Criteria of Waist Circumference According to Computed Tomography-Measured Visceral Fat Area and the Clustering of Cardiovascular Risk Factors

Transcription

1 ORIGINAL ARTICLE Epiemiology Circ J 29; 73: Criteria of Waist Circumference Accoring to Compute Tomography-Measure Visceral Fat Area an the Clustering of Cariovascular Risk Factors Hietoshi Kashihara, PhD; Jung Su Lee, PhD*; Kiyoshi Kawakubo, MD**; Masaki Tamura, MD; Akira Akabayashi, MD Backgroun: The purpose of this stuy was to etermine the iscriminate gener-specific cutoff values of waist circumference (WC) for etecting the clustering of cariovascular risk factors (CCRF), which reflects the intraabominal visceral fat area (VFA) using a large Japanese population. Methos an Results: The stuy population consiste of 6,736 men an women who unerwent a perioic health check-up an ha a compute tomography scan for VFA measurement at the PL Tokyo Health Care Center in Tokyo, Japan. The CCRF was efine accoring to the Japanese Committee of the Criteria for Metabolic Synrome. The iscriminate values for etecting the CCRF were teste using receiver operating characteristics analysis. The iscriminate values of VFA were 3. cm 2 with 68.7% sensitivity an 6.8% specificity for men an 69. cm 2 with 8% sensitivity an 7.% specificity for women. The WC values corresponing to the VFA were 89. cm for men an 86.3 cm for women. The iscriminate values of VFA an WC were not substantially ifferent between people with or without raise bloo pressure. Conclusions: For the early etection an management of the CCRF an VFA in the primary screening setting, the iscriminate an convenient gener-specific WC criteria woul be 9 cm for men an 85 cm for women. (Circ J 29; 73: ) Key Wors: Cariovascular risk factors; Intra-abominal visceral fat area (VFA); Metabolic synrome; Screening; Waist circumference A nationwie prevention strategy for cariovascular iseases, incluing a specific health check-up an health eucation for the metabolic synrome, starte with the fiscal year 28 in Japan. The purpose of the strategy is to promote the early etection an management of the clustering of cariovascular risk factors (CCRF); thus, the waist circumference (WC) measurement was introuce to health check-ups for the first time in Japan. The CCRF that occurs in the same iniviual appears to confer a substantial aitional cariovascular risk over an above the sum of the risk associate with each abnormality.,2 Multiple risk factors that commonly appear together are now known as the metabolic synrome, 3 which increases the morbiity an mortality of cariovascular iseases. 4 7 The metabolic synrome has ramatically increase in evelope countries, incluing Japan. 8,9 Intra-abominal visceral fat accumulation is wiely recognize as a cause of metabolic isorers. 5 A simple an convenient inicator that reflects the visceral fat mass is neee in the primary screening setting. Measuring WC is a simple tool an WC is wiely recognize as a measure of metabolic abnormalities. 6 2 The Japanese Committee of the Criteria for Metabolic Synrome (JCCMS) 2 propose criteria for practical use in 25, an WC was inclue as a primary crucial criterion. The JCCMS current criteria for WC for men an women are 85 an 9 cm, respectively. 2 The WC cutoff value was etermine using the cm 2 area of intra-abominal visceral fat area (VFA), 6 because the mean number of metabolic isorers (ie, hyperglycemia, yslipiemia, an hypertension) increase with VFA, an the average number of metabolic isorers was more than. at cm 2 of VFA. 22 However, that stuy s population was relatively small an gener ha not been consiere when evaluating the VFA values to etect the CCRF. 6 Only a few comparable Japanese stuies have been conucte to simultaneously reassess appropriate VFA an WC cutoff values. 22,23 These stuies propose much smaller cutoff values of VFA an WC for women than the current Japanese criteria. 