Journal of Human Hypertension (2014), 1–5 & 2014 Macmillan Publishers Limited All rights reserved 0950-9240/14 www.nature.com/jhh
ORIGINAL ARTICLE
Impact of metabolic syndrome components on incident stroke subtypes: a Chinese cohort study Y-C Chen1, C-A Sun2, T Yang3, C-H Chu4, C-H Bai5, S-L You6,7, L-C Hwang8, C-H Chen9, C-Y Wei10 and Y-C Chou1,7 Limited evidence is available on the risk differences in the development of stroke subtypes in relation to particular clustering patterns of the metabolic syndrome (MetS) components. A follow-up study of a Chinese cohort involving 10 292 individuals was performed to assess the roles of cluster patterns of the MetS components in the prediction of incident stroke subtypes. During follow-up, there were 161 incident cases of ischemic strokes and 41 incident cases of hemorrhagic strokes. Among MetS components, only the hypertensive trait was associated with significantly elevated risks of both ischemic and hemorrhagic strokes. Furthermore, MetS with hypertension as components was associated with increased risk of ischemic and hemorrhagic strokes (adjusted hazards ratio (95% confidence interval) was 2.96 (1.94–4.50) and 2.93 (1.25–6.90), respectively) as compared with those who had neither hypertension nor MetS. Notably, as the number of the MetS components increased, the risk of ischemic stroke significantly and dose-dependently increased. This implies a cumulative effect of MetS components in elevating the risk of ischemic stroke. These findings suggest that MetS comprises heterogenous clusters with respect to the risk of developing the subtype of stroke. Journal of Human Hypertension advance online publication, 16 January 2014; doi:10.1038/jhh.2013.152 Keywords: cohort study; hemorrhagic stroke; ischemic stroke; metabolic syndrome
INTRODUCTION Stroke accounts for about 10% of all deaths in industrialised countries and is responsible for a vast burden of disability in the community.1 The main feature of stroke epidemiology is large ethnic variations in incidence and mortality from stroke. The incidence and mortality from stroke are higher among Asian populations than those among Western populations.1,2 In Taiwan, stroke has been the second leading cause of death since 1983.3 In addition, the percentage of the subtype of stroke was 71% for cerebral infarction and 22% for cerebral hemorrhage in Taiwan.4 It has been noted that the incidence rate of stroke in Taiwan is higher than that reported from the United States and cerebral hemorrhages are more common among people in Taiwan than among occidental people.4 Accordingly, there is a need for studies on etiologies of stroke in Taiwan in order to identify effective methods of prevention. The metabolic syndrome (MetS) is a highly prevalent constellation of vascular risk factors, including elevated blood pressure, elevated blood glucose, obesity and dyslipidemia.5 Several prospective studies have shown that the MetS was associated with increased morbidity and mortality not only for patients with cardiovascular disease6 but also for patients with stroke events.7 However, previous investigations only included ischemic stroke as a defined outcome measure.8–12 Therefore, little is known about the risk differences for MetS in the occurrence of stroke subtypes. Moreover, it is unclear whether a particular MetS clusters predict
the development of stroke subtypes. Accordingly, we performed this study to examine the incidence of stroke (including ischemic stroke and hemorrhagic stroke) in a cohort of Chinese subjects with and without MetS. We also assessed the roles of several clusters of the MetS components in the prediction of incident stroke subtypes.
MATERIALS AND METHODS Study population The study population consisted of subjects who participated in the Taiwanese Survey on Prevalences of Hypertension, Hyperglycemia, and Hyperlipidemia (TwSHHH) and a subsequent follow-up study (TwSHHH-II). The detailed study design and data collection have been reported elsewhere.13,14 The initial TwSHHH was conducted in 2002 based on a multistage random sample of civilian, non-institutionalized population in Taiwan. There were 10 292 individuals randomly selected for the TwSHHH. Of them, 7578 (73.6%) completed a questionnaire and 6600 (64.1%) permitted additional blood pressure and other biomarker measurements. These 6600 subjects who completed all examinations in the TwSHHH were enrolled in the TwSHHH-II initiated in 2007. In the current study, subjects without data on blood pressure, anthropometric and other biomarker measurements (including fasting plasma glucose, high-density lipoprotein cholesterol (HDL-C) and triglycerides) (n ¼ 65) and those who had a previous history of stroke based on questionnaire or medical claim data (n ¼ 204) were excluded. Accordingly, a total of 6331 subjects were included in the study. Differences in age distributions were not statistically
1 Graduate Institute of Life Sciences, National Defense Medical Center, Taipei City, Taiwan, Republic of China; 2Department of Public Health, College of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan, Republic of China; 3Department of Health Business Administration, Meiho University, Pingtung, Taiwan, Republic of China; 4Department of Surgery, Tri-Service General Hospital, National Defense Medical Center, Taipei City, Taiwan, Republic of China; 5School of Public Health, College of Public Health and Nutrition, Taipei Medical University, Taipei City, Taiwan, Republic of China; 6Genomics Research Center, Academia Sinica, Taipei City, Taiwan, Republic of China; 7School of Public Health, National Defense Medical Center, Taipei City, Taiwan, Republic of China; 8Department of Medicine, Mackay Medical College, New Taipei City, Taiwan, Republic of China; 9Digestive Disease Center, Changhua Show-Chwan Memorial Hospital, Changhua County, Taiwan, Republic of China and 10Department of Neurology, Chang Bing Show-Chwan Memorial Hospital, Changhua County, Taiwan, Republic of China. Correspondence: Dr Y-C Chou, School of Public Health, National Defense Medical Center, No. 161, Section 6, Min-Chuan East Road, Neihu District, Taipei City, Taiwan 114, Republic of China. E-mail: [email protected] Received 5 August 2013; revised 9 December 2013; accepted 11 December 2013
Metabolic syndrome and incident stroke Y-C Chen et al
2 significant between those participating in the TwSHHH and nonparticipants. The protocols for the TwSHHH and TwSHHH-II were approved by the Institutional Review Board at the Bureau of Health Promotion, Department of Health, Executive Yuan in Taiwan. Written informed consent was obtained from all participants in the TwSHHH and the TwSHHH-II. At study entry, participants underwent questionnaire interviews and anthropometric measurements, including body weight, height and waist circumference, by well-trained nurses under a standardized protocol. Arterial blood pressure was obtained using a random-zero mercury column sphygmomanometer. The systolic blood pressure and diastolic blood pressure were measured two times in the left arm of seated participants. A third blood pressure measurement was made if the first two blood pressure readings differed by 410 mm Hg. The average of the two closest readings was calculated to determine the reported blood pressure for each participant. Body mass index was calculated as weight divided by height squared (kg m 2). A blood sample was collected into an EDTA anticoagulant tube for each participant after a 12-h overnight fast. Standard enzymatic methods were used to determine serum cholesterol and triglycerides. Electrophoresis was performed to measure HDL-C and lowdensity lipoprotein cholesterol. Plasma glucose was measured by the hexokinase glucose-6-phosphate dehydrogenase procedure.
