Subtypes of Hypertension and Risk of Stroke in Rural Chinese Adults

Transcription

1 Original Article Subtypes of Hypertension and Risk of Stroke in Rural Chinese Adults Zhaoqing Sun, 1 Xiaodan Han, 1 Liqiang Zheng, 1 Xingang Zhang, 1 Jue Li, 2 Dayi Hu, 2 and Yingxian Sun 1 background Hypertension is a definitive risk factor for stroke. We examined the associations between hypertension subtypes and stroke incidence in rural Chinese adults. methods We conducted a prospective study in a sample of 38,949 rural Chinese adults aged 35 years and free from stroke at baseline. The subtypes of hypertension were defined as isolated systolic hypertension (ISH), isolated diastolic hypertension (IDH), systolic and diastolic hypertension (SDH), and managed hypertension (MHT). The relative risks of stroke with the subtypes of hypertension, compared with normotensive subjects, were estimated using the Cox model after adjusting for age, sex, and other confounders. results The prevalence of hypertension was as follows: SDH = 18.6%, ISH = 10.2%, IDH = 5.8%, and MHT = 0.4%. During a total of 168,023 person-years of follow-up, 858 subjects developed stroke. The incidence rate per 100,000 person-years of overall first-ever stroke was The SDH group was at the highest risk of stroke among all the hypertensive subjects. The hazard ratio was 2.13 (95% confidence interval (CI) = ) for all stroke, 2.74 (95% CI = ) for hemorrhagic stroke, and 1.92 (95% CI = ) for ischemic stroke. conclusions SDH patients warrant the most attention for prevent of stroke. However, ISH and IDH are also independent predictors of stroke. The risk factors for stroke, especially hypertension, were not controlled in the rural population we studied. More aggressive efforts to control the risk factors for stroke in rural areas are needed. Keywords: blood pressure; China; epidemiology; hypertension; rural population; stroke. doi: /ajh/hpt197 Stroke is one of the most common causes of death and a leading cause of long-term disability worldwide. 1 The incidence and mortality of stroke in China are also high, and the burden of stroke is growing in China. Stroke is currently the leading cause of death in China, 2,3 and hypertension is the most important modifiable risk factor for stroke. 4,5 Recent data suggest a rapid increase in the number of cases of hypertension in China, especially in rural areas. 6 8 More than half of the Chinese population lives in rural regions, so any changes in stroke incidence and mortality in rural populations have enormous public health and economic consequences. How to take a useful and scientific public health measure for hypertension and stroke should be in consideration. As we all know, hypertension includes 3 categories isolated diastolic hypertension (IDH), isolated systolic hypertension (ISH), and diastolic with systolic hypertension (systolic diastolic hypertension (SDH)). 9 Several studies have reported a clear relationship between ISH or SDH and stoke However, little data reveal the effects of IDH on the risk of stroke, even though patients with IDH constitute a large portion (14% 24%) of the total hypertensive population. 10,12 Further, IDH patients are less likely to receive clinical attention and treatment than those with ISH or SDH The purpose of this study was to describe the incidence of stroke and subtypes of hypertension and to examine the relationship between hypertension subtypes and the risk of stroke in rural Chinese adults. METHODS Study design and study population The procedures followed were in accordance with ethical standards of the responsible committee on human experimentation of China Medical University. A large-scale, crosssectional survey was conducted from 2004 to 2006 in the rural areas of Fuxin County, Liaoning Province, where there are 35 towns and approximately 640,000 rural people. The study adopted a multistage, stratified clustering sampling