2 Therefore, it is necessary to reconfirm the gener-specific correlation between VFA an CCRF an to clarify the gener-specific cutoff values of WC corresponing to the new VFA values using a large, general Japanese population with a wie range of boy mass. Raise bloo pressure (BP) is well known as the most common component of the metabolic synrome, 9,24 an the (Receive March 22, 29; revise manuscript receive May 26, 29; accepte May 28, 29; release online August 4, 29) PL Tokyo Health Care Center, *Department of Health Promotion Sciences, Grauate School of Meicine, The University of Tokyo, **Department of Foo Sciences & Nutrition, Kyoritsu s University an Department of Biomeical Ethics, School of Public Health, Grauate School of Meicine, The University of Tokyo, Tokyo, Japan Mailing aress: Jung Su Lee, PhD, Department of Health Promotion Sciences, Grauate School of Meicine, The University of Tokyo, 7-3- Hongo, Bunkyo-ku, Tokyo 3-33, Japan. jslee@m.u-tokyo.ac.jp All rights are reserve to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp

2 882 KASHIHARA H et al. Table. Characteristics of the Stuy Population N 5,8,656 Age (years) 49.2± ± 3 74 Boy mass inex* 25.2± ± WC (cm) 88.3± ± Intra VFA (cm 2 ) ± ± Patients with risk factors** 2, % % risk factor, % % 2 risk factors % 66 4.% 3 risk factors 55 3.% 7 % Mean ± SD or number of subjects an proportion of subjects (%). *Boy weight (kg)/height (m) 2. **Risk factors: raise BP, raise triglyceries an/or reuce HDL-C, an raise fasting plasma glucose. WC, waist circumference; VFA, visceral fat areas; BP, bloo pressure; HDL-C, high-ensity lipoprotein-cholesterol. most attributable risk factor for all-cause mortality in the Japanese. 25 However, it is still not clear whether the characteristic features iffer between those who have raise BP an those who o not. Thus, it is imperative to compare the gener-specific iscriminate cutoff values of VFA an WC for etecting CCRF between cases with or without raise BP. The present stuy use a relatively large Japanese general population to: () etermine gener-specific iscriminate cutoff values of VFA for etecting the CCRF, (2) etermine gener-specific cutoff values of WC reflecting the new VFA values, an (3) compare the iscriminate gener-specific cutoff values of VFA an WC for etecting CCRF between cases with or without raise BP. Methos Stuy Population The eligible stuy population consiste of 6,736 men an women (age range: 3 74 years ol) who unerwent an annual perioic health check-up an also ha a compute tomography (CT) scan for VFA measurement from October 25 to December 28 at the PL Tokyo Health Care Center in Tokyo, Japan. People who were on meication for iabetes, hyperlipiemia, or hypertension were exclue from the stuy population. This center is a perioic health examination facility mainly for insure employees an their family members. The present stuy was conucte in accorance with the ethical principles state in the Declaration of Helsinki (by the Worl Meical Association). The autonomy of the participants was fully respecte an security of personal information was provie to participants. Measurements The WC (cm) was measure at the navel in a staning position to the nearest. cm by a technician when the participants were in a morning fasting state. Systolic an iastolic BP (SBP, DBP) were measure using an automatic BP monitor (UDEX Super type, ELK Corporation, Tokyo, Japan) after the participant ha reste while seate for 3 min. Bloo was taken from the anterior cubital vein in the fasting state, an triglyceries (TG), highensity lipoprotein-cholesterol (HDL-C), an fasting plasma glucose (FPG) were analyze using enzymatic methos with an automatic analyzer (Hitachi 745 Automatic Analyzer, Hitachi, Tokyo, Japan). Quality control for bloo testing was performe every ay using poole stanar bloo samples. Coefficients of variation of TG, HDL-C, an FPG in this measurement system were less than.%. The CCRF of the metabolic synrome was efine as 2 or more of the following 3 risk factors accoring to the JCCMS criteria 2 : () TG 5 mg/l, an/or HDL-C <4 mg/l, (2) SBP 3 mmhg an/or DBP 85 mmhg, an (3) FPG mg/l. The VFA was measure from CT cross-sectional scans at the navel (ROBUSTO; Hitachi, Tokyo, Japan) accoring to the reporte stanar metho. 