Clinical Modification (ICD-9-CM) was used to classify stroke events. Ischemic strokes were determined using the ICD-9-CM codes 433 (occlusion and stenosis of the precerebral arteries, n ¼ 14), 434 (occlusion of the cerebral arteries, n ¼ 134) and 436 (acute but ill-defined cerebrovascular disease, n ¼ 13). Hemorrhagic strokes were determined using the ICD-9-CM codes 430 (subarachnoid hemorrhage, n ¼ 11), 431 (intracerebral hemorrhage, n ¼ 27) and 432 (other and unspecified intracranial hemorrhage, n ¼ 3). Unclassified strokes were also recorded (n ¼ 117), but were not included in the subtype analysis. When participants experienced stroke more than one occasion, only the first stroke event was taken into consideration. Subjects who experienced incident strokes or died from any cause were censored as of the date of the event. The remaining subjects were censored in December 2006 if their medical claim data indicated that they were free from strokes until that time.
Statistical analysis
Ascertainment of incident stroke
Person years were calculated as the sum of individual follow-up time until the occurrence of incident strokes, death or the end of 2006. The hazard ratios (HRs) and the respective 95% confidence intervals (CIs) of incident stroke subtypes were determined using the Cox proportional hazards model with reference to the risk in subjects without the MetS, without each of the MetS components, or with none of the components when appropriate. We adjusted for age (years), sex, smoking status (never, former and current smokers), alcohol intake category (never, former and current drinkers) and family history of cardiovascular disease. It has been noted that hypertension is a significant and independent risk factor for stroke.17 In the evaluation of the potential importance of the hypertensive component per se in conjunction with the actual definition of MetS in the risk of incident stroke subtypes, several mutually exclusive subgroups were defined. Group 1, the reference group, comprised individuals who had neither the hypertensive component nor the full definition of MetS (two or less MetS components). Group 2 comprised subjects who met criteria of MetS but lacked the hypertensive component in the criteria. Group 3 comprised participants who had hypertensive component but had not fulfilled the definition of MetS, and group 4 comprised individuals who had fulfilled the definition of MetS with the hypertensive component in the definition. Data management and statistical analyses were performed with SAS version 9.2 (SAS Institute, Cary, NC, USA). All of the statistical tests were two-sided with an alpha level of 0.05.
Stroke events were ascertained using computerized record linkages of the national identification numbers of the survey participants to the National Death Certificate System and the National Health Insurance Database (NHID) in Taiwan. Because reporting death events is mandatory in Taiwan, the completeness of the registry is nearly 100%.15 In addition, the NHID provides complete data regarding outpatient visits, hospital admissions, prescriptions, diseases and vital statistics for 99% of the Taiwanese population.16 Of the participants, 99.5% were covered by the national health insurance; thus, their medical histories were obtained from the claims data. The International Classification of Diseases, 9th Revision,
RESULTS The final analytic sample comprised 6331 subjects with a mean (±s.d.) age of 42.2 (±16.5) years, and 47.6% were men. Overall, 1212 individuals had MetS, yielding a prevalence rate of 19.1% (men: 21.3%, women: 17.2%). Table 1 shows the baseline characteristics for MetS and non-MetS subjects. The MetS group was significantly older (mean age 53.3 years) than the non-MetS
Definition of MetS In this study, MetS was defined according to the criteria set by a joint statement of the International Diabetes Federation Task Force on Epidemiology, and Prevention; the International Atherosclerosis Society; and the International Association for the Study of Obesity.5 Subjects who fulfilled three of the following five criteria were defined as having MetS: (1) hypertension: a blood pressure of at least 130/85 mm Hg or use of anti-hypertensive medications, (2) hypertriglyceridemia: serum triglyceride levels of at least 150 mg dl 1 or use of medications for elevated triglycerides, (3) reduced HDL-C: HDL-C levels o40 mg dl 1 in men and o50 mg dl 1 in women or use of drug treatment for reduced HDL-C, (4) hyperglycemia: a fasting plasma glucose level of 100 mg dl 1 or more or use of drug treatment for elevated glucose and (5) central obesity: waist circumference X90 cm in men and X80 cm in women.
Table 1.
Baseline characteristics of the study participants according to the status of MetS
Age (years) Men (%) BMI (kg m 2) Waist circumference (cm) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Serum total cholesterol (mg dl 1) Serum HDL-C (mg dl 1) Serum LDL cholesterol (mg dl 1) Serum triglycerides (mg dl 1) Fasting serum glucose (mg dl 1) Cigarette smokers (%) Alcohol drinkers (%) Family history of CVD (%)
Without MetS (n ¼ 5119)
With MetS (n ¼ 1212)
P-value
39.6±15.7 642 (53.0) 22.7±3.4 76.8±10.0 111.2±15.5 72.6±10.2 178.6±34.7 57.5±13.7 111.6±25.5 104.3±56.5 88.9±17.9 1159 (22.6) 518 (10.1) 479 (9.4)
53.3±15.1 570 (47.0) 26.9±3.7 91.1±9.0 130.6±18.0 83.5±10.9 199.5±41.3 47.0±16.3 129.6±28.9 215.2±115.1 113.3±44.9 395 (32.6) 203 (16.8) 193 (15.9)
o0.0001 o0.001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001
Abbreviations: BMI, body mass index; CVD, cardiovascular disease; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; MetS, metabolic syndrome.
Journal of Human Hypertension (2014) 1 – 5
& 2014 Macmillan Publishers Limited
Metabolic syndrome and incident stroke Y-C Chen et al
3 group was (mean age 39.6 years) (Po0.001). As expected, MetS subjects exhibited significantly higher BMI, waist circumference, arterial blood pressure, lipid profile, triglyceride and fasting plasma glucose levels than did non-MetS subjects (Po0.001). The prevalence of cigarette smoking and alcohol consumption was significantly higher in the MetS group compared with the non-MetS group (Po0.001). After 28 749.95 follow-up person-years, we documented 161 incident cases of ischemic stroke and 41 incident cases of hemorrhagic stroke. The incidence rates of ischemic stroke and hemorrhagic stroke were 5.53 per 1000 person-years and 1.40 per 1000 person-years, respectively. We determined the distribution of the clustering patterns of MetS components among the participants with MetS at baseline. Table 2 indicates that the prevalence of the hypertension combined with additional MetS components was as follows: 38.6% of subjects exhibited an additional two components, demonstrating a primary clustering pattern defined by hypertension, central obesity and hypertriglyceridemia (15.3%); 27.2% of subjects exhibited three additional components, demonstrating a primary clustering pattern of hypertension, central obesity, hypertriglyceridemia and hyperglycemia as companions (11.0%);
Table 2. Distribution of the clustering patterns of MetS components among the participants with MetS at baseline (n ¼ 1212) Clustering pattern
No. of subjects
Prevalence (%)
55 9 101 82 186 34 33 69 20 131 34 134 33 130 53 108
4.5 0.9 8.3 6.8 15.3 2.8 2.7 5.7 1.7 10.8 2.8 11.0 2.7 10.7 4.4 8.9
HT þ HG þ TG HT þ HG þ HDL HT þ HG þ CO HT þ TG þ HDL HT þ TG þ CO HT þ HDL þ CO HG þ TG þ HDL HG þ TG þ CO HG þ HDL þ CO TG þ HDL þ CO HT þ HG þ TG þ HDL HT þ HG þ TG þ CO HT þ HG þ HDL þ CO HT þ TG þ HDL þ CO HG þ TG þ HDL þ CO HT þ HG þ TG þ HDL þ CO
Abbreviations: CO, central obesity; HDL, reduced high-density lipoprotein cholesterol; HG, hyperglycemia; HT, hypertension; TG, hypertriglyceridemia; MetS, metabolic syndrome.