scheme that included samples from the northern, southern, western, eastern, and central regions. According to the population, 3 towns were selected from the southern region, 2 Correspondence: Yingxian Sun (sunyingxian@medmail.com.cn). Initially submitted June 14, 2013; date of first revision July 8, 2013; accepted for publication September 29, 2013; online publication October 25, Division of Cardiology, Shengjing Hospital of China Medical University, Shenyang, P.R. China; 2 Heart, Lung and Blood Vessel Center, Tongji University, Shanghai, P.R. China. American Journal of Hypertension, Ltd All rights reserved. For Permissions, please journals.permissions@oup.com American Journal of Hypertension 27(2) February

2 Sun et al. from the eastern region, and 1 town from each of the other 3 regions. Rural villages in the proximity of these towns were selected from different geographic areas. In total, 8 towns and 84 rural villages were selected from these various regions of the country. Exclusion criteria included a history of tumors, congestive heart failure, or pregnancy. We intended that 53,400 subjects aged 35 years should be surveyed; actually, a total of 45,925 people were surveyed between 2004 and 2006, and the response rate was 85%. 6,8 Of the 45,925 subjects at baseline, 3,579 were not included in the follow-up study because study participants contact information was unavailable. Overall, 42,346 participants at baseline were eligible to participate in the follow-up study, 39,845 individuals completed the follow-up, and 38,949 individuals free from stroke at baseline were available for the calculation of disease incidence. Baseline measurement All baseline surveys were conducted by local doctors, who were trained by our research staff. Training was conducted before beginning the study and included the purpose of the research, how to administer the questionnaire, the method of measurement, and the study procedures. A strict test was conducted after the training, and only those who scored perfectly on the test became investigators. Data on demographic variables (age, sex, and ethnicity), smoking status, use of alcohol, information on antihypertensive medications, and history of coronary heart disease (CHD) at baseline were obtained by interviews with a standard epidemiological questionnaire. Alcohol drinking was defined as alcohol consumption of at least 8 grams per week during the last year. 14 Current smoking was defined as smoking at least 1 cigarette every day for at least 1 year. Body weight and height were measured with subjects wearing light clothing and no shoes. Body mass index (BMI) was calculated as the weight in kilograms divided by the height in square meters. Diabetes and lipid disorder were defined according to selfreport. History of CHD at baseline was positive if a CHD was reported during the baseline interview and confirmed by medical records. A detailed description of the methods has been presented elsewhere. 15 A trained and certified observer used an American Heart Association protocol to perform 3 blood pressure (BP) measurements for each participant. The participant rested in a seated position for 5 minutes before the measurements. Participants were advised to avoid alcohol consumption, cigarette smoking, coffee and tea consumption, and exercise for at least 30 min before the measurements. The research staff used a standardized electronic sphygmomanometer (HEM-741C; Omron, Tokyo, Japan) and 1of 4 cuff sizes (pediatric, regular adult, large, or thigh), which was chosen based on each participants s arm circumference. The average of the 3 readings was used for analysis. Subtypes of hypertension were defined according to the baseline systolic BP (SBP), diastolic BP (DBP), and antihypertensive treatment: ISH (SBP 140 mm Hg and DBP < 90 mm Hg); SDH (SBP 140 mm Hg and DBP 90 mm Hg); IDH (SBP < 140 mm Hg and DBP 90 mm Hg) for the participants