26 Statistical Analysis First, the gener-specific iscriminate cutoff values of VFA or WC for etecting the CCRF were teste using receiver operating characteristics (ROC) analysis. The sensitivity an specificity were calculate for every cm 2 of VFA or cm of WC, an the iscriminate point for the CCRF (the least istance from upper-left to VFA or WC ROC curve) was etermine. The calculation formula is ( ) 2 + ( Specificity) 2. The area uner the ROC curve (AUC) an 95% confience interval (CI) were also calculate. Next, the gener-specific WC cutoff value accoring to the VFA cutoff value was etermine using simple regression analysis between VFA an WC. Further, the gener-specific iscriminate cutoff values of VFA or WC for etecting the CCRF were compare between those with raise BP (SBP 3 mmhg an/or DBP 85 mmhg) an those without raise BP using ROC analysis. Results Table shows the characteristics of the stuy population. The mean age was 49.2 an 5.4 years for men an women, respectively. The mean boy mass inex (BMI) was 25.2 for men an 23.2 for women. The percentage of people who ha 2 or more cariovascular risk factors was 7.% for men an 4.4% for women. Figure shows the gener-specific ROC curve of VFA for etecting the CCRF. AUC was.7 (95%CI 9.72) for men an 3 (95%CI.79 7) for women, an the AUC was statistically significant for both men an women. From the ROC curve, iscriminate VFA values for etecting the CCRF were 3. cm 2 for men an 69. cm 2 for women. At these cutoff points, sensitivity an specificity were 68.7% an 6.8%, respectively, for men, an 8%

3 Waist Cutoff Values Using VFA an Risk Clustering 883 -Specificity - VFA = 3 cm 2 = 87 Specificity = 8 -Specificity - VFA = 69 cm 2 = 8 Specificity =.7 VFA = cm 2 =.384 Specificity =.94 -Specificity -Specificity Figure. Gener-specific iscriminate value of intra-abominal visceral fat area (VFA) for clustering of cariovascular risk factors using receiver operating characteristics (ROC) analysis., least istance from upper-left to intra abominal VFA ROC. WC = 85 cm 2 = 54 Specificity =.333 WC = 84 cm 2 =.74 Specificity = 33 WC = 89 cm 2 = 8 Specificity = 2 WC = 9 cm 2 = 25 Specificity = 4 -Specificity -Specificity Figure 2. Gener-specific iscriminate value of waist circumference (WC) for clustering of cariovascular risk factors using receiver operating characteristics (ROC) analysis., least istance from upper-left to WC ROC cm 2 cm 2 5 VFA = 3.42 WC 2.7 R = 27 3 cm 2 69 cm cm cm VFA = 2.64 WC 59. R = cm cm Figure 3. Gener-specific iscriminate value of waist circumference (WC) reflecting the intra-abominal visceral fat area (VFA).

4 884 KASHIHARA H et al. Table 2. Comparison Between Patients With an Without Raise BP Without raise BP With raise BP Without raise BP With raise BP N 3,886,94, Age (years) 48.± ±. 5.7±. 56.5±9. Boy mass inex* 24.9± ± ± ±4. Clustering risk factors** 232 (6.) 63 (52.8) 7 (.2) 56 (24.7) Mean ± SD or number of subjects (%). *Boy weight (kg)/height (m) 2. **Patients who have 2 or more of the following risk factors: raise systolic BP an/or iastolic BP, raise triglyceries an/or reuce HDL-C, an raise fasting plasma glucose. Abbreviations see in Table. Table 3. Gener-Specific Discriminate Cutoff VFA an WC for Detecting Clustering of Cariovascular Risk Factors Incluing Patients With an Without Raise BP Without raise BP With raise BP Without raise BP With raise BP VFA Value of least istance*, Specificity AUC %CI of AUC** WC Value of least istance* Specificity AUC %CI of AUC** *Least istance from upper-left to ROC curve. **95%CI of AUC. CI, confience interval; AUC, area uner the receiver operating characteristics (ROC) curve. Other abbreviations see in Table. an 7.%, respectively for women. When cm 2 of VFA was use for women, specificity was 9.4%, but sensitivity was only 38.4%. Figure 2 shows the gener-specific ROC curve of WC for etecting the CCRF. AUC was 5 (95%CI 3 7) for men an.73 (95%CI 7.78) for women, an the AUC was statistically significant for both men an women. From the ROC curve, iscriminate WC values for etecting the CCRF were 89. cm for men an 84. cm for women. At these cutoff points, sensitivity an specificity were 6.8% an 6%, respectively, for men, an 74.