Table 3.
and 8.9% of subjects had a complete set of components of MetS. Table 3 shows that subjects with MetS had a significantly higher risk of ischemic stroke than did non-MetS subjects after adjusting for potential confounding covariates (adjusted HR ¼ 1.81, 95% CI ¼ 1.31–2.48). However, no significant association was determined between MetS and the risk of hemorrhagic stroke (adjusted HR ¼ 1.19, 95% CI ¼ 0.61–2.33). When we considered the stroke risk in relation to individual MetS components, only the hypertensive component was associated with significantly elevated risks of both ischemic and hemorrhagic strokes after mutual adjusting for individual MetS components; the adjusted HR (95% CI) values for both ischemic and hemorrhagic strokes were 2.22 (1.52–3.23) and 4.19 (1.92–9.15), respectively. In addition, as the number of the MetS components increased, the HR of ischemic stroke significantly and dose-dependently increased. The multivariableadjusted HRs for ischemic stroke for subjects with 1, 2, 3 and more than 4 MetS components were 2.18 (95% CI ¼ 1.09–4.36), 3.14 (95% CI ¼ 1.60–6.16), 3.10 (95% CI ¼ 1.54–6.21) and 5.40 (95% CI ¼ 2.73–10.68), respectively, when compared with those without any of the MetS components (P for trend was o0.001). However, no significant dose–response gradient was determined for incidence of hemorrhagic stroke and the number of MetS components (Figure 1). Table 4 shows the associations of risks of incident stroke subtypes with clustering patterns of MetS components. In Cox regression model adjusting for age, sex, cigarette smoking and alcohol intake habits, and family history of cardiovascular disease (with having neither MetS nor hypertension as the reference group), the HRs (95% CIs) for ischemic stroke associated with MetS without hypertension, hypertension without MetS and MetS with hypertension, respectively, were 1.50 (0.67–3.37), 2.28 (1.45–3.38) and 2.96 (1.94–4.50). The HRs (95% CIs) for hemorrhagic stroke associated with hypertension without MetS and MetS with hypertension were 3.74 (1.63–8.57) and 2.93 (1.25–6.90), respectively. Of notice was no any case of hemorrhagic stroke in subjects with MetS but without hypertension.
DISCUSSION In this population-based follow-up study, we determined that the presence of hypertension per se or MetS defined by the presence of hypertension significantly predicted the incidences of ischemic and hemorrhagic strokes. In addition, as the number of MetS components increased, the risk of ischemic stroke increased, exhibiting a significant and dose–response trend. These findings are crucial because the relationships between clustering patterns
Risk of stroke subtypes in relation to the full definition of MetS and its components (n ¼ 6331) No. of subjects (%)
Ischemic stroke No. of event
a
Metabolic syndrome Central obesityc Hypertensionc Reduced HDL-Cc Hyperglycemiac Hypertriglyceridemiac
1212 1887 1775 1457 1063 1967
(19.1) (29.8) (28.0) (23.0) (16.8) (31.1)
77 87 116 50 61 86
Hemorrhagic stroke
HR (95% CI) b
1.81 1.12d 2.22d 1.28d 1.20d 1.36d
(1.31–2.48) (0.80–1.57) (1.52–3.23) (0.90–1.82) (0.86–1.67) (0.98–1.89)
No. of event 14 16 30 11 12 15
HR (95% CI) 1.19b 0.73d 4.19d 1.38d 1.09d 0.66d
(0.61–2.33) (0.37–1.44) (1.92–9.15) (0.66–2.86) (0.54–2.22) (0.33–1.31)
Abbreviations: CI, confidence interval; HDL-C, high-density lipoprotein cholesterol; HR, hazard ratio; MetS, metabolic syndrome. aCompared with individuals without metabolic syndrome (less than three components). bAdjusted for age, sex, smoking status, drinking status and family history of cardiovascular disease. c Compared with individuals without the specific component. dAdjusted for age, sex, smoking status, drinking status, family history of cardiovascular disease and other MetS components.
& 2014 Macmillan Publishers Limited
Journal of Human Hypertension (2014) 1 – 5
Metabolic syndrome and incident stroke Y-C Chen et al
4 population. Further, a primary clustering pattern of the MetS components among subjects with MetS was hypertension combined with central obesity and hypertriglyceridemia (45.9% shown in Table 2). Hence, the absence of hypertension largely overlaps with the absence of central obesity. Consequently, the absence of hypertension in the definition of MetS leaves out hyperglycemia and reduced level of HDL-C to form a diagnosis of the syndrome. As a result, hypertension was the hard core of MetS to contribute to an increased risk of ischemic and hemorrhagic strokes. The MetS without hypertension as components was not a significant factor for the risk of stroke. Of notice was a significant positive dose–response relationship between the numbers of MetS components and the risk of ischemic stroke. This implies a cumulative effect of MetS components in elevating the risk of ischemic stroke. These study results can be explained by the different underlying causes of ischemic and hemorrhagic strokes; the former is mainly a consequence of atherosclerosis and the latter is mainly caused by hypertension and cerebral aneurysms. Indeed, a previous report examining six communities in Chongqing, China, revealed that hypertension yielded the highest HR for hemorrhagic stroke.18 In addition, investigators who examined Finnish and Swedish cohorts determined that exhibiting all MetS components did not predict the risk of hemorrhagic stroke, and other than hypertension, no single MetS component predicted the risk of hemorrhagic stroke.19 Likewise, several studies documented that a full definition of MetS did not perform better than its individual components in prediction of coronary artery disease.20,21 Taken together, these findings suggest that MetS comprises heterogenous clusters with respect to stroke and cardiovascular disease risks. The present study has certain strengths and limitations. This study examined data regarding baseline cluster patterns of MetS components as risk predictors for the subtypes of incident strokes. Because the current study used a representative sample of an ethnic Chinese population, the findings can likely be generalized to all Chinese adults. However, the duration of the follow-up period was potentially inadequate, and the small number of stroke events observed (hemorrhagic strokes in particular), potentially limited the comparison of the cluster patterns of MetS components as they relate to stroke events. In addition, epidemiological studies have typically confirmed that stroke event rates are closely related to the severity of carotid atherosclerosis22,23 and atrial fibrillation.24 However, the present study did not collect sufficient information regarding these covariates to include them in the data analysis; they may be residual confounding factors. Finally, no repeated measurements of the MetS components were performed. In conclusion, the MetS comprises heterogeneous clusters with respect to the risk of stroke subtypes. More intensive efforts should be directed toward prevention and management of the individual components of MetS, particularly hypertension, to reduce the burden of the subtype of stroke.
of MetS and the risk of stroke subtypes have not been thoroughly characterized. Our investigation demonstrated that hypertension was the hard core of MetS associated with a significantly elevated risk of ischemic and hemorrhagic strokes. Equally important, we found that a higher number of MetS components was significantly associated with increased risk of ischemic stroke. However, no significant positive dose–response relationship emerged between the number of MetS components and the risk of incident hemorrhagic stroke. Our analysis showed that central obesity and hypertension were main traits to form MetS (82.3 and 74.7%, respectively, as shown in Table 2) in this ethnic Chinese
Figure 1. Adjusted HRs for the number of MetS components to predict incident strokes. The HRs were adjusted for age, sex, smoking status, drinking status and family history of cardiovascular disease. Subjects without any of the MetS components were the reference group. (a) Ischemic strokes: P for trend o0.001 and (b) hemorrhagic strokes: P for trend ¼ 0.207.