with or without antihypertensive treatment; managed hypertension (MHT) was defined as BP under control by antihypertensive treatment (SBP < 140 mm Hg and DBP < 90 mm Hg); and normotensive was defined as no history of hypertension and SBP < 140 mm Hg and DBP < 90 mm Hg at baseline. Follow-up data collection The follow-up examination was conducted between July 2010 and October We conducted new home visits with interviews to capture all stroke events. Deaths were identified by hospital records and direct contact with the participant s families using the International Classification of Diseases, Ninth Revision, Clinical Modification. We confirmed that stroke was the cause of death on the basis of autopsy reports, death certificates, medical record abstractions, or information obtained from family members. Stroke was defined according to the World Health Organization MONICA criteria: a sudden onset of focal (or global) disturbance of cerebral function lasting >24 hours (unless interrupted by surgery or death) with no apparent nonvascular cause. 16 The definition included patients presenting with clinical signs and symptoms suggestive of complete stroke, including ischemic stroke, intracerebral hemorrhage stroke, and subarachnoid hemorrhage stroke. A stroke episode that occurred >28 days after the first stroke was considered a recurrent stroke. The classification of ischemic stroke, intracerebral hemorrhage stroke, and subarachnoid hemorrhage stroke was based on clinical presentation, and confirmation by computed tomography or magnetic resonance imaging was required, with reference to the MONICA criteria. Undetermined stroke included all cases that either had not had a computed tomography scan or could not be classified by computed tomography or magnetic resonance imaging scan findings. Although the stroke events included first-ever and recurrent stroke, only firstever strokes were used for analysis. At the conclusion of the study, the endpoint assessment committee, which was blinded to the study participants baseline risk factor information, independently reviewed all materials. Statistical analysis The incidence rate was denoted by case load/100,000 person years. Relative risks (hazard ratio (HR)) and the corresponding 95% confidence intervals (CIs) were derived from Cox proportional hazards regression models for different hypertension subtypes compared with the normotensive group, adjusting for antihypertensive treatment, age, sex, BMI, smoking, drinking, diabetes, lipid disorder, CHD, and/ or SBP. The proportionality assumption was evaluated by scaled Schoenfeld residuals, and the global fit of the models was evaluated by graphically examining the cumulative hazards function relative to the Cox Snell residuals. All analyses were performed with SPSS statistical software version 12.0 (SPSS, Chicago, IL). A two-sided P < 0.05 was accepted as indicating statistical significance. 194 American Journal of Hypertension 27(2) February 2014

3 Hypertension Subtypes and Stroke in rural Chinese RESULTS Baseline date Baseline date is presented in Table 1. The most common subtype of hypertension was SDH (18.6%), followed by ISH (10.2%), IDH (5.8%), and MHT (0.4%). In our population, 20.5% of the participants were taking antihypertensive medication, but only 1.1% of the hypertension patients were under control. All subtypes had more women and older patients compared with the normotensive group, with the exception of the IDH group. Compared with the normotensive group, all hypertension subtypes had a higher mean BMI and a higher prevalence of CHD. All hypertension subtypes also included more patients with lipid disorder and diabetes, with the exception of the IDH group. The differences in drinking and smoking habits between hypertensive and normotensive groups were significant in the IDH and SDH groups. Subtypes of hypertension and the risks of stroke Hypertension