% an 63.3%, respectively, for women. However, specificity was only 33.3% when 85 cm of JCCSM WC was use for men, an sensitivity was only 42.5% when 9 cm of JCCSM WC was use for women. Gener-specific WC values corresponing to the iscriminate VFA values were confirme using a scattergram (Figure 3). Each WC cutoff point corresponing to a VFA of 3. cm 2 for men an 69. cm 2 for women was calculate using a simple regression formula. Results inicate that the WC cutoff points were 89. cm for men an 86.3 cm for women. Table 2 compares the characteristic features of those with an without raise BP. The mean age an BMI were significantly higher in people with raise BP than those without in both men an women. A proportion of people who ha CCRF was also significantly higher in people with raise BP than those without raise BP in both men an women. Table 3 shows the comparison of iscriminate VFA an WC cutoff values for etecting the CCRF using ROC analysis between those with an without raise BP. The iscriminate VFA cutoff value for men with raise BP was 8. cm 2, which was slightly higher than the 3. cm 2 for men without raise BP. The iscriminate VFA cutoff value was 7. cm 2 an 73. cm 2 for women with an without raise BP, respectively. The iscriminate WC cutoff value for men both with an without raise BP was 89. cm. The iscriminate WC cutoff value was 84. cm an 83. cm for women with an without raise BP, respectively. The iscriminate WC cutoff values i not iffer between those with an without raise BP in both men an women. Discussion The purpose of the present stuy was to etermine the gener-specific iscriminate criteria of WC for etecting the CCRF of metabolic synrome using a large number of Japanese men an women from the general population with a wie range of boy mass. From the stuy results, the gener-specific cutoff values of VFA for etecting the CCRF is 3. cm 2 for men an 69. cm 2 for women, an the WC values corresponing to these VFA values are 89. cm for men an 86.3 cm for women. When WC for etecting the CCRF was evaluate using the ROC analysis, the iscriminate WC cutoff values are 89. cm for men an 84. cm for women. These WC cutoff values are quite similar to the WC values corresponing to the VFA values etecting the CCRF. Therefore, for the early etection an management of the CCRF an VFA in the primary screen-

5 Waist Cutoff Values Using VFA an Risk Clustering ing setting, the iscriminate an convenient gener-specific WC criteria woul be 9 cm for men an 85 cm for women. The other stuy purpose was to compare the iscriminate gener-specific cutoff values of VFA an WC for etecting CCRF between those with an without raise BP. The iscriminate cutoff value of VFA was 5 cm 2 larger in men with raise BP than that in men without raise BP. However, the iscriminate VFA value was 7 cm 2 for women with raise BP an 2 cm 2 smaller than the value for women without raise BP. These ifferences were too small to iscuss any specific correlation between raise BP an VFA in men an women. Further stuy focusing on this issue is neee. From the present stuy, we propose a WC of 9 cm for men an 85 cm for women, which are the iscriminate an convenient gener-specific WC criteria. The cutoff value of 9 cm for men is larger than that of the JCCMS criteria 6,2 an the cutoff value of 85 cm for women is smaller. The criteria of JCCMS were etermine using the cm 2 of VFA 6 among a relatively small stuy population (775 men an 48 women). In that analysis, gener ifference was not consiere. WC was measure using gener-specific simple regression analysis between WC an VFA only for 554 men an 94 women. 6 In the present stuy, specificity was only 33.3% for the 85 cm cutoff value of the JCCMS 6,2 for men. In contrast, sensitivity for etecting the CCRF was only 42.5% for the 9 cm cutoff value of the JCCMS for women. In Japan, a nationwie strategy incluing a specific health checkup an health eucation about the metabolic synrome was initiate. In the specific health checkup, WC is a prerequisite conition of the initial assessment. To achieve the purpose of the strategy, a sensitive WC cutoff value is neee to ifferentiate among those people who have CCRF. The CCRF of the metabolic synrome is prevalent even in non-obese people an increases the risk of ischemic heart isease an stroke 27 an cariovascular mortality 28 in the Japanese. If the sensitivity of WC is too low, most of the people who have CCRF will not be inclue in the target population for the early etection an management of CCRF. Therefore, it is important to etermine the iscriminate cutoff value of WC with high sensitivity. Previous limite Japanese stuies 29 3 also recommene much smaller WC values for women than the JCCMS criteria. However, they i not consier the WC value with the irectly measure VFA using a CT scan. In the present stuy, VFA an WC were measure by traine technicians, an the measuring site was at the navel accoring to the JCCMS criteria. A few comparable Japanese stuies have been conucte to simultaneously reconfirm appropriate VFA an WC cutoff values. 