Table 4.
Joint effect of the hypertensive component and the full definition of MetS on the risk of incident stroke subtypes
Hypertension
Absence Absence Presence Presence
MetS
No. of subjects
Absence Presence Absence Presence
4250 306 869 906
Ischemic stroke
Hemorrhage stroke
No. of cases
HRa (95% CI)
No. of cases
HRa (95% CI)
38 7 46 70
Reference 1.50 (0.67–3.37) 2.28 (1.45–3.58) 2.96 (1.94–4.50)
11 0 16 14
Reference — 3.74 (1.63–8.57) 2.93 (1.25–6.90)
Abbreviations: CI, confidence interval; HR, hazard ratio; MetS, metabolic syndrome. aHazard ratios were adjusted for age, sex, smoking status, drinking status and family history of cardiovascular disease.
Journal of Human Hypertension (2014) 1 – 5
& 2014 Macmillan Publishers Limited
Metabolic syndrome and incident stroke Y-C Chen et al
5 What is known about topic: Metabolic syndrome was associated with increased morbidity and mortality for patients with stroke events. However, limited evidence is available on the risk differences in the development of the stroke subtypes in relation to particular clustering patterns of the metabolic syndrome components.
6 7 8
What this study adds: The MetS comprises heterogenous clusters with respect to the risk of developing the subtype of stroke. Among MetS components, only the hypertensive trait was associated with significantly elevated risks of both ischemic and hemorrhagic strokes. Furthermore, MetS with hypertension as components was associated with increased risk of ischemic and hemorrhagic strokes as compared with those who had neither hypertension nor MetS. Notably, as the number of the MetS components increased, the risk of ischemic stroke significantly and dose-dependently increased. This implies a cumulative effect of MetS components in elevating the risk of ischemic stroke. More intensive efforts should be directed toward prevention and management of the individual components of MetS, particularly hypertension, to reduce the burden of the subtype of stroke.
9 10
11
12
13
14
CONFLICT OF INTEREST
15
The authors declare no conflict of interest. 16
ACKNOWLEDGEMENTS The authors thank the Bureau of Health Promotion, Department of Health for administrative support and nurses at study areas for data collection. Sources of funding: This study was supported by a grant from the Bureau of Health Promotion, Department of Health (DOH-95-HP-2103), Executive Yuan, Taiwan, Republic of China.
17 18
19
DISCLAIMER The sponsor had no role in study design, collection, analysis and interpretation of data; writing the report and the decision to submit the report for publication.
20
REFERENCES
21
1 Feigin VL, Lawes CMM, Bennett DA, Anderson CS. Stroke epidemiology: a review of population-based study of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2003; 2: 43–53. 2 Khaw KT. Epidemiology of stroke. J Neurol Neurosurg Psychiatry 1996; 61: 333–338. 3 National Health Administration, Executive Yuan, Republic of China. Health Statistics: I. Vital statistics, 1976-2011. 4 Hu HH, Sheng WY, Chu FL, Lan CF, Chiang BN. Incidence of stroke in Taiwan. Stroke 1992; 23: 1237–1241. 5 Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA 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
& 2014 Macmillan Publishers Limited
22
23
24
federation; international atherosclerosis society; and international association for the study of obesity. Circulation 2009; 120: 1640–1645. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005; 365: 1415–1428. Towfighi A, Ovbiagele B. Metabolic syndrome and stroke. Curr Diabetes Rep 2008; 8: 37–41. Kurl S, Laukkanen JA, Niskanen L, Laaksonen D, Sivenius J, Nyyssonen K et al. Metabolic syndrome and the risk of stroke in middle-aged men. Stroke 2006; 37: 806–811. Chen HJ, Bai CH, Yeh WT, Chiu HC, Pan WH. Influence of metabolic syndrome and general obesity on the risk of ischemic stroke. Stroke 2006; 37: 1060–1064. Iso H, Sato S, Kitamura A, Imano H, Kiyama M, Yamagishi K et al. Metabolic syndrome and the risk of ischemic heart disease and stroke among Japanese men and women. Stroke 2007; 38: 1744–1751. Boden-Albala B, Sacco RL, Lee HS, Grahame-Clarke C, Rundek T, Elkind MV et al. Metabolic syndrome and ischemic stroke risk: Northern Manhattan Study. Stroke 2008; 39: 30–35. Rodriguez-Colon SM, Mo J, Duan Y, Liu J, Caulfield JE, Jin X et al. Metabolic syndrome clusters and the risk of incident stroke: The Atherosclerosis Risk in Communities (ARIC) Study. Stroke 2009; 40: 200–205. Yang T, Chu CH, Bai CH, You SL, Chou YC, Chou WY et al. Uric acid as a risk marker for metabolic syndrome: a Chinese cohort study. Atherosclerosis 2012; 220: 525–531. Yang T, Chu CH, Bai CH, You SL, Chou YC, Hwang LC et al. Uric acid concentration as a risk marker for blood pressure progression and incident hypertension: a Chinese cohort study. Metabolism 2012; 61: 1747–1755. Lu TH, Chang HY, Hwu CM, Chiu HC, Yin WH, Pan WH. Comparison of official coders versus physician panel in assignment of underlying cause of death. J Formos Med Assoc 2001; 100: 365–369. Wen CP, Tsai SP, Chung WS. A. 10-year experience with universal health insurance in Taiwan: measuring changes in health and health disparity. Ann Intern Med 2008; 148: 258–267. Woo D, Haverbusch M, Sekar P, Kissela B, Khoury J, Schneider A et al. Effect of untreated hypertension on hemorrhagic stroke. Stroke 2004; 35: 1703–1708. Zhang WW, Liu CY, Wang YJ, Xu ZQ, Chen Y, Zhou HD. Metabolic syndrome increases the risk of stroke: a 5-year follow-up study in a Chinese population. J Neurol 2009; 256: 1493–1499. Qiao Q, Laatikainen T, Zethelius B, Stegmayr B, Eliasson M, Jousilahti P et al. Comparison of definitions of metabolic syndrome in relation to the risk of developing stroke and coronary heart disease in Finnish and Swedish cohorts. Stroke 2009; 40: 337–343. Iribarren C, Go AS, Husson G, Sidney S, Fair JM, Quertermous T et al. Metabolic syndrome and early-onset coronary artery disease: is the whole greater than its parts? J Am Coll Cardiol 2006; 48: 1800–1807. Hanefeld M, Koehler C, Gallo S, Benke I, Ott P. Impact of the individual components of the metabolic syndrome and their different combinations on the prevalence of atherosclerotic vascular disease in type 2 diabetes: the Diabetes in Germany (DIG) study. Cardiovasc Diabetol 2007; 6: 13–19. Handa N, Matsumoto M, Maeda H, Hougaku H, Kamada T. Ischemic stroke events and carotid atherosclerosis: results of the Osaka Follow-up Study for Ultrasonographic Assessment of Carotid Atherosclerosis (the OSACA Study). Stroke 1995; 26: 1781–1786. Magai Y, Kitagawa K, Sakaguchi M, Shimizu Y, Hashimoto H, Yamagami H et al. Significance of earlier carotid atherosclerosis for stroke subtypes. Stroke 2001; 32: 1780–1785. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991; 22: 983–988.
