Proposed Criteria for Metabolic Syndrome in Japanese Based on Prospective Evidence
The Hisayama Study
Background and Purpose— The current criteria of metabolic syndrome (MetS) are not based on evidence derived from prospective studies on cardiovascular disease (CVD).
Methods— In a 14-year follow-up study of 2452 community-dwelling Japanese individuals aged ≥40 years, we examined which of the MetS criteria are most predictive for the development of CVD. During the follow-up, 246 first-ever CVD events occurred.
Results— An optimal cutoff point of waist circumference for predicting CVD was 90 cm in men (age-adjusted hazard ratio=1.81; 95% CI, 1.19 to 2.74; P=0.005) and 80 cm in women (age-adjusted hazard ratio=1.46; 95% CI, 0.99 to 2.16; P=0.05). A comparison of MetS criteria showed that the modified Japanese criteria using this cutoff point instead of the original definition were the strongest predictor of CVD events in both sexes (men: age-adjusted hazard ratio=2.58; 95% CI, 1.65 to 4.02; P<0.001; women: age-adjusted hazard ratio=2.39; 95% CI, 1.65 to 3.48; P<0.001). These observations remained robust even after adjustment for other confounding factors. According to this criteria set, only in the presence of central obesity, the hazard ratios for future CVD increased significantly as the number of MetS components increased, and a significant relationship was identified from 2 or more MetS components compared with individuals who had no MetS component.
Conclusions— Our findings suggest that the optimal cutoff point of waist circumference is 90 cm in men and 80 cm in women and that the modified Japanese criteria of MetS with this cutoff point as an essential component better predict CVD in the general Japanese population.
Metabolic syndrome (MetS) consists of a clustering of cardiovascular risk factors, and individuals with this condition have an elevated risk of developing cardiovascular diseases and type 2 diabetes.1 Practical and valuable criteria must be established promptly, because the prevalence of metabolic disorders has been increasing rapidly in recent years in Japan and other countries.2–4 Over the past decade, several institutions have proposed various criteria in attempts to define MetS as a diagnostic category. Among these, the criteria of the National Cholesterol Education Program’s Adult Treatment Panel III (NCEP) has most often been used in the literature.5 In this criteria set, the cutoff points of waist circumference were 102 cm in men and 88 cm in women; this parameter comprised a component of this syndrome but not a prerequisite for its diagnosis. However, this cutoff level may be unsuitable for Asian populations. For Japanese, 2 sets of diagnostic criteria of MetS exist at the present time, resulting in a great deal of confusion in clinical practice. One set is proposed by the Japanese Society of Internal Medicine (Japanese criteria)6; in these criteria, waist circumference is defined as an essential component, and its cutoff value is 85 cm for men and 90 cm for women.6 The other criteria set is offered by the International Diabetes Federation (IDF), in which ethnic-specific waist circumference cutoff points are used as a requirement of diagnosis.7 The IDF recommended cutoff levels of 90 cm in men and 80 cm in women for central obesity in Japanese individuals. In the current knowledge, it remains unclear which of these criteria or cutoff points of waist circumference are a better predictor of the development of cardiovascular disease (CVD) in the general population of Japanese. There has also been controversy as to whether the component of waist circumference should be considered a prerequisite for a diagnosis of MetS.8 The aim of the present article is to derive a better definition from the existing MetS criteria for predicting CVD in a prospective study of a defined general population of Japanese.
Materials and Methods
In 1988, a screening survey for the present study was performed in the town of Hisayama, a suburb of the Fukuoka metropolitan area on Japan’s Kyushu Island. The age and occupational distributions and nutritional intake of the population were almost identical to those of Japan as a whole based on data from the national census and nutrition survey.9 A detailed description of this survey was published previously.9 Briefly, a total of 2736 residents aged ≥40 years (80.7% of the total population of this age group) consented to participate in the examination and underwent a comprehensive assessment. After the exclusion of 102 subjects who had a history of coronary heart disease or stroke, as determined by a questionnaire and medical records, one subject for whom no blood sample was obtained, 120 subjects who had already eaten breakfast, and 61 subjects for whom waist circumference was not measured, the remaining 2452 subjects (1050 men and 1402 women) were enrolled in this study.