22,23 Miyawaki et al propose WC cutoff values of 86 cm for men an 77 cm for women, 22 whereas Oka et al propose WC cutoff values 89.8 cm for men an 82.3 cm for women. 23 The cutoff values of WC for women were much smaller than the current Japanese criteria, 2 as in the present stuy, an the results of these stuies, especially those of Oka et al, 23 were quite similar to the results of the present stuy. Therefore, the cutoff values of 9 cm for men an 85 cm for women seem to be simple an convenient for gener-specific WC criteria for the Japanese population. The present stuy population was a general population comprising insure people an their families living in or aroun the Tokyo metropolitan area. The mean BMI an the prevalence of cariovascular risk factors of the nationwie representative sample use in the Japanese Health an 885 Nutrition Survey 9 were quite similar to the present stuy. Thus, the stuy population was not biase against the general Japanese population. The stuy analyses are base on cross-sectional ata; therefore, a cause an effect relationship between WC an cariovascular iseases cannot be rawn. In orer to more accurately efine the cutoff values for WC in the Japanese population, we shoul compare WC cutoff values with the evelopment of iabetes or cariovascular isease. Further stuy using large representative samples in various communities in Japan with a prospective stuy esign is neee. There is a growing boy of evience that aiponectin is strongly associate with metabolic isorers an visceral obesity an plays a crucial role in the pathogenesis of the metabolic synrome an cariovascular isease in the Japanese. 32,33 However, we coul not investigate the correlation between aiponectin an CCRF or the iscriminate values of VFA an WC, because the present stuy population consiste of people who unerwent an annual perioic health check-up without aiponectin examination. Further stuy focusing on this issue is neee. Conclusion For the early etection an management of CCRF in the primary screening setting, we conclue that the iscriminate criteria of gener-specific WC with relatively high sensitivity an specificity are 9 cm for men an 85 cm for women. Acknowlegments The authors are eeply grateful to the stuy participants for their cooperation. References. Sattar N, Gaw A, Scherbakova O, For I, O Reilly DS, Haffner SM, et al. Metabolic synrome with an without C-reactive protein as a preictor of coronary heart isease an iabetes in the West of Scotlan Coronary Prevention Stuy. Circulation 23; 8: Golen SH, Folsom AR, Coresh J, Sharrett AR, Szklo M, Brancati F. Risk factor grouping relate to insulin resistance an their synergistic effects on subclinical atherosclerosis: The atherosclerosis risk in communities stuy. Diabetes 22; 5: International Diabetes Feeration. The IDF consensus worlwie efinition of the metabolic synrome, (accesse May, 27). 4. Lakka HM, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, Tuomilehto J, et al. The metabolic synrome an total an cariovascular isease mortality in mile-age men. JAMA 22; 288: Isomaa B, Almgren P, Tuomi T, Forsen B, Lahti K, Nissen M, et al. Cariovascular morbiity an mortality associate with the metabolic synrome. Diabetes Care 2; 24: Gruny SM, Brewer HB Jr, Cleeman JI, Smith SC Jr, Lenfant C. American Heart Association, National Heart, Lung, an Bloo Institute: Definition of metabolic synrome: Report of the National Heart, Lung, an Bloo Institute/American Heart Association conference on scientific issues relate to efinition. Circulation 24; 9: Hu G, Qiao Q, Tuomilehto J, Balkau B, Borch-Johnsen K, Pyorala K; DECODE Stuy Group. Prevalence of the metabolic synrome an its relation to all-cause an cariovascular mortality in noniabetic European men an women. Arch Intern Me 24; 64: National Cholesterol Eucation Program (NCEP) Expert Panel on the Detection, Evaluation, an Treatment of High Bloo Cholesterol in Aults (Aults Treatment Panel III). Thir Report of the National Cholesterol Eucation program (NCEP) Expert Panel on Detection, Evaluation, an Treatment of High Bloo Cholesterol in Aults (Ault Treatment Panel III) final report. Circulation 22; 6: Ministry of Health, Labour an Welfare. National Health an Nutri-