Journal of Human Hypertension (2014) 1 – 5
ORIGINAL ARTICLE
Impact of metabolic syndrome components on incident stroke subtypes: a Chinese cohort study Y-C Chen1, C-A Sun2, T Yang3, C-H Chu4, C-H Bai5, S-L You6,7, L-C Hwang8, C-H Chen9, C-Y Wei10 and Y-C Chou1,7 Limited evidence is available on the risk differences in the development of stroke subtypes in relation to particular clustering patterns of the metabolic syndrome (MetS) components. A follow-up study of a Chinese cohort involving 10 292 individuals was performed to assess the roles of cluster patterns of the MetS components in the prediction of incident stroke subtypes. During follow-up, there were 161 incident cases of ischemic strokes and 41 incident cases of hemorrhagic strokes. Among MetS components, only the hypertensive trait was associated with significantly elevated risks of both ischemic and hemorrhagic strokes. Furthermore, MetS with hypertension as components was associated with increased risk of ischemic and hemorrhagic strokes (adjusted hazards ratio (95% confidence interval) was 2.96 (1.94–4.50) and 2.93 (1.25–6.90), respectively) as compared with those who had neither hypertension nor MetS. Notably, as the number of the MetS components increased, the risk of ischemic stroke significantly and dose-dependently increased. This implies a cumulative effect of MetS components in elevating the risk of ischemic stroke. These findings suggest that MetS comprises heterogenous clusters with respect to the risk of developing the subtype of stroke. Journal of Human Hypertension advance online publication, 16 January 2014; doi:10.1038/jhh.2013.152 Keywords: cohort study; hemorrhagic stroke; ischemic stroke; metabolic syndrome
INTRODUCTION Stroke accounts for about 10% of all deaths in industrialised countries and is responsible for a vast burden of disability in the community.1 The main feature of stroke epidemiology is large ethnic variations in incidence and mortality from stroke. The incidence and mortality from stroke are higher among Asian populations than those among Western populations.1,2 In Taiwan, stroke has been the second leading cause of death since 1983.3 In addition, the percentage of the subtype of stroke was 71% for cerebral infarction and 22% for cerebral hemorrhage in Taiwan.4 It has been noted that the incidence rate of stroke in Taiwan is higher than that reported from the United States and cerebral hemorrhages are more common among people in Taiwan than among occidental people.4 Accordingly, there is a need for studies on etiologies of stroke in Taiwan in order to identify effective methods of prevention. The metabolic syndrome (MetS) is a highly prevalent constellation of vascular risk factors, including elevated blood pressure, elevated blood glucose, obesity and dyslipidemia.5 Several prospective studies have shown that the MetS was associated with increased morbidity and mortality not only for patients with cardiovascular disease6 but also for patients with stroke events.7 However, previous investigations only included ischemic stroke as a defined outcome measure.8–12 Therefore, little is known about the risk differences for MetS in the occurrence of stroke subtypes. Moreover, it is unclear whether a particular MetS clusters predict
the development of stroke subtypes. Accordingly, we performed this study to examine the incidence of stroke (including ischemic stroke and hemorrhagic stroke) in a cohort of Chinese subjects with and without MetS. We also assessed the roles of several clusters of the MetS components in the prediction of incident stroke subtypes.
MATERIALS AND METHODS Study population The study population consisted of subjects who participated in the Taiwanese Survey on Prevalences of Hypertension, Hyperglycemia, and Hyperlipidemia (TwSHHH) and a subsequent follow-up study (TwSHHH-II). The detailed study design and data collection have been reported elsewhere.13,14 The initial TwSHHH was conducted in 2002 based on a multistage random sample of civilian, non-institutionalized population in Taiwan. There were 10 292 individuals randomly selected for the TwSHHH. Of them, 7578 (73.6%) completed a questionnaire and 6600 (64.1%) permitted additional blood pressure and other biomarker measurements. These 6600 subjects who completed all examinations in the TwSHHH were enrolled in the TwSHHH-II initiated in 2007. In the current study, subjects without data on blood pressure, anthropometric and other biomarker measurements (including fasting plasma glucose, high-density lipoprotein cholesterol (HDL-C) and triglycerides) (n ¼ 65) and those who had a previous history of stroke based on questionnaire or medical claim data (n ¼ 204) were excluded. Accordingly, a total of 6331 subjects were included in the study. Differences in age distributions were not statistically
1 Graduate Institute of Life Sciences, National Defense Medical Center, Taipei City, Taiwan, Republic of China; 2Department of Public Health, College of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan, Republic of China; 3Department of Health Business Administration, Meiho University, Pingtung, Taiwan, Republic of China; 4Department of Surgery, Tri-Service General Hospital, National Defense Medical Center, Taipei City, Taiwan, Republic of China; 5School of Public Health, College of Public Health and Nutrition, Taipei Medical University, Taipei City, Taiwan, Republic of China; 6Genomics Research Center, Academia Sinica, Taipei City, Taiwan, Republic of China; 7School of Public Health, National Defense Medical Center, Taipei City, Taiwan, Republic of China; 8Department of Medicine, Mackay Medical College, New Taipei City, Taiwan, Republic of China; 9Digestive Disease Center, Changhua Show-Chwan Memorial Hospital, Changhua County, Taiwan, Republic of China and 10Department of Neurology, Chang Bing Show-Chwan Memorial Hospital, Changhua County, Taiwan, Republic of China. Correspondence: Dr Y-C Chou, School of Public Health, National Defense Medical Center, No. 161, Section 6, Min-Chuan East Road, Neihu District, Taipei City, Taiwan 114, Republic of China. E-mail: [email protected] Received 5 August 2013; revised 9 December 2013; accepted 11 December 2013
Metabolic syndrome and incident stroke Y-C Chen et al
2 significant between those participating in the TwSHHH and nonparticipants. The protocols for the TwSHHH and TwSHHH-II were approved by the Institutional Review Board at the Bureau of Health Promotion, Department of Health, Executive Yuan in Taiwan. Written informed consent was obtained from all participants in the TwSHHH and the TwSHHH-II. At study entry, participants underwent questionnaire interviews and anthropometric measurements, including body weight, height and waist circumference, by well-trained nurses under a standardized protocol. Arterial blood pressure was obtained using a random-zero mercury column sphygmomanometer. The systolic blood pressure and diastolic blood pressure were measured two times in the left arm of seated participants. A third blood pressure measurement was made if the first two blood pressure readings differed by 410 mm Hg. The average of the two closest readings was calculated to determine the reported blood pressure for each participant. Body mass index was calculated as weight divided by height squared (kg m 2). A blood sample was collected into an EDTA anticoagulant tube for each participant after a 12-h overnight fast. Standard enzymatic methods were used to determine serum cholesterol and triglycerides. Electrophoresis was performed to measure HDL-C and lowdensity lipoprotein cholesterol. Plasma glucose was measured by the hexokinase glucose-6-phosphate dehydrogenase procedure.
Clinical Modification (ICD-9-CM) was used to classify stroke events. Ischemic strokes were determined using the ICD-9-CM codes 433 (occlusion and stenosis of the precerebral arteries, n ¼ 14), 434 (occlusion of the cerebral arteries, n ¼ 134) and 436 (acute but ill-defined cerebrovascular disease, n ¼ 13). Hemorrhagic strokes were determined using the ICD-9-CM codes 430 (subarachnoid hemorrhage, n ¼ 11), 431 (intracerebral hemorrhage, n ¼ 27) and 432 (other and unspecified intracranial hemorrhage, n ¼ 3). Unclassified strokes were also recorded (n ¼ 117), but were not included in the subtype analysis. When participants experienced stroke more than one occasion, only the first stroke event was taken into consideration. Subjects who experienced incident strokes or died from any cause were censored as of the date of the event. The remaining subjects were censored in December 2006 if their medical claim data indicated that they were free from strokes until that time.