subtypes and stroke risks are presented in Table 2. The mean follow-up time was 4.3 years. During 168,023 person-years of follow-up, the total incidence rate of stroke was 5.1 (n = 858), with more cases of ischemic stroke (n = 458) and less cases of hemorrhagic stroke(n = 348); the rest was undetermined stroke (n = 27). We explored the effect of baseline SBP with stroke incidence with univariable Cox regression analysis. We found that there is a direct relationship of SBP and stroke (all stroke: HR = 1.15, 95% CI = ; ischemic stroke: HR = 1.15, 95% CI = ; hemorrhagic stroke: HR = 1.16, 95% CI = ). After adjusting for baseline age, sex, BMI, smoking, drinking, lipid disorder, diabetes, CHD, antihypertensive treatment, the SDH subtype had the highest risk of developing all kinds of stoke (HR = 2.13, 95% CI = ). The same result can be seen in ischemic stroke (HR = 1.92, 95% CI = ) and hemorrhagic stroke (HR = 2.74, 95% CI = ).The IDH group had a slightly higher HR than the ISH group for all kinds of stroke (HR = 1.58, 95% CI = vs. HR = 1.42, 95% CI = ) and for ischemic stroke (HR = 1.97, 95% CI = vs. HR = 1.60, 95% CI = ). When we compared the multivariable models, we added baseline SBP as a covariable to the same models, the associations among hypertension subtypes and stroke are largely attenuated, and significance is no longer observed. Only the SDH group had an increased relative risk of hemorrhagic stroke compared with the normotensive group (HR = 1.63, 95% CI = ). The associations of subtypes of hypertension with the risks of stroke in normal-weight (BMI < 25 kg/m 2 ) and overweight or obese (BMI 25 kg/m 2 ) participants are shown in Table 3. For those with BMI <25 kg/m 2, after adjusting for the confounders mentioned above, the patients with SDH had 2 times the risk of developing any kind of stroke (HR = 2.21, 95% CI = ), 3 times the risk of developing hemorrhagic stroke (HR = 2.97, 95 % CI = ), and 2 times the risk of ischemic stroke (HR = 1.89, 95% CI = ) compared with the normotensive group. The ISH and IDH group showed similar risk on stroke incidence. For those with BMI 25 kg/m 2, only the SDH group ws at higher risk of any kind of stroke incidence compared with the normotensive group. Table 1. Baseline characteristics of study participants Characteristic Normotensive group ISH group IDH group SDH group MHT group Total No. (%) 25,323 (65.0) 3,989 (10.2) 2,256 (5.8) 7,235 (18.6) 146 (0.4) 38,949 Antihypertensive treatment, no. (%) 0 (0) 436 (10.9)* 67 (3.0)* 2,148 (29.7)* 146 (100)* 2,797 (7.2) Age, y, mean ± SD 48.2 ± ± 13.0* 48.8 ± ± 11.4* 54.2 ± 10.3* ± Male, no. (%) 12,776 (50.5) 1,844 (46.2)* 1,216 (53.9) 3,452 (47.7)* 49 (33.6)* 19,337 (49.6) SBP, mm Hg, mean ± SD 121 ± ± 14* 129 ± 7* 164 ± 20* 127 ± 9* ± DBP, mm Hg, mean ± SD 76 ± 8 81 ± 7* 93 ± 5* 100 ± 9* 79 ± 7* ± BMI, kg/m 2, mean ± SD 23.0 ± ± 3.4* 23.6 ± 3.1* 24.1 ± 3.5* 23.9 ± 3.6* ± 3.10 DM, no. (%) 72 (0.3) 26 (0.7)* 9 (0.4) 26 (0.7)* 5 (3.4)* 176 (0.5) Lipid disorder, no. (%) 326 (1.3) 157 (3.9)* 47 (2.1) 608 (8.4)* 19 (13.0)* 1,157 (3.0) Smoking, no. (%) 10,301 (40.7) 1,635 (41.0) 1,004 (44.5)* 3,107 (42.9)* 46 (31.5) 16,093 (41.3) Drinking, no. (%) 7,725 (30.5) 1,146 (28.7) 813 (36.0)* 2,380 (32.9)* 27 (18.5) 12,091 (31.0) CHD, no. (%) 495 (2.0) 223 (5.6)* 70 (3.1)* 584 (8.1)* 36 (24.7)* 1,408 (3.6) Abbreviations: BMI, body mass index; CHD, coronary heart disease; DBP, diastolic blood pressure; DM, diabetes mellitus; IDH, isolated diastolic hypertension; ISH, isolated systolic hypertension; MHT, managed hypertension; SBP, systolic blood pressure; SDH, systolic and diastolic hypertension. *P < 0.05, for t test or χ 2 test, compared with normotensive group. American Journal of Hypertension 27(2) February

