and renal events8 and the absence of secondary causes, such as renovascular disease, renal failure, pheochromocytoma, aldosteronism, or other causes of secondary hypertension or mendelian forms (monogenic).9 Patients with essential hypertension were diagnosed according to SBP ≥140 mm Hg or DBP ≥90 mm Hg or self-reported use of antihypertensive medication.10
Abnormal total cholesterol (TC) was defined as TC >5.18 mmol/L, abnormal triglyceride (TG) as >1.70 mmol/L, abnormal low-density lipoprotein cholesterol (LDL-C) as >3.37 mmol/L, and abnormal high-density lipoprotein cholesterol as <1.04 mmol/L. Dyslipidemia was defined as any of the followings being abnormal: TC, TG, LDL-C, or high-density lipoprotein cholesterol according to Chinese Guidelines on Prevention and Treatment of Dyslipidemia in Adults.11 Diabetes mellitus was defined according to the current American Diabetes Association guidelines (fasting plasma glucose ≥7.0 mmol/L and 2-h plasma glucose ≥11.1 mmol/L).12
Questionnaire responses provided data on the following patient characteristics. (1) Demographic characteristics: name, sex, date of birth, ethnicity, occupation, education, marital status, community name, ID number, and so on. (2) Health-related behaviors: smoking, passive smoking, alcohol consumption, and physical activity. Levels of both leisure and occupational physical activity were assessed as described in a previous study.13 Categories of alcohol intake were determined according to definitions of the National Institute on Alcohol Abuse and Alcoholism.14 Current smokers were defined as those who smoke >1 cigarette per day; nonsmokers were defined as those who had quit smoking within the previous 12 months, had never smoked, or smoked <1 cigarette per day.15 (3) Diet: we adopted the international commonly used food frequency questionnaire to document the diet of the patients for the past year. This questionnaire mainly inquired about the intake frequency of grains, vegetables, fruits, and other foods rich in folic acid and vitamins. (4) Emotional state: Zung self-rating depression scale was applied16; if the score was >50, we defined this as depression according to China’s standards. (5) Disease status and medication use: duration of hypertension and family history of IS and CHD; use of antihypertensive medication, including diuretics, receptor blockers, calcium antagonists, angiotensin-converting enzyme inhibitors, angiotensin II receptor antagonists, α-blockers, compound preparations, and traditional Chinese medicine.
Blood samples were collected by venipuncture from each participant after overnight fasting. The collection tubes were immediately placed in an ice container and transferred to the Clinical Laboratory of the Chronic Diseases Hospital in the Nanshan District within 1 hour. tHcy levels were measured within 4 h. Additional plasma samples were immediately frozen and stored at −80°C for future analysis.
Blood tests included fasting glucose, TC, LDL-C, TG, uric acid (UA), plasma tHcy, and serum creatinine. TC, LDL-C, glucose, and TG were measured using enzymatic methods, UA was detected by quantitative determination with uricase, creatinine was detected by the Jaffe method, and a circulating enzymatic method was adopted for the measurement of Hcy. All of these indicators were tested with an automatic biochemical analyzer (Hitachi 7080).
Folic Acid Intervention
Intervention patients had tHcy levels ≥15 μmol/L during the baseline survey and agreed to participate in the intervention program. Patients took a daily dose of 0.8 mg folic acid, 2 tablets each time (0.4 mg per tablet; Beijing SL Pharmaceutical, H10970079), or took 1 enalapril maleate/folic acid tablet daily (10 mg enalapril maleate, 0.8 mg folic acid; Shenzhen Osa Pharmaceutical, H20103723). In the follow-up intervention, CHSC clinicians evaluated the intake of folic acid and enalapril maleate/folic acid tablets by documenting the medication time, dosage, adverse reactions, compliance, and withdrawal time and reasons. Patients who took folic acid or enalapril maleate/folic acid tablets for >1 month were defined as the exposure group; patients with adverse effects were not included in the exposure group.