Statistical analysis
Ascertainment of incident stroke
Person years were calculated as the sum of individual follow-up time until the occurrence of incident strokes, death or the end of 2006. The hazard ratios (HRs) and the respective 95% confidence intervals (CIs) of incident stroke subtypes were determined using the Cox proportional hazards model with reference to the risk in subjects without the MetS, without each of the MetS components, or with none of the components when appropriate. We adjusted for age (years), sex, smoking status (never, former and current smokers), alcohol intake category (never, former and current drinkers) and family history of cardiovascular disease. It has been noted that hypertension is a significant and independent risk factor for stroke.17 In the evaluation of the potential importance of the hypertensive component per se in conjunction with the actual definition of MetS in the risk of incident stroke subtypes, several mutually exclusive subgroups were defined. Group 1, the reference group, comprised individuals who had neither the hypertensive component nor the full definition of MetS (two or less MetS components). Group 2 comprised subjects who met criteria of MetS but lacked the hypertensive component in the criteria. Group 3 comprised participants who had hypertensive component but had not fulfilled the definition of MetS, and group 4 comprised individuals who had fulfilled the definition of MetS with the hypertensive component in the definition. Data management and statistical analyses were performed with SAS version 9.2 (SAS Institute, Cary, NC, USA). All of the statistical tests were two-sided with an alpha level of 0.05.
Stroke events were ascertained using computerized record linkages of the national identification numbers of the survey participants to the National Death Certificate System and the National Health Insurance Database (NHID) in Taiwan. Because reporting death events is mandatory in Taiwan, the completeness of the registry is nearly 100%.15 In addition, the NHID provides complete data regarding outpatient visits, hospital admissions, prescriptions, diseases and vital statistics for 99% of the Taiwanese population.16 Of the participants, 99.5% were covered by the national health insurance; thus, their medical histories were obtained from the claims data. The International Classification of Diseases, 9th Revision,
RESULTS The final analytic sample comprised 6331 subjects with a mean (±s.d.) age of 42.2 (±16.5) years, and 47.6% were men. Overall, 1212 individuals had MetS, yielding a prevalence rate of 19.1% (men: 21.3%, women: 17.2%). Table 1 shows the baseline characteristics for MetS and non-MetS subjects. The MetS group was significantly older (mean age 53.3 years) than the non-MetS
Definition of MetS In this study, MetS was defined according to the criteria set by a joint statement of the International Diabetes Federation Task Force on Epidemiology, and Prevention; the International Atherosclerosis Society; and the International Association for the Study of Obesity.5 Subjects who fulfilled three of the following five criteria were defined as having MetS: (1) hypertension: a blood pressure of at least 130/85 mm Hg or use of anti-hypertensive medications, (2) hypertriglyceridemia: serum triglyceride levels of at least 150 mg dl 1 or use of medications for elevated triglycerides, (3) reduced HDL-C: HDL-C levels o40 mg dl 1 in men and o50 mg dl 1 in women or use of drug treatment for reduced HDL-C, (4) hyperglycemia: a fasting plasma glucose level of 100 mg dl 1 or more or use of drug treatment for elevated glucose and (5) central obesity: waist circumference X90 cm in men and X80 cm in women.
Table 1.
Baseline characteristics of the study participants according to the status of MetS
Age (years) Men (%) BMI (kg m 2) Waist circumference (cm) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Serum total cholesterol (mg dl 1) Serum HDL-C (mg dl 1) Serum LDL cholesterol (mg dl 1) Serum triglycerides (mg dl 1) Fasting serum glucose (mg dl 1) Cigarette smokers (%) Alcohol drinkers (%) Family history of CVD (%)
Without MetS (n ¼ 5119)
With MetS (n ¼ 1212)
P-value
39.6±15.7 642 (53.0) 22.7±3.4 76.8±10.0 111.2±15.5 72.6±10.2 178.6±34.7 57.5±13.7 111.6±25.5 104.3±56.5 88.9±17.9 1159 (22.6) 518 (10.1) 479 (9.4)
53.3±15.1 570 (47.0) 26.9±3.7 91.1±9.0 130.6±18.0 83.5±10.9 199.5±41.3 47.0±16.3 129.6±28.9 215.2±115.1 113.3±44.9 395 (32.6) 203 (16.8) 193 (15.9)
o0.0001 o0.001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001 o0.0001
Abbreviations: BMI, body mass index; CVD, cardiovascular disease; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; MetS, metabolic syndrome.
Journal of Human Hypertension (2014) 1 – 5
& 2014 Macmillan Publishers Limited
Metabolic syndrome and incident stroke Y-C Chen et al
3 group was (mean age 39.6 years) (Po0.001). As expected, MetS subjects exhibited significantly higher BMI, waist circumference, arterial blood pressure, lipid profile, triglyceride and fasting plasma glucose levels than did non-MetS subjects (Po0.001). The prevalence of cigarette smoking and alcohol consumption was significantly higher in the MetS group compared with the non-MetS group (Po0.001). After 28 749.95 follow-up person-years, we documented 161 incident cases of ischemic stroke and 41 incident cases of hemorrhagic stroke. The incidence rates of ischemic stroke and hemorrhagic stroke were 5.53 per 1000 person-years and 1.40 per 1000 person-years, respectively. We determined the distribution of the clustering patterns of MetS components among the participants with MetS at baseline. Table 2 indicates that the prevalence of the hypertension combined with additional MetS components was as follows: 38.6% of subjects exhibited an additional two components, demonstrating a primary clustering pattern defined by hypertension, central obesity and hypertriglyceridemia (15.3%); 27.2% of subjects exhibited three additional components, demonstrating a primary clustering pattern of hypertension, central obesity, hypertriglyceridemia and hyperglycemia as companions (11.0%);
Table 2. Distribution of the clustering patterns of MetS components among the participants with MetS at baseline (n ¼ 1212) Clustering pattern
No. of subjects
Prevalence (%)
55 9 101 82 186 34 33 69 20 131 34 134 33 130 53 108
4.5 0.9 8.3 6.8 15.3 2.8 2.7 5.7 1.7 10.8 2.8 11.0 2.7 10.7 4.4 8.9
HT þ HG þ TG HT þ HG þ HDL HT þ HG þ CO HT þ TG þ HDL HT þ TG þ CO HT þ HDL þ CO HG þ TG þ HDL HG þ TG þ CO HG þ HDL þ CO TG þ HDL þ CO HT þ HG þ TG þ HDL HT þ HG þ TG þ CO HT þ HG þ HDL þ CO HT þ TG þ HDL þ CO HG þ TG þ HDL þ CO HT þ HG þ TG þ HDL þ CO
Abbreviations: CO, central obesity; HDL, reduced high-density lipoprotein cholesterol; HG, hyperglycemia; HT, hypertension; TG, hypertriglyceridemia; MetS, metabolic syndrome.