4 Sun et al. Table 2. Incidence and relative risk of stroke by subtypes of hypertension Risk of stroke Normotensive group ISH group IDH group SDH group MHT group Total Persons-years of follow up 108,619 17,521 9,587 31, ,023 All stroke Number of cases Incidence per 100,000 person-years , Model 1 Referent 0.353, 1.42 ( ) 0.455, 1.58 ( ) 0.755, 2.13 ( ) 0.174, 1.19 ( ) Model 2 Referent 0.035, 0.97 ( ) 0.374, 1.45 ( ) 0.228, 1.26 ( ) 0.232, 1.26 ( ) Hemorrhagic stroke Number of cases Incidence per 100,000 person years Model 1 Reference 0.228, 1.26 ( ) 0.038,1.04 ( ) 1.007,2.74 ( ) 0.271,1.31 ( ) Model 2 Reference 0.151, 0.86 ( ) 0.042,0.96 ( ) 0.486,1.63 ( ) 0.335,1.40 ( ) Ischemic stroke Number of cases incidence per 100,000 person- years Model 1 Reference 0.467, 1.60 ( ) 0.678, 1.97 ( ) 0.651, 1.92 ( ) 0.219, 0.80 ( ) Model 2 Reference 0.042, 1.04 ( ) 0.591, 1.81 ( ) 0.077, 1.08 ( ) 0.165, 0.85 ( ) Model 1 was adjusted for baseline age, sex, body mass index, smoking, drinking, lipid disorder, diabetes, coronary heart disease, and antihypertensive treatment. Model 2 was adjusted for baseline age, sex, body mass index, smoking, drinking, lipid disorder, diabetes, coronary heart disease, antihypertensive treatment, and systolic blood pressure. Abbreviations: IDH, isolated diastolic hypertension; ISH, isolated systolic hypertension; MHT, managed hypertension; SDH, systolic and diastolic hypertension. 196 American Journal of Hypertension 27(2) February 2014

5 Hypertension Subtypes and Stroke in rural Chinese Table 3. Relative risk of stroke by hypertension subtypes and body mass index Normotensive group ISH group IDH group SDH group MHT group BMI < 25 kg/m 2 All stroke Referent 0.428, 1.53 ( ) 0.408, 1.50 ( ) 0.792, 2.21 ( ) 0.354, 1.43 ( ) Ischemic stroke Referent 0.558, 1.75 ( ) 0.591, 1.81 ( ) 0.637, 1.89 ( ) 0.557, 0.57 ( ) Hemorrhagic stroke Referent 0.234, 1.26 ( ) 0.026, 1.03 ( ) 1.087, 2.97 ( ) 0.755, 2.13 ( ) BM 25 kg/m 2 All stroke Referent 0.088, 0.92 ( ) 0.178, 1.20 ( ) 0.461, 1.59 ( ) 0.635, 0.53 ( ) Ischemic stroke Referent 0.222, 0.80 ( ) 0.413, 1.51 ( ) 0.478, 1.61 ( ) 0.005, 0.99 ( ) Hemorrhagic stroke Referent 0.178, 1.20 ( ) 0.158, 0.85 ( ) 0.650, 1.92 ( ) Adjusted for baseline age, sex, smoking, drinking, lipid disorder, diabetes, coronary heart disease, and antihypertensive treatment. Abbreviations: BMI, body mass index; IDH, isolated diastolic hypertension; ISH, isolated systolic hypertension; MHT, managed hypertension; SDH, systolic and diastolic hypertension. Discussion This study is the first epidemiologic study on stroke to be conducted in the rural region of Liaoning Province since The results showed basic information about the epidemiologic date of hypertension and stroke in rural parts of China. Our study reported that the prevalence of hypertension in the follow-up group was 35.0%. Among hypertensive patients, only 20.5% were taking antihypertensive medication, and 1.1% of patients achieved BP control. SDH was the most common subtype, followed by ISH and IDH. The prevalence differed by age group and the IDH had a significant lower mean age. Our finding was similar to that of recent studies. 12,17 In our study, the incidence rate per 100,000 person years of overall first-ever stroke was Compared with recent stroke studies conducted in Chinese urban centers and Western countries, the incidence of stroke in our study is high. The medical conditions in the rural Chinese region in our study were relatively poor, and risks factors for stroke, especially hypertension, lipid disorder, and diabetes, were uncontrolled. At the same time, people in rural areas acquired Westernized lifestyles and dietary habits in recent years. All the above-mentioned reasons may explain the change. Several recent studies focused on the effect of hypertension subtypes on the risk of stroke. A study of 26,587 Chinese subjects reported that the HRs of stroke were 2.16 for IDH, 2.35 for ISH, and 2.96 for SDH compared with normotensive subjects. 12 The China Hypertension Epidemiology Follow-Up Study reported that the HRs of stroke were 1.85 for IDH, 2.20 for ISH, and 3.74 for SDH among 169,871 participants from China. 