Qualitative data are expressed as proportion (%) and were analyzed using the chi-squared test. Quantitative data are presented as mean±standard deviation; variables with normal distributions were analyzed by the Student’s t test, whereas variables with skewed distributions were analyzed using the Mann–Whitney U test. We used the Markov Chain Monte Carlo method of multiple imputation to ensure that the imputed data set had a monotone missing pattern through 20 imputations. The multiple imputation method is commonly used and is flexible for variables with missing values.17 The tHcy levels were divided into 3 categories (normal, <15 μmol/L; mildly elevated, 15–30 μmol/L; moderately elevated, 30–100 μmol/L),18 tHcy levels ≥15 μmol/L were defined as hyperhomocysteinemia.18
The hazard ratios (HRs) and 95% confidence intervals (CIs) of IS and CHD were calculated according to the tHcy levels using the Cox regression model. The model comprised the following: M0, not adjusted; M1, adjusted for 13 factors (including age, sex, education, smoking, alcohol consumption, BMI, physical activity, diabetes mellitus, depression, family history of stroke, years of hypertension, antihypertensive medication, and use of folic acid); M2, adjusted for 18 factors (including the above 13 factors, SBP, TC, glucose, TG, and LDL-C). Based on model M2, we calculated the area under the curve value under the receiver-operating characteristic (ROC) curve to evaluate the effect of the model. Power analysis was simulated with Power and Sample Size Calculation software.19
As shown in Figure I in the online-only Data Supplement, among the total of 5935 participants as of April 2011, 202 presented with IS and 270 with CHD (25 had both IS and CHD) and were thus excluded. Of the remaining 5488 participants with an average follow-up of 2.7 years, 327 were lost to follow-up (5.96%): 138 persons refused to continue, 81 returned to their home town, 65 provided the wrong telephone number or wrong address, 29 moved, and 14 died (not of IS or CHD). Ultimately, 5161 participants were successfully followed and included in this prospective study.
Table 1 presents the baseline characteristics of the participants according to sex. Of the 5488 participants, 2712 were males (49.42%) and 2776 were females (50.58%). There were no differences between males and females concerning physical activity, depression, antihypertensive medication, and baseline SBP. Education, smoking, alcohol consumption, salt intake, oil intake, vegetable intake, fruit intake, years of hypertension, age, BMI, DBP, and levels of TC, LDL-C, UA, glucose, and creatinine showed statistically significant differences between males and females (P<0.05).
Comparison of Baseline Characteristics Between Participants Lost and Not Lost to Follow-Up
To assess the information bias of the participants lost to follow-up, we compared the baseline characteristics between those lost and not lost. Table I in the online-only Data Supplement demonstrates that only fruit intake, physical activity, and SBP differed significantly between the 2 groups (P<0.05).
Comparison of Indicators Between Baseline and Follow-Up
Table II in the online-only Data Supplement shows that age, BMI, waist circumference, SBP, and levels of TC, LDL-C, TG, glucose, creatinine, and tHcy were significantly higher after 2.7 years of follow-up than at baseline, although DBP and UA levels were significantly lower.
After following-up the 5488 participants for 2.7 years, we identified 197 with IS and 7 who died of IS. We also observed 157 with CHD and 5 who died of CHD. Another 14 died for reasons other than IS or CHD. The main type of IS was cerebral infarction (mainly hemorrhagic IS, 75%), intracerebral hemorrhage (17%), subarachnoid hemorrhage (5%), nontraumatic intracranial hemorrhage (1%), and other (2%). The proportions of CHD types were angina pectoris (76%), myocardial infarction (17%), silent myocardial ischemia (4%), ischemic heart disease (2%), and sudden death (1%).
Association of tHcy With IS and CHD
We analyzed the tHcy levels based on sex and age (Table 2). The mean tHcy levels were 13.60 μmol/L in all participants, 15.96 μmol/L in males, and 11.70 μmol/L in females. Males had higher tHcy levels and a higher prevalence of hyperhomocysteinemia than females in each age group (P<0.0001). Of note, tHcy levels and prevalence of hyperhomocysteinemia increased with age.
The participants were further divided into 2 groups based on tHcy levels (<15 and ≥15 μmol/L). Table 3 shows that after 2.7 years of follow-up, the incidence of IS was 3.82% in hypertensive patients, 6.18% in the exposure group (tHcy ≥15 μmol/L), and 2.84% in the control group (tHcy <15 μmol/L). The HRs (95% CIs) of IS caused by hyperhomocysteinemia were 2.18 (1.65–2.89), 2.40 (1.56–3.67), and 2.73 (1.83–4.08) for all participants, males, and females, respectively.
After 2.7 years of follow-up, the incidence of CHD was 3.04% in hypertensive patients, 3.16% in the exposure group (tHcy ≥15 μmol/L), and 2.92% in the control group (tHcy <15 μmol/L). The HRs (95% CIs) of CHD caused by hyperhomocysteinemia (tHcy ≥15 μmol/L) were 1.46 (1.06–2.02) in all participants, 1.19 (0.76–1.88) in males, and 2.26 (1.40–3.64) in females (Table 3). According to power calculations, our sample size provided >80% power to detect the relative associations of tHcy with IS and CHD risk.