Table 3.
and 8.9% of subjects had a complete set of components of MetS. Table 3 shows that subjects with MetS had a significantly higher risk of ischemic stroke than did non-MetS subjects after adjusting for potential confounding covariates (adjusted HR ¼ 1.81, 95% CI ¼ 1.31–2.48). However, no significant association was determined between MetS and the risk of hemorrhagic stroke (adjusted HR ¼ 1.19, 95% CI ¼ 0.61–2.33). When we considered the stroke risk in relation to individual MetS components, only the hypertensive component was associated with significantly elevated risks of both ischemic and hemorrhagic strokes after mutual adjusting for individual MetS components; the adjusted HR (95% CI) values for both ischemic and hemorrhagic strokes were 2.22 (1.52–3.23) and 4.19 (1.92–9.15), respectively. In addition, as the number of the MetS components increased, the HR of ischemic stroke significantly and dose-dependently increased. The multivariableadjusted HRs for ischemic stroke for subjects with 1, 2, 3 and more than 4 MetS components were 2.18 (95% CI ¼ 1.09–4.36), 3.14 (95% CI ¼ 1.60–6.16), 3.10 (95% CI ¼ 1.54–6.21) and 5.40 (95% CI ¼ 2.73–10.68), respectively, when compared with those without any of the MetS components (P for trend was o0.001). However, no significant dose–response gradient was determined for incidence of hemorrhagic stroke and the number of MetS components (Figure 1). Table 4 shows the associations of risks of incident stroke subtypes with clustering patterns of MetS components. In Cox regression model adjusting for age, sex, cigarette smoking and alcohol intake habits, and family history of cardiovascular disease (with having neither MetS nor hypertension as the reference group), the HRs (95% CIs) for ischemic stroke associated with MetS without hypertension, hypertension without MetS and MetS with hypertension, respectively, were 1.50 (0.67–3.37), 2.28 (1.45–3.38) and 2.96 (1.94–4.50). The HRs (95% CIs) for hemorrhagic stroke associated with hypertension without MetS and MetS with hypertension were 3.74 (1.63–8.57) and 2.93 (1.25–6.90), respectively. Of notice was no any case of hemorrhagic stroke in subjects with MetS but without hypertension.
DISCUSSION In this population-based follow-up study, we determined that the presence of hypertension per se or MetS defined by the presence of hypertension significantly predicted the incidences of ischemic and hemorrhagic strokes. In addition, as the number of MetS components increased, the risk of ischemic stroke increased, exhibiting a significant and dose–response trend. These findings are crucial because the relationships between clustering patterns
Risk of stroke subtypes in relation to the full definition of MetS and its components (n ¼ 6331) No. of subjects (%)
Ischemic stroke No. of event
a
Metabolic syndrome Central obesityc Hypertensionc Reduced HDL-Cc Hyperglycemiac Hypertriglyceridemiac
1212 1887 1775 1457 1063 1967
(19.1) (29.8) (28.0) (23.0) (16.8) (31.1)
77 87 116 50 61 86
Hemorrhagic stroke
HR (95% CI) b
1.81 1.12d 2.22d 1.28d 1.20d 1.36d
(1.31–2.48) (0.80–1.57) (1.52–3.23) (0.90–1.82) (0.86–1.67) (0.98–1.89)
No. of event 14 16 30 11 12 15
HR (95% CI) 1.19b 0.73d 4.19d 1.38d 1.09d 0.66d
(0.61–2.33) (0.37–1.44) (1.92–9.15) (0.66–2.86) (0.54–2.22) (0.33–1.31)
Abbreviations: CI, confidence interval; HDL-C, high-density lipoprotein cholesterol; HR, hazard ratio; MetS, metabolic syndrome. aCompared with individuals without metabolic syndrome (less than three components). bAdjusted for age, sex, smoking status, drinking status and family history of cardiovascular disease. c Compared with individuals without the specific component. dAdjusted for age, sex, smoking status, drinking status, family history of cardiovascular disease and other MetS components.
& 2014 Macmillan Publishers Limited
Journal of Human Hypertension (2014) 1 – 5
Metabolic syndrome and incident stroke Y-C Chen et al
4 population. Further, a primary clustering pattern of the MetS components among subjects with MetS was hypertension combined with central obesity and hypertriglyceridemia (45.9% shown in Table 2). Hence, the absence of hypertension largely overlaps with the absence of central obesity. Consequently, the absence of hypertension in the definition of MetS leaves out hyperglycemia and reduced level of HDL-C to form a diagnosis of the syndrome. As a result, hypertension was the hard core of MetS to contribute to an increased risk of ischemic and hemorrhagic strokes. The MetS without hypertension as components was not a significant factor for the risk of stroke. Of notice was a significant positive dose–response relationship between the numbers of MetS components and the risk of ischemic stroke. This implies a cumulative effect of MetS components in elevating the risk of ischemic stroke. These study results can be explained by the different underlying causes of ischemic and hemorrhagic strokes; the former is mainly a consequence of atherosclerosis and the latter is mainly caused by hypertension and cerebral aneurysms. Indeed, a previous report examining six communities in Chongqing, China, revealed that hypertension yielded the highest HR for hemorrhagic stroke.18 In addition, investigators who examined Finnish and Swedish cohorts determined that exhibiting all MetS components did not predict the risk of hemorrhagic stroke, and other than hypertension, no single MetS component predicted the risk of hemorrhagic stroke.19 Likewise, several studies documented that a full definition of MetS did not perform better than its individual components in prediction of coronary artery disease.20,21 Taken together, these findings suggest that MetS comprises heterogenous clusters with respect to stroke and cardiovascular disease risks. The present study has certain strengths and limitations. This study examined data regarding baseline cluster patterns of MetS components as risk predictors for the subtypes of incident strokes. Because the current study used a representative sample of an ethnic Chinese population, the findings can likely be generalized to all Chinese adults. However, the duration of the follow-up period was potentially inadequate, and the small number of stroke events observed (hemorrhagic strokes in particular), potentially limited the comparison of the cluster patterns of MetS components as they relate to stroke events. In addition, epidemiological studies have typically confirmed that stroke event rates are closely related to the severity of carotid atherosclerosis22,23 and atrial fibrillation.24 However, the present study did not collect sufficient information regarding these covariates to include them in the data analysis; they may be residual confounding factors. Finally, no repeated measurements of the MetS components were performed. In conclusion, the MetS comprises heterogeneous clusters with respect to the risk of stroke subtypes. More intensive efforts should be directed toward prevention and management of the individual components of MetS, particularly hypertension, to reduce the burden of the subtype of stroke.
of MetS and the risk of stroke subtypes have not been thoroughly characterized. Our investigation demonstrated that hypertension was the hard core of MetS associated with a significantly elevated risk of ischemic and hemorrhagic strokes. Equally important, we found that a higher number of MetS components was significantly associated with increased risk of ischemic stroke. However, no significant positive dose–response relationship emerged between the number of MetS components and the risk of incident hemorrhagic stroke. Our analysis showed that central obesity and hypertension were main traits to form MetS (82.3 and 74.7%, respectively, as shown in Table 2) in this ethnic Chinese
Figure 1. Adjusted HRs for the number of MetS components to predict incident strokes. The HRs were adjusted for age, sex, smoking status, drinking status and family history of cardiovascular disease. Subjects without any of the MetS components were the reference group. (a) Ischemic strokes: P for trend o0.001 and (b) hemorrhagic strokes: P for trend ¼ 0.207.