The subjects were followed prospectively from December 1988 to November 2002 by repeated health examinations. Health status was checked yearly by mail or telephone for any subjects who did not undergo a regular examination or who had moved out of town. We also established a daily monitoring system among the study team and local physicians or members of the town’s Health and Welfare Office. When a subject died, an autopsy was performed at the Departments of Pathology of Kyushu University. During the follow-up period, 479 subjects died, of whom 362 (75.6%) underwent autopsy. Only one subject was lost to follow-up.
Definition of Cardiovascular Events
CVD was defined as the development of ischemic stroke or coronary heart disease. Each CVD case was coded according to the International Classification of Disease, Ninth Revision (ICD-9) from 1988 to 1996 and Tenth Revision (ICD-10) from 1997 to 2002. Stroke was defined as a sudden onset of nonconvulsive and focal neurological deficit persisting for ≥24 hours.2 Each diagnosis of ischemic stroke (ICD-9: 434, ICD-10: I63) was made by 2 neurologists (Y.K. and Y.T.) separately using collected clinical and pathological information including brain CT/MRI and autopsy findings based on the Classification of Cerebrovascular Disease III proposed by the National Institute of Neurological Disorders and Stroke.10 Coronary heart disease included acute myocardial infarction (ICD-9: 410, ICD-10: I21), silent myocardial infarction (ICD-9: 412, ICD-10: I25.2), sudden cardiac death within 1 hour after the onset of acute illness (ICD-9: 798.1, ICD-10: I96.0), or coronary artery disease followed by coronary artery bypass surgery (ICD-9: E878.2, ICD-10: Z95.1) or angioplasty (ICD-9: E879.0, ICD-10: Z95.5).2 Acute myocardial infarction was diagnosed when a subject met at least 2 of the following criteria: (1) typical symptoms, including prolonged severe anterior chest pain; (2) cardiac enzyme levels more than twice the upper limit of the normal range; (3) evolving diagnostic electrocardiographic changes; and (4) morphological changes, including local asynergy of cardiac wall motion on echocardiography, persistent perfusion defect on cardiac scintigraphy, or myocardial necrosis or scars ≥1 cm long accompanied by coronary atherosclerosis at autopsy. Silent myocardial infarction was defined as myocardial scarring without any historical indication of clinical symptoms or abnormal cardiac enzyme changes. During the 14-year follow-up, 246 first-ever cardiovascular events (131 men and 115 women) occurred. Of these, there were 145 ischemic strokes (66 men and 79 women) and 125 cases of coronary heart disease (78 men and 47 women).
Risk Factor Measurements
At the baseline examination, waist circumference was measured by a trained staff member at the umbilical level with the subject standing. Body height and weight were measured in light clothing without shoes, and body mass index (BMI) was calculated.
To measure blood glucose and lipid levels, blood samples were collected from an antecubital vein between 8:00 and 10:30 am after an overnight fast of at least 12 hours. Blood for glucose assay was obtained by venipuncture into tubes containing sodium fluoride (NaF), and plasma glucose levels were determined by the glucose–oxidase method. Serum total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride concentrations were determined enzymatically. Blood pressure was measured 3 times using a standard mercury sphygmomanometer in the sitting position after the subject rested for at least 5 minutes. Freshly voided urine samples were collected at the screening, and proteinuria was defined as 1+ or more using a reagent strip. Electrocardiographic abnormalities were defined as left ventricular hypertrophy (Minnesota Code 3 to 1) and/or ST depression (Minnesota code 4–1, 2, 3).
Each participant completed a self-administered questionnaire covering medical history, smoking habits, alcohol intake, and exercise. The questionnaire was checked by trained interviewers at the screening. Smoking habits and alcohol intake were classified as either current habitual use or not. Those subjects who engaged in sports or other forms of exertion ≥3 times a week during their leisure time made up a regular exercise group.
Definition of Metabolic Syndrome
The Third Report of the NCEP Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults criteria5 of MetS include the presence of at least 3 of 5 factors: central obesity (waist circumference >102 cm in men, >88 cm in women), elevated blood pressure (blood pressure ≥130/85 mm Hg and/or current use of antihypertensive agents), elevated fasting plasma glucose (≥6.1 mmol/L and/or current use of antidiabetic medication), reduced HDL cholesterol (<1.03 mmol/L for men, <1.29 mmol/L for women), and elevated triglycerides (≥1.68 mmol/L).