Different tHcy Levels and the Prediction of IS and CHD
Table 4 demonstrates that under the M2 model (adjusted by 18 factors), the HRs (95% CIs) for IS in the highest tHcy category (≥30 μmol/L) compared with normal levels of tHcy (<15 μmol/L) were 4.96 (3.03–8.12) in all participants, 6.11 (3.44–10.85) in males, and 1.84 (0.52–6.46) in females. Compared with the reference group (tHcy <15 μmol/L), the HRs (95% CIs) of hyperhomocysteinemia (tHcy ≥15 μmol/L) were 1.82 (1.17–2.84) in all participants, 2.19 (1.40–3.43) in males, and 1.60 (1.09–3.19) in females.
Furthermore, under the M2 model, there was no significant difference in the risk of CHD between participants with the highest tHcy category (tHcy ≥30 μmol/L) and those with normal levels of tHcy (<15 μmol/L). In addition, compared with the group with tHcy <15 μmol/L, there was no significant association between hyperhomocysteinemia and the risk of CHD.
Based on the M2 model, the area under the curve of the ROC curve using the logistic regression model (prediction of different levels of tHcy leading to IS) was 0.71 to 0.74. After eliminating tHcy from the model, the logistic regression (area under the curve of ROC) was 0.70 to 0.73. Using the same model but with CHD instead of IS, the area under the curve of the ROC did not differ from that of IS to any great extent (data not shown).
Effect of Folic Acid Supplementation on tHcy Levels
Among the 5488 hypertensive patients in the follow-up study, 501 took folic acid for an average duration of 2.5 years. Table 5 reveals that compared with the nonintervention group, age, UA, and creatinine were higher in the intervention group, whereas there was no significant difference between the 2 groups regarding BMI, waist circumference, SBP, DBP, TC, glucose, TG, and LDL-C.
Figure demonstrates that after taking folic acid for 2.5 years, compared with baseline data, the plasma tHcy level in the intervention group declined 6.7 μmol/L on average. Folic acid supplementation exerted the most significant effect in the first 3 months, after which tHcy levels declined at a slower pace. In the nonintervention group, tHcy level did not vary extensively.
Vascular risk is stronger in individuals with hypertension combined with hyperhomocysteinemia.20 However, few studies have focused on the effects of tHcy on IS and CHD in hypertensive patients. In this population-based, prospective cohort study, we hypothesized that hyperhomocysteinemia might increase the risk of IS and CHD in Chinese hypertensive patients. Our findings indicate that hyperhomocysteinemia is associated with IS but not CHD in hypertensive patients and that folic acid supplementation reduces tHcy levels. To our knowledge, this prospective cohort study is the first to show a potential association of tHcy level with IS susceptibility in hypertensive patients in China.
Both our baseline survey and cohort study indicated that elevated tHcy was associated with IS, and the effect of higher tHcy levels on IS risk tended to be stronger in females. Consistent with our findings, Hung et al showed that resistant hypertension was associated with a higher risk of IS, especially in women and elderly patients.21 Moreover, Yan et al reported that women with mild hypertension had a higher risk of stroke than men and showed that a 10 mm Hg increase in SBP was associated with a 38% increased risk of stroke in women.22 The Framingham Heart Study also suggested an overall higher lifetime stroke susceptibility in women.23 Estrogen and estrogen–progestogen therapy can reduce homocysteine levels in menopausal women.24 Another study suggested that higher levels of estrogen–homocysteine conjugates could lead to a lowering of homocysteine-induced cardiovascular disease (CVD) risk.25 Here, we observed that elderly women tended to have higher tHcy levels, perhaps because of the lower estrogen levels in older women.
At present, the beneficial effect of folic acid supplementation remains controversial. It is reported that supplementation with vitamins B6 and B12 plus folic acid could reduce blood pressure in patients with essential hypertension.26 Indeed, a substantial body of evidence supports a role of plasma homocysteine in the pathogenesis of hypertension.27 However, an updated Cochrane review demonstrated that clinical trials using low and high doses of folic acid, vitamin B12, and vitamin B6 had not shown any clinical benefit28 and claimed that the effects of folate might be influenced by the duration of treatment.29 The average duration of folic acid supplementation in our study was 2.5 years, which reduced the tHcy level by an average of 6.7 μmol/L (27.92%). Our results therefore suggest that folic acid supplementation can be beneficial for hypertension by reducing the level of tHcy.