Table 4.
Joint effect of the hypertensive component and the full definition of MetS on the risk of incident stroke subtypes
Hypertension
Absence Absence Presence Presence
MetS
No. of subjects
Absence Presence Absence Presence
4250 306 869 906
Ischemic stroke
Hemorrhage stroke
No. of cases
HRa (95% CI)
No. of cases
HRa (95% CI)
38 7 46 70
Reference 1.50 (0.67–3.37) 2.28 (1.45–3.58) 2.96 (1.94–4.50)
11 0 16 14
Reference — 3.74 (1.63–8.57) 2.93 (1.25–6.90)
Abbreviations: CI, confidence interval; HR, hazard ratio; MetS, metabolic syndrome. aHazard ratios were adjusted for age, sex, smoking status, drinking status and family history of cardiovascular disease.
Journal of Human Hypertension (2014) 1 – 5
& 2014 Macmillan Publishers Limited
Metabolic syndrome and incident stroke Y-C Chen et al
5 What is known about topic: Metabolic syndrome was associated with increased morbidity and mortality for patients with stroke events. However, limited evidence is available on the risk differences in the development of the stroke subtypes in relation to particular clustering patterns of the metabolic syndrome components.
6 7 8
What this study adds: The MetS comprises heterogenous clusters with respect to the risk of developing the subtype of stroke. Among MetS components, only the hypertensive trait was associated with significantly elevated risks of both ischemic and hemorrhagic strokes. Furthermore, MetS with hypertension as components was associated with increased risk of ischemic and hemorrhagic strokes as compared with those who had neither hypertension nor MetS. Notably, as the number of the MetS components increased, the risk of ischemic stroke significantly and dose-dependently increased. This implies a cumulative effect of MetS components in elevating the risk of ischemic stroke. More intensive efforts should be directed toward prevention and management of the individual components of MetS, particularly hypertension, to reduce the burden of the subtype of stroke.
9 10
11
12
13
14
CONFLICT OF INTEREST
15
The authors declare no conflict of interest. 16
ACKNOWLEDGEMENTS The authors thank the Bureau of Health Promotion, Department of Health for administrative support and nurses at study areas for data collection. Sources of funding: This study was supported by a grant from the Bureau of Health Promotion, Department of Health (DOH-95-HP-2103), Executive Yuan, Taiwan, Republic of China.
17 18
19
DISCLAIMER The sponsor had no role in study design, collection, analysis and interpretation of data; writing the report and the decision to submit the report for publication.
20
REFERENCES
21
1 Feigin VL, Lawes CMM, Bennett DA, Anderson CS. Stroke epidemiology: a review of population-based study of incidence, prevalence, and case-fatality in the late 20th century. Lancet Neurol 2003; 2: 43–53. 2 Khaw KT. Epidemiology of stroke. J Neurol Neurosurg Psychiatry 1996; 61: 333–338. 3 National Health Administration, Executive Yuan, Republic of China. Health Statistics: I. Vital statistics, 1976-2011. 4 Hu HH, Sheng WY, Chu FL, Lan CF, Chiang BN. Incidence of stroke in Taiwan. Stroke 1992; 23: 1237–1241. 5 Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA 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
& 2014 Macmillan Publishers Limited
22
23
24
federation; international atherosclerosis society; and international association for the study of obesity. Circulation 2009; 120: 1640–1645. Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005; 365: 1415–1428. Towfighi A, Ovbiagele B. Metabolic syndrome and stroke. Curr Diabetes Rep 2008; 8: 37–41. Kurl S, Laukkanen JA, Niskanen L, Laaksonen D, Sivenius J, Nyyssonen K et al. Metabolic syndrome and the risk of stroke in middle-aged men. Stroke 2006; 37: 806–811. Chen HJ, Bai CH, Yeh WT, Chiu HC, Pan WH. Influence of metabolic syndrome and general obesity on the risk of ischemic stroke. Stroke 2006; 37: 1060–1064. Iso H, Sato S, Kitamura A, Imano H, Kiyama M, Yamagishi K et al. Metabolic syndrome and the risk of ischemic heart disease and stroke among Japanese men and women. Stroke 2007; 38: 1744–1751. Boden-Albala B, Sacco RL, Lee HS, Grahame-Clarke C, Rundek T, Elkind MV et al. Metabolic syndrome and ischemic stroke risk: Northern Manhattan Study. Stroke 2008; 39: 30–35. Rodriguez-Colon SM, Mo J, Duan Y, Liu J, Caulfield JE, Jin X et al. Metabolic syndrome clusters and the risk of incident stroke: The Atherosclerosis Risk in Communities (ARIC) Study. Stroke 2009; 40: 200–205. Yang T, Chu CH, Bai CH, You SL, Chou YC, Chou WY et al. Uric acid as a risk marker for metabolic syndrome: a Chinese cohort study. Atherosclerosis 2012; 220: 525–531. Yang T, Chu CH, Bai CH, You SL, Chou YC, Hwang LC et al. Uric acid concentration as a risk marker for blood pressure progression and incident hypertension: a Chinese cohort study. Metabolism 2012; 61: 1747–1755. Lu TH, Chang HY, Hwu CM, Chiu HC, Yin WH, Pan WH. Comparison of official coders versus physician panel in assignment of underlying cause of death. J Formos Med Assoc 2001; 100: 365–369. Wen CP, Tsai SP, Chung WS. A. 10-year experience with universal health insurance in Taiwan: measuring changes in health and health disparity. Ann Intern Med 2008; 148: 258–267. Woo D, Haverbusch M, Sekar P, Kissela B, Khoury J, Schneider A et al. Effect of untreated hypertension on hemorrhagic stroke. Stroke 2004; 35: 1703–1708. Zhang WW, Liu CY, Wang YJ, Xu ZQ, Chen Y, Zhou HD. Metabolic syndrome increases the risk of stroke: a 5-year follow-up study in a Chinese population. J Neurol 2009; 256: 1493–1499. Qiao Q, Laatikainen T, Zethelius B, Stegmayr B, Eliasson M, Jousilahti P et al. Comparison of definitions of metabolic syndrome in relation to the risk of developing stroke and coronary heart disease in Finnish and Swedish cohorts. Stroke 2009; 40: 337–343. Iribarren C, Go AS, Husson G, Sidney S, Fair JM, Quertermous T et al. Metabolic syndrome and early-onset coronary artery disease: is the whole greater than its parts? J Am Coll Cardiol 2006; 48: 1800–1807. Hanefeld M, Koehler C, Gallo S, Benke I, Ott P. Impact of the individual components of the metabolic syndrome and their different combinations on the prevalence of atherosclerotic vascular disease in type 2 diabetes: the Diabetes in Germany (DIG) study. Cardiovasc Diabetol 2007; 6: 13–19. Handa N, Matsumoto M, Maeda H, Hougaku H, Kamada T. Ischemic stroke events and carotid atherosclerosis: results of the Osaka Follow-up Study for Ultrasonographic Assessment of Carotid Atherosclerosis (the OSACA Study). Stroke 1995; 26: 1781–1786. Magai Y, Kitagawa K, Sakaguchi M, Shimizu Y, Hashimoto H, Yamagami H et al. Significance of earlier carotid atherosclerosis for stroke subtypes. Stroke 2001; 32: 1780–1785. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991; 22: 983–988.
Journal of Human Hypertension (2014) 1 – 5