In the IDF criteria,7 central obesity must be present for a diagnosis of MetS in addition to at least 2 of the other 4 factors. The IDF categories for Asians are: waist circumference ≥90 cm for men and ≥80 cm for women; blood pressure ≥130/85 mm Hg and/or current use of antihypertensive agents; fasting plasma glucose ≥5.6 mmol/L and/or current use of antidiabetic medication; HDL cholesterol <1.03 mmol/L in men and <1.29 mmol/L in women; and triglycerides ≥1.68 mmol/L.
Unlike the other criteria, the Japanese criteria6 consist of 4 factors, because they deal with HDL cholesterol and triglycerides together and the cutoff value of waist circumference is larger in women than in men. MetS in the Japanese criteria was diagnosed in individuals who had a high waist circumference (≥85 cm in men and ≥90 cm in women) plus any 2 of the following: (1) blood pressures of ≥130/85 mm Hg and/or current use of antihypertensive medicine; (2) fasting plasma glucose ≥6.1 mmol/L and/or current use of antidiabetic medication; and (3) triglycerides ≥1.68 mmol/L and/or HDL cholesterol <1.03 mmol/L in men and women. Additionally, we created 2 new criteria. The modified NCEP and Japanese criteria used a waist circumference of ≥90 cm in men and ≥80 cm in women instead of the original cutoff points.
The SAS software package Version 8.2 (SAS Institute, Cary, NC) was used to perform all statistical analyses. Serum triglycerides were transformed into logarithms to improve the skewed distribution. The age- and multivariate-adjusted hazard ratios (HRs) and 95% CIs were estimated with the use of the Cox proportional hazards model. To find the cutoff value of abdominal obesity, we also plotted receiver operating characteristic curves. In this method, the optimal cutoff value of abdominal obesity was defined by maximizing the sensitivity and specificity to the development of CVD.11 In addition, population-attributable risk percent was estimated for various MetS criteria sets with the following formula: prevalence×(HR−1)/[prevalence×(HR−1)+1].
This study was conducted with the approval of the Ethics Committee of the Faculty of Medicine, Kyushu University, and written informed consent was obtained from all of the participants.
Table 1 shows the subjects’ baseline clinical characteristics by sex. The prevalence of MetS defined by the NCEP, IDF for Asians, modified NCEP, and modified Japanese criteria was significantly higher in women than in men, whereas the prevalence of MetS by the Japanese criteria was higher in men. Mean values of waist circumference, systolic and diastolic blood pressures, fasting plasma glucose and triglyceride levels, and frequencies of elevated blood pressure, fasting plasma glucose, and triglycerides, reduced HDL cholesterol, proteinuria, electrocardiographic abnormalities, alcohol intake, and smoking habits were significantly higher in men than in women, whereas women had higher total and HDL cholesterol concentrations. Mean age and BMI and frequency of regular exercise did not differ between the sexes.
To compare the ability to predict CVD at each published cutoff level of waist circumference among the NCEP, IDF, and Japanese MetS criteria, we estimated the age-adjusted HRs and 95% CIs by sex (Table 2). In men, the age-adjusted HR of incident CVD was significantly higher in subjects with a waist of ≥90 cm (IDF criteria for Asians) than in those with a smaller waist (age-adjusted HR=1.81; 95% CI, 1.19 to 2.74; P=0.005), whereas in women, this association was marginally significant at the cutoff level of ≥80 cm (age-adjusted HR=1.46; 95% CI, 0.99 to 2.16; P=0.05). The levels of central obesity determined by the cutoff levels of waist circumference proposed by the NCEP, IDF for Europids, and Japanese criteria were not significant predictors of CVD in either sex.
In the analysis with the receiver operating characteristic curve method, the cutoff point defined as the maximum combination of sensitivity and specificity was 80.2 cm for men and 81.5 cm for women. This cutoff point significantly predicted CVD in women but did not in men (men: age-adjusted HR=1.30; 95% CI, 0.91 to 1.85; P=0.15; women: age-adjusted HR=1.61; 95% CI, 1.11 to 2.35; P=0.01).