13 A Western stroke study reported HRs of stroke were 1.14 for IDH, 1.36 for ISH, and 2.71 for SDH among 3,267 Finnish men. 21 Our results support the findings of other recent studies that showed a significantly stronger association of SDH with all kinds of stroke compared with other subtypes of hypertension. Although most previous studies reported that individuals with ISH had the highest risk of stroke, 10,22,23 considering the population group and decade character, individuals with SDH should receive the most attention for stroke prevention and protection from stroke-related injury. Different from previous studies, when we compared the multivariable models, we added baseline SBP as a covariable to the same model. We found that the associations among hypertension subtypes and stroke are largely attenuated and significance is no longer observed. That the categories do not add much to the BP values themselves may be a useful finding. We will further study this. We observed significant associations between the IDH and ISH groups and the development of ischemic stroke, which is consistent with recent Chinese studies. 12,13 However, in our study, risk for hemorrhagic stroke was not significant in the ISH and IDH groups. This observation might be explained by several factors. First, our study was based on first-term follow-up (4 5 years of follow-up). This short evaluation period may show a lower risk in the ISH and IDH group. 12,24 Second, subjects in the IDH group had a lower mean age, American Journal of Hypertension 27(2) February

6 Sun et al. indicating a shorter duration of hypertension. Finally, we identified only 10 cases of hemorrhagic stroke in the IDH group and 51 cases in the ISH group; the bias introduced by this small sample also may explain the inconsistent results. Our study has several limitations. The main limitation of our study is that the associated clinical conditions were selfreported. This may have underestimated the prevalence of these conditions. This also limited the possibility to evaluate the impact of the associated conditions on the incidence of stroke in a reliable manner. We did not collect atrial fibrillation data for the participants, and this may be reduce the study s power. Also, the size of the MHT sample was small, so we did not observe the expected result. Large sample sizes are needed in the future to confirm the results In conclusion, all of the hypertension subtypes were associated with an increased incidence of stroke. SDH should receive the most attention for stroke prevention and protection from stroke-related injury. Many interrelated effects and risk factors require further study. We observed an unsatisfactory control rate of hypertension in the rural population, and, therefore, a widespread program including diagnosis, monitoring, and treatment of hypertension is critically needed. We believe that a public health strategy that includes educational and environmental interventions could be effective. The medical conditions and healthcare strategies in the rural Chinese population warrant increased attention. Acknowledgments We thank our subjects for their enthusiastic participation. This research was supported by grants from the key technology Research and Development program of Liaoning Province ( ). DISCLOSURE The authors declared no conflict of interest. References 1. Bonita R, Mendis S, Truelsen T, Bogousslavsky J, Toole J, Yatsu F. The global stroke initiative. Lancet Neurol 2004; 3: Thorvaldsen P, Asplund K, Kuulasmaa K, Rajakangas AM, Schroll M. Stroke incidence, case fatality, and mortality in the WHO MONICA Project. Stroke 1995; 26: Huang J, Yu J, Hu D, Wu Y, Lu J, Li Y, Huang Y, Azen SP, Dustin LD, Detrano RC. Major causes of death among men and women in China. N Engl J Med 2005; 353: Kearney P, Whelton M, Reynolds K, Muntner P, He J. Global burden of hypertension: analysis of worldwide data. Lancet 2005; 365: Lopez A, Mathers C, Ezzati M, Jamison D, Murray C. Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet 2003; 367: Sun Z, Zheng L, Wei Y, Li J, Zhang X, Zhang X, Liu S, Xu C, Li J, Zhao F, Dong G, Hu D, Sun Y. The prevalence of prehypertension and hypertension among rural adults in Liaoning province of China. Clin Cardiol 2007; 30: Gu D, Reynolds K, Wu X, Chen J, Duan X, Muntner P, Huang G, Reynolds RF, Su S, Whelton PK, He J. Prevalence, awareness, treatment, and control of hypertension in China. Hypertension 2002; 40: Sun Z, Zheng L, Detrano R, Zhang D, Zhang X, Xu C, Li J, Liu S, Li J, Hu D, Sun Y. The accelerating epidemic of hypertension among rural Chinese women: results from Liaoning Province. Am J Hypertens 2008; 21: Kaplan NM, Victor RG. Kaplan s Clinical Hypertension. Lippincott Williams and Wilkins: Philadelphia, PA, Hozawa A, Ohkubo T, Nagai K, Kikuya M, Matsubara M, Tsuji I, Ito S, Satoh H, Hisamichi S, Imai Y. Prognosis of isolated systolic and isolated diastolic hypertension as assessed by self-measurement of blood pressure at home: the Ohasama Study. Arch Intern Med 2000; 160: Strandberg TE, Salomaa VV, Vanhanen HT, Pitkala K, Miettinen TA. Isolated diastolic hypertension, pulse pressure, and mean arterial pressure as predictors of mortality during a follow-up of up to 32 years. J Hypertens 2002; 20: Fang XH, Zhang XH, Yang QD, Dai XY, Su FZ, Rao ML, Wu SP, Du XL, Wang WZ, Li SC. Subtype hypertension and risk of stroke in middle-aged and older Chinese: a 10-year follow-up study. Stroke 2006; 37: Kelly TN, Gu D, Chen J, Huang JF, Chen JC, Duan X, Wu X, Yau CL, Whelton PK, He J. Hypertens subtype and risk of cardiovascular disease in Chinese adults. Circulation 2008; 118: National Institute on Alcohol Abuse and Alcoholism: Health risks and benefits of alcohol consumption. Alcohol Res Health. 2000; 1: Zhang X, Sun Z, Zhang X, Zheng L, Liu S, Xu C, Li J, Zhao F, Li J, Hu D, Sun Y. Prevalence and associated factors of overweight and obesity in a Chinese rural population. Obesity (Silver Spring) 2008; 16: Tunstall-Pedoe H, for the WHO MONICA Project Principle Investigators. The World Health Organization MONICA Project (Monitoring of Trends and Determinants in Cardiovascular Disease): a major international collaboration. J Clin Epidemiol 1988; 41: Huang JF, Wildman RP, Gu DF, Muntner P, Su SY, He J. Prevalence of isolated systolic and isolated diastolic hypertension subtypes in China. Am J Hypertens 2004; 17: Jiang B, Wang WZ, Chen HL, Hong Z, Yang QD, Wu SP, Du XL, Bao QJ. Incidence and trends of stroke and its subtypes in China: results from three large cities. Stroke 2006; 37: Ministry of Health. Chinese Health Statistical Digest Ministry of Health, People s Republic of China: Beijing, Corso G, Bottacchi E, Giardini G, De la Pierre F, Meloni T, Pesenti Campagnoni M, Ponzetti C, Veronese Morosini M. Communitybased study of stroke incidence in the Valley of Aosta, Italy. Carecerebrovascular Aosta Registry: years Neuroepidemiology 2009; 32: Strandberg TE, Salomaa VV, Vanhanen HT, Pitkala K, Miettinen TA. Isolated diastolic hypertension, pulse pressure, and mean arterial pressure as predictors of mortality during a follow-up of up to 32 years. J Hypertens 2002; 20: Nielsen W, Lindenstrom E, Vestbo J, Jenson G. Is diastolic hypertension an independent risk factor for stroke in the presence of normal systolic blood pressure in the middle-aged and elderly? Am J Hypertens 1997; 10: Petrovich H, Curb D, Bloom-Marcus M. Isolated systolic hypertension and risk of stroke in Japanese-American men. Stroke 1995; 26: Fang XH, Longstreth WT Jr, Li SC, Kronmal RA, Cheng XM, Wang WZ, Wu SP, Du XL, Dai XY. Longitudinal study of blood pressure and stroke in over 37,000 people in China. Cerebrovasc Dis 2001; 11: American Journal of Hypertension 27(2) February 2014