Before adjusting for the traditional risk factors of CVD, we noted a significant association between tHcy and the risk of CHD in females, but this association did not remain after adjusting for the risk factors. A meta-analysis revealed that lifelong moderate homocysteine elevation had little or no effect on CHD.30 However, in contrast to our study, another study declared that serum tHcy was an independent predictor of CHD in the elderly population.31
The major constituent of CVD in western populations is CHD; however, it is stroke in the Chinese population.32 A randomized trial demonstrated that multivitamin supplementation decreased the mortality of stroke,32 but not CHD in Chinese individuals.33 Genetic variants of the methylenetetrahydrofolate reductase gene can modulate tHcy levels; the TT genotype of the methylenetetrahydrofolate reductase C677T gene is related to greater risk of IS6, but not CHD,34 and the T allele is more frequent in Asians than Europeans.6 The above may partially explain the significant association between tHcy and IS, but not CHD in our study.
Elevated tHcy level is a clotting factor, which is capable of increasing the risk of stroke >4-fold in patients with atrial fibrillation.35 Stroke and myocardial infarction have different pathogenesis. Most IS are embolic and because of cerebral venous thrombosis with infarction from stasis. Elevated tHcy levels increase thrombosis and the risk of cardioembolic strokes, whereas myocardial infarctions are mainly caused by plaque rupture and occlusion of a coronary artery, with the thrombosis being secondary to the occlusion.36 There is a strong positive association between stroke from atrial fibrillation and age, and tHcy levels above 14 μmol/L are present in 40% of patients >80 years.37
Previous studies have indicated that vitamin B therapy might reduce the risk of stroke but not coronary events38 and that the nonsignificant effect of vitamin B therapy might be related to metabolic deficiency of vitamin B12 and impaired renal function.39 A randomized controlled trial demonstrated that renal failure may alter the response to vitamin B therapy.40 Vitamin B therapy is helpful in patients with good renal function, but harmful in patients with markedly impaired renal function.39 Metabolic B12 deficiency is present in 30% of patients with vascular disease who are >70 years of age, thus higher doses of B12 are necessary for these patients.41 However, high doses of cyanocobalamin may increase the levels of thiocyanate in patients with renal failure.42 Elevation of tHcy and asymmetrical dimethylarginine increases vascular risk in patients with renal failure.43 Methylcobalamin reduces the levels of tHcy and asymmetrical dimethylarginine in dialysis patients,44 whereas cyanocobalamin does not lower levels of asymmetrical dimethylarginine. Therefore, Spence has suggested that methylcobalamin instead of cyanocobalamin should be used in patients with renal impairment.39
A previous study tested the sensitivity, specificity, and accuracy of carotid-color Doppler in detecting dolichocarotids (DCs) and explored the 5-year incidence of cerebro-CVD (transient ischemic attack, IS, myocardial infarction, and cardiovascular death) in patients with DCs. The findings suggested that patients with DCs were susceptible to potentially disabling and fatal events and that presence of DCs was related to cerebro-vascular symptoms.45 In addition, a positive association existed between DCs and nonischemic dilated cardiomyopathy, and women were more likely to develop DCs.46 Future studies should pay attention to the role of DCs in the development of stroke.
In addition, elevated heart rate is a physiopathological state that can affect CVD risk.47 Heart rate reduction is already regarded as a therapeutic target in several cardiac conditions, such as ischemic heart disease and heart failure. Ivabradine reduces the heart rate and is a reliable pharmacological drug for the reduction of morbidity and mortality related to coronary artery diseases and chronic heart failure.48 Hyperpolarization and cyclic nucleotide channels may also be important targets for the reduction of heart rate.49
The advantages of this study were as follows. (1) We studied a large community-based sample of hypertensive patients, and participants were homogeneous regarding their genetic background and environmental exposure. (2) Data were collected under rigid quality control, and important covariables were measured and controlled in the analysis. (3) The prospective cohort study enabled us to obtain a less biased association between exposure variables and outcome events. (4) The findings of our study have potential value in diverse circumstances. At present, most of the literature concerning the relationship between tHcy and susceptibility to IS and CHD has involved healthy individuals rather than hypertensive patients. Unlike Europe and the United States, China does not emphasize or promote the importance of taking folic acid routinely; thus, people in China potentially have lower levels of folic acid and relatively high tHcy levels. It is therefore necessary to gather data from the Chinese population to provide a comprehensive evaluation of the potential influence of tHcy and folic acid on the risk of developing IS and CHD. (5) We have controlled for many traditional risk factors related to IS and CHD and have applied hierarchical analysis to explore the sensitivity and ensure the reliability of the results.