Age- and multivariate-adjusted HRs and population-attributable risk percents of various MetS criteria for the development of CVD were estimated by sex (Table 3). The age-adjusted analyses showed that MetS defined by all of the criteria sets, except for the Japanese one in men, was a significant risk factor for CVD. Among these, MetS as determined by the modified Japanese criteria was the strongest predictor for the development of CVD in both sexes (men: age-adjusted HR=2.58; 95% CI, 1.65 to 4.02; P<0.001; women: age-adjusted HR=2.39; 95% CI, 1.65 to 3.48; P<0.001). These findings remained substantially unchanged even after adjustment for the following confounding factors: age, serum total cholesterol, proteinuria, electrocardiographic abnormalities, alcohol intake, smoking habits, and regular exercise. When we divided CVD into ischemic stroke and coronary heart disease, the age-adjusted incidence of ischemic stroke was significantly higher in subjects with MetS defined by the modified Japanese criteria than those without MetS for both sexes (men: 18.0 versus 5.2 per 1000 person-years. P<0.001: women: 9.2 versus 4.0, P<0.001). The same was true incidence of coronary heart disease in both sexes (mean: 10.4 versus 6.4, P=0.003; women: 6.7 versus 2.0, P<0.001). These associations remained significant even after adjustment for the previously mentioned confounding factors (ischemic stroke: HR=3.07; 95% CI, 1.68 to 5.61; P<0.001, in men; HR=2.21; 95% CI, 1.39 to 3.51; P<0.001, in women; coronary heart disease: HR=2.37; 95% CI, 1.28 to 4.39; P=0.006, in men; HR=2.91; 95% CI, 1.62 to 5.22; P<0.001, in women). On the other hand, the multivariate-adjusted population-attributable risk percents for MetS defined by the IDF, modified NCEP, and modified Japanese criteria were comparably higher than those for MetS defined by the other criteria in both sexes, and all of the population-attributable risk percents were larger in women than in men.
To investigate the necessity of central obesity defined as a waist circumference of ≥90 cm in men and ≥80 cm in women for predicting CVD in the modified Japanese criteria, the previously mentioned risk factor-adjusted HR according to the number of MetS components other than waist circumference were estimated by the presence or absence of central obesity (Table 4). In the subjects who had central obesity, the HR of CVD increased significantly as the number of MetS components increased, whereas this trend was not observed in the subjects without central obesity. In the subjects with central obesity, the risk of CVD significantly increased if subjects had 2 or more MetS components compared with individuals who had no MetS component (one component: adjusted HR=1.13; 95% CI, 0.53 to 2.40; P=0.74; 2 components: adjusted HR=2.47; 95% CI, 1.21 to 5.04; P=0.01; 3 components: adjusted HR=3.09; 95% CI, 1.40 to 6.79; P=0.005). Similar relationships were found when CVD was stratified into ischemic stroke and coronary heart disease.
Because diabetes and hypertension are strong risk factors for CVD, we examined both the combined and separate effects of MetS and diabetes or hypertension on the development of CVD. As shown in Table 5, compared with nondiabetic subjects without MetS, nondiabetic subjects with MetS had significantly higher multivariate-adjusted HR of ischemic stroke (adjusted HR=1.65; 95% CI, 1.04 to 2.62; P=0.03); HR was markedly higher than that in diabetic subjects with MetS (adjusted HR=5.35; 95% CI, 3.28 to 8.73; P<0.001). However, no elevation was found in diabetic subjects without MetS. Similar associations were observed for coronary heart disease. Likewise, the multivariate-adjusted HR of ischemic stroke was significantly higher in normotensive subjects with MetS (adjusted HR=2.13; 95% CI, 1.03 to 4.39; P=0.04) and in hypertensive subjects with MetS (adjusted HR=3.17; 95% CI, 2.01 to 5.02; P<0.001) but was not significant in hypertensive subjects without MetS. Similar patterns were seen for coronary heart disease. Significant interactions between MetS and diabetes were revealed in the risk of ischemic stroke and coronary heart disease (P<0.01), whereas the interactions between MetS and hypertension were not significant.
Using data from a 14-year follow-up study of a general Japanese population, we demonstrated that the optimal cutoff point of waist circumference for predicting CVD in Japanese was 90 cm in men and 80 cm in women. In the comparison of various MetS criteria, the modified Japanese criteria set, which uses this cutoff point instead of the original one, was a better predictor for incident CVD in both sexes. According to this criteria set, in subjects with central obesity only, the HR of future CVD increased as the number of MetS components increased, and a significantly elevated risk was identified in subjects who had ≥2 MetS components compared with those who had no MetS component. Furthermore, the significant effects of MetS on the development of ischemic stroke and coronary heart disease were independent of hypertension and diabetes. These findings suggest that the modified Japanese criteria are better for predicting CVD in Japanese.
The existence of different criteria sets for MetS has caused a great deal of confusion in routine practice in Japan. Whereas the IDF criteria are recommended internationally, the Japanese criteria are commonly used in Japan. The established MetS criteria are based mainly on “expert” opinions, and the evidence derived from prospective studies is scarce.12 Thus, it remains uncertain whether the threshold at which each MetS component is defined as positive or negative is optimal or even useful for predicting the risk of CVD. The findings of our study indicate that the definition of MetS by the modified Japanese criteria confers greater accuracy in predicting CVD events compared with the other ones. There are some possible explanations for this superiority. First, this criteria set adopted the optimal cutoff value of waist circumference for predicting vascular events in the present cohort. An optimal cutoff point of waist circumference for having cardiovascular risk factors has been discussed extensively in several cross-sectional studies of Asian populations. Hara et al showed in a receiver operating characteristic analysis that 85 cm for men and 78 cm for women were the best values for predicting other MetS features in a Japanese population.13 Similar analyses reported that 90 cm in men and 84 cm in women was optimal in Japanese American14 and 85 cm in men and 80 cm in women in Chinese populations.15,16 However, no studies showed an optimal cutoff value of waist circumference for CVD risk in a prospective cohort design. Our finding is the first evidence that the optimal cutoff point of waist circumference for predicting CVD was 90 cm in men and 80 cm in women in a general Japanese population. This evidence might be extrapolated to other Asian populations having similar physiques and genetics.
Second, when we used our modified Japanese criteria, the HR of cardiovascular events rose obviously as the number of MetS components increased only in subjects with central obesity. Thus, to treat waist circumference as an essential component would likely improve the precision of the prediction of cardiovascular events in the current subjects. There has been controversy over the necessity of central obesity for the diagnosis of MetS. In NIPPON DATA90, a Japanese cohort study, the risk of CVD death increased significantly as the number of MetS components rose both in nonobese participants and obese ones.8 On the other hand, in the present study, a clear trend in the risk of CVD occurrence was observed only in the subjects with central obesity. This inconsistency in findings might be caused by the difference in populations and the definition of obesity. In the NIPPON DATA90 study, BMI was substituted for waist circumference in the MetS definition. However, there is often remarkable heterogeneity of waist circumference among individuals with similar BMI values. It has been also shown that, among obese individuals, waist circumference indicates an increased risk of CVD, and this association is independent of the risk predicted by increased BMI.17 Thus, the use of BMI instead of waist circumference may lead to a misdiagnosis of MetS.
In the present study, the risk of CVD occurrence was higher for the modified Japanese MetS criteria than for the IDF criteria despite the identical condition regarding central obesity. One possibility for this is that the definitions of hyperglycemia and dyslipidemia are different between the 2 sets of MetS criteria. In our subjects, the definitions of hyperglycemia and dyslipidemia in Japanese criteria were superior to those in the other criteria for the prediction of the development of CVD (data not shown). These facts may explain why the modified Japanese criteria had a higher HR. Further studies are needed to optimize the cutoff points of fasting plasma glucose and lipid levels for predicting cardiovascular events.