The limitations of this study also need to be noted. (1) The representativeness of the population may be a limitation because only the patients who were registered in the health management system of CHSCs were included in our cohort study; patients in the early stages of hypertension were not recruited. (2) Selection bias may be an issue because we did not include the healthy population. (3) Although we considered many traditional risk factors, there may be other confounding factors not included in this study. Circulating homocysteine is influenced by many factors, such as nutritional deficiencies, lifestyle factors, physiological conditions, and drugs. IS and CHD are complex diseases that are influenced by both genetic and environmental factors and their interactions. Future functional research is needed to elaborate the mechanisms of the interaction between tHcy and folic acid in hypertensive patients. (4) Many types of IS and CHD exist; owing to the limited number in each subgroup, we did not perform subgroup analysis regarding the different forms of IS and CHD. (5) Another limitation is the possible recall bias during data collection. However, we have tried to minimize recall bias through rigorous training in survey methods and application of standard operating procedures during data collection.
This population-based, prospective cohort study suggests that it is necessary to replenish folic acid in Chinese patients with hypertension. Strategies for the prevention and treatment of hypertension, including folic acid supplementation, are necessary to prevent stroke.22
We are grateful for the help of the doctors and nurses in the abovementioned health centers in data and sample collection.
Sources of Funding
The research was supported by grants from Shenzhen Nanshan Bureau of Science and Technology (2010058), K.C.Wong Magna Fund in Ningbo University, the National Natural Science Foundation of China (81402745), the Natural Science Foundation of Ningbo City (2014A610268), the Key Program of Education Commission of Zhejiang Province (Z201017918), the Natural Science Foundation of Zhejiang Province (LQ13H260002), Zhejiang Province Scientific Research Projects of Education (Y201326971), the Ministry of Education, Humanities and Social Sciences project (14YJC630046), China Medical Board Health 2020 Project (08–929), and China Medical Board Distinguished Professorships Project (G16916400/F510000).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.115.009111/-/DC1.
- Received February 12, 2015.
- Revision received May 4, 2015.
- Accepted May 5, 2015.
- © 2015 American Heart Association, Inc.
- Jacques PF,
- Kalmbach R,
- Bagley PJ,
- Russo GT,
- Rogers G,
- Wilson PW,
- et al
- Holmes MV,
- Newcombe P,
- Hubacek JA,
- Sofat R,
- Ricketts SL,
- Cooper J,
- et al
- Carretero OA,
- Oparil S.
- Mancia G,
- De Backer G,
- Dominiczak A,
- Cifkova R,
- Fagard R,
- Germano G,
- et al
- 12.↵American Diabetes Association.Executive summary: standards of medical care in diabetes–2009. Diabetes Care. 2009;32(suppl 1):S6–S12.
- Kriska AM,
- Saremi A,
- Hanson RL,
- Bennett PH,
- Kobes S,
- Williams DE,
- et al
- Halanych JH,
- Safford MM,
- Kertesz SG,
- Pletcher MJ,
- Kim YI,
- Person SD,
- et al
- Barnard J,
- Meng XL.
- Clarke R,
- Bennett DA,
- Parish S,
- Verhoef P,
- Dötsch-Klerk M,
- Lathrop M,
- et al
- Gopinath B,
- Flood VM,
- Rochtchina E,
- Thiagalingam A,
- Mitchell P.
- Mark SD,
- Wang W,
- Fraumeni JF Jr,
- Li JY,
- Taylor PR,
- Wang GQ,
- et al
- Lewis SJ,
- Ebrahim S,
- Davey Smith G.
- Poli D,
- Antonucci E,
- Cecchi E,
- Marcucci R,
- Liotta AA,
- Cellai AP,
- et al
- Spence D.
- Spence JD.
- Koyama K,
- Yoshida A,
- Takeda A,
- Morozumi K,
- Fujinami T,
- Tanaka N.
- Alsagaff MY,
- Thaha M,
- Aminuddin M,
- Yogiarto RM,
- Yogiantoro M,
- Tomino Y.
- Koyama K,
- Ito A,
- Yamamoto J,
- Nishio T,
- Kajikuri J,
- Dohi Y,
- et al