In our study, there was no large difference in waist circumference between our men and women (82.0 cm versus 81.1 cm). On the other hand, the optimal cutoff point of waist circumference for predicting CVD was lower in women (80 cm) than in men (90 cm). It is known that men are prone to intra-abdominal fat accumulation, whereas women are prone to subcutaneous fat accumulation.18 Because men would have more intra-abdominal fat than women at a given waist circumference, it may be valid to select a lower cutpoint of waist circumference for men than for women. However, recent epidemiological studies using CT revealed that women who had more visceral fat tended to have more metabolic risk factors for CVD compared with men.19,20 The cause of this sex difference is uncertain but may be related to a higher amount of hepatic free fatty acid delivery derived from visceral fat in women than in men.21 These findings imply that it is reasonable to choose the lower cutoff point of waist circumference for women than for men. Furthermore, the population-attributable risk percents for any MetS criteria sets were larger in women than in men. These findings also suggest that MetS has a stronger influence on women than on men.
The American Diabetes Association/European Association for the Study of Diabetes says that MetS has been imprecisely defined, that its pathogenesis is uncertain, and that its value as a CVD risk marker is doubtful. Furthermore, it recommends that clinicians should evaluate and manage all CVD risk factors without regard to whether a patient meets the criteria for a diagnosis of MetS. Certainly, Sone et al documented that the diagnosis of MetS using the modified NCEP criteria was not useful for predicting CVD in patients with diabetes.22 However, our stratified analysis indicated that MetS is a significant risk factor for CVD in both nondiabetic and normotensive individuals. Moreover, the present study revealed that the risk of CVD was higher in subjects with MetS than in those with diabetes or hypertension. These results imply that MetS plays a main role in the development of CVD in the general population, including patients with mild diabetes and hypertension. In the general Japanese population, blood pressure levels decreased significantly with time due to the increment in the use of antihypertensive medication, whereas metabolic disorders greatly increased in recent periods.2,23 Even with advances in therapeutic agents, it is difficult to treat MetS and diabetes because lifestyle modifications are also needed. These disorders remain large problems for the prevention of CVD, especially in developed countries.
Additionally, our subjects showed a synergistic effect between MetS and diabetes for the development of CVD. The conditions of MetS are accompanied by adipokine disorders, inducing inflammatory cytokines and immune response, and endothelial dysfunction, which promotes the development of atherosclerosis.24 On the other hand, hyperglycemia in diabetes itself directly affects the progression of atherosclerosis through the increase in nonenzymatic glycation of proteins and lipids,25 the production of reactive oxygen species,26 and the activation of protein kinase C27 isoform and the hexosamine biosynthetic pathway.28 It is therefore speculated that MetS and diabetes mutually enhance the risk of CVD by distinct mechanisms.
In our men, the cutoff value of waist circumference derived from the receiver operating characteristic analysis (80.2 cm) was much lower than that derived from the cohort study (90 cm), and the former was not a significant predictor of incident CVD in the follow-up study. This suggests that a value defined by maximizing the sensitivity and specificity would be not always best.
The strengths of our study include its longitudinal population-based design, the long duration of follow-up, the sufficient number of CVD events, and the almost perfect follow-up of subjects. However, 2 limitations of the present study should be discussed. One is that the diagnosis of MetS was based on a single measurement of its components at baseline as was the case in other epidemiological studies. During the follow-up, risk factor levels were changed due to modifications in lifestyle or medication, and misclassification of MetS was possible. This would weaken the association found in this study, biasing the results toward the null hypothesis. Therefore, the true association may be stronger than that shown in our study. The other limitation is that the present study lacked information on drugs, fibrates, and nicotinic acid, affecting the metabolism of HDL cholesterol and triglycerides. However, these medications were rarely used in our country at this study’s 1988 baseline. This suggests that such a bias did not invalidate the present findings.
In conclusion, the present analysis has clearly demonstrated that the optimal cutoff point of waist circumference is 90 cm in men and 80 cm in women and that the modified Japanese criteria of MetS with this cutoff point as an essential component better predicted CVD in the general Japanese population than did the other criteria sets. Furthermore, the increasing effects of MetS on the development of ischemic stroke and coronary heart disease were independent of hypertension and diabetes. High-risk strategies using this criteria set offer additional protection against CVD.
We thank the staff of the Division of Health and Welfare of Hisayama for their cooperation in this study.
Sources of Funding
This study was supported in part by a Grant-in-Aid for Scientific Research C (No. 20591063) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
- Received July 12, 2008.
- Revision received September 30, 2008.
- Accepted October 22, 2008.
Ford ES. Risks for all-cause mortality, cardiovascular disease, and diabetes associated with the metabolic syndrome: a summary of the evidence. Diabetes Care. 2005; 28: 1769–1778.
Kubo M, Kiyohara Y, Kato I, Tanizaki Y, Arima H, Tanaka K, Nakamura H, Okubo K, Iida M. Trends in the incidence, mortality, and survival rate of cardiovascular disease in a Japanese community: the Hisayama Study. Stroke. 2003; 34: 2349–2354.
Ford ES, Giles WH, Mokdad AH. Increasing prevalence of the metabolic syndrome among US adults. Diabetes Care. 2004; 27: 2444–2449.
Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001; 285: 2486–2497.
Kadota A, Hozawa A, Okamura T, Kadowak T, Nakmaura K, Murakami Y, Hayakawa T, Kita Y, Okayama A, Nakamura Y, Kashiwagi A, Ueshima H. Relationship between metabolic risk factor clustering and cardiovascular mortality stratified by high blood glucose and obesity: NIPPON DATA90, 1990–2000. Diabetes Care. 2007; 30: 1533–1538.
Ohmura T, Ueda K, Kiyohara Y, Kato I, Iwamoto H, Nakayama K, Nomiyama K, Ohmori S, Yoshitake T, Shinkawa A, Hasuo Y, Fujishima M. Prevalence of type 2 (non-insulin-dependent) diabetes mellitus and impaired glucose tolerance in the Japanese general population: the Hisayama Study. Diabetologia. 1993; 36: 1198–1203.
Special report from the National Institute of Neurological Disorders and Stroke. Classification of cerebrovascular diseases III. Stroke. 1990; 21: 637–676.
Franks PW, Olsson T. Metabolic syndrome and early death: getting to the heart of the problem. Hypertension. 2007; 49: 10–12.
Hara K, Matsushita Y, Horikoshi M, Yoshiike N, Yokoyama T, Tanaka H, Kadowaki T. A proposal for the cutoff point of waist circumference for the diagnosis of metabolic syndrome in the Japanese population. Diabetes Care. 2006; 29: 1123–1124.
Hayashi T, Boyko EJ, McNeely MJ, Leonetti DL, Kahn SE, Fujimoto WY. Minimum waist and visceral fat values for identifying Japanese Americans at risk for the metabolic syndrome. Diabetes Care. 2007; 30: 120–127.
Ashwell M. Obesity in men and women. Int J Obes Relat Metab Disord. 1994; 18 (suppl 1): S1–7.
Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY, Vasan RS, Murabito JM, Meigs JB, Cupples LA, D'Agostino RB Sr, O'Donnell CJ. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007; 116: 39–48.
Sone H, Mizuno S, Fujii H, Yoshimura Y, Yamasaki Y, Ishibashi S, Katayama S, Saito Y, Ito H, Ohashi Y, Akanuma Y, Yamada N. Is the diagnosis of metabolic syndrome useful for predicting cardiovascular disease in Asian diabetic patients? Analysis from the Japan Diabetes Complications Study. Diabetes Care. 2005; 28: 1463–1471.
Kubo M, Kiyohara Y, Ninomiya T, Tanizaki Y, Yonemoto K, Doi Y, Hata J, Oishi Y, Shikata K, Iida M. Decreasing incidence of lacunar vs other types of cerebral infarction in a Japanese population. Neurology. 2006; 66: 1539–1544.
Kirstein M, Brett J, Radoff S, Ogawa S, Stern D, Vlassara H. Advanced protein glycosylation induces transendothelial human monocyte chemotaxis and secretion of platelet-derived growth factor: role in vascular disease of diabetes and aging. Proc Natl Acad Sci U S A. 1990; 87: 9010–9014.
Esposito K, Nappo F, Marfella R, Giugliano G, Giugliano F, Ciotola M, Quagliaro L, Ceriello A, Giugliano D. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation. 2002; 106: 2067–2072.
Koya D, King GL. Protein kinase C activation and the development of diabetic complications. Diabetes. 1998; 47: 859–866.
James LR, Tang D, Ingram A, Ly H, Thai K, Cai L, Scholey JW. Flux through the hexosamine pathway is a determinant of nuclear factor κB-dependent promoter activation. Diabetes. 2002; 51: 1146–1156.