6 886 KASHIHARA H et al. tion Survey. Daiichi-Shuppan: Tokyo, 26 (in Japanese).. Fujioka S, Matsuzawa Y, Tokunaga K, Tarui S. Contribution of intraabominal fat accumulation to the impairment of glucose an lipi metabolism in human obesity. Metabolism 987; 36: Kobayashi H, Nakamura T, Miyaoka K, Nishia M, Funahashi T, Yamashita S, et al. Visceral fat accumulation contributes to insulin resistance, small-size low-ensity lipoprotein, an progression of coronary artery isease in mile-age non-obese Japanese men. Circ J 2; 65: Kanai H, Matsuzawa Y, Kotani K, Keno Y, Kobatake T, Nagai Y, et al. Close correlation of intra-abominal fat accumulation to hypertension in obese women. Hypertension 99; 6: Fujimoto WY, Newell-Morris LL, Grote M, Bergstrom RW, Shuman WP. Visceral fat obesity an morbiity: NIDDM an atherogenic risk in Japanese-American men an women. Int J Obes 99; 5: Desprès JP, Moorjani M, Ferlan M, Tremblay A, Lupien PJ, Naeau A, et al. Aipose tissue istribution an plasma lipoprotein levels in obese women: Importance of intra-abominal fat. Arteriosclerosis 989; 9: Sparrow AD, Borkan GA, Gerzof SG, Wisnewski C, Silbert CK. Relationship of fat istribution to glucose tolerance: Results of compute tomography in male participants of the Normative Aging Stuy. Diabetes 986; 35: The Examination Committee of the Criteria for Obesity Disease in Japan, Japan Society for the stuy of Obesity. New criteria for Obesity Disease in Japan. Circ J 22; 66: Lee JS, Aoki K, Kawakubo K, Gunji A. A stuy on inices of boy fat istribution for screening of obesity. J Occup Health 995; 37: Lee JS, Kawakubo K. A useful inex highly correlate with coronary risk factors for community base obesity screening. Jpn J Public Health 999; 46: 89 2 (in Japanese). 9. Pouliot MC, Despres JP, Lemieux S, Moorjani S, Bouchar C, Tremblay A, et al. Waist circumference an abominal sagittal iameter: Best simple anthropometric inexes of abominal visceral aipose tissue accumulation an relate cariovascular risk in men an women. Am J Cariol 994; 73: Kissebah AH, Vyelingum N, Murray R, Evans DJ, Hartz AJ, Kalkhoff RK, et al. Relation of boy fat istribution to metabolic complications of obesity. J Clin Enocrinol Metab 982; 54: The Examination Committee of the Criteria for Metabolic Synrome in Japan. Definition an Criteria of the Metabolic Synrome in Japan. J Jpn Society Intern Me 25; 94: 88 2 (in Japanese). 22. Miyawaki T, Hirata M, Moriyama K, Sasaki Y, Aono H, Saito N, et al. Metabolic synrome in Japanese iagnose with visceral fat measurement by compute tomography. Proc Jpn Aca 25; 8: Oka R, Kobayashi J, Yagi K, Tanii H, Miyamoto S, Asano A, et al. Reassessment of the cutoff values of waist circumference an visceral fat area for ientifying Japanese subjects at risk for the metabolic synrome. Diabetes Res Clin Pract 28; 79: Kobayashi J, Nishimura K, Matoba M, Maekawa N, Mabuchi H. Generation an gener ifferences in the components contributing to the iagnosis of the metabolic synrome accoring to the Japanese criteria. Circ J 27; 7: Yamamoto T, Nakamura Y, Hozawa A, Okamura T, Kaowaki T, Hayakawa T, et al. Low-risk profile for cariovascular isease an mortality in Japanese. Circ J 28; 72: Yoshizumi T, Nakamura T, Yamane M, Islam AHMW, ju M, Yamasaki K, et al. Abominal fat: Stanarize technique for measurement at CT. Raiology 999; 2: Iso H, Sato S, Kitamura A, Imano H, Kiyama M, Yamagishi K, et al. Metabolic synrome an the risk of ischemic heart isease an stroke among Japanese men an women. Stroke 27; 38: Kaota A, Hozawa A, Okamura T, Kaowak T, Nakmaura K, Murakami Y, et al. NIPPON DATA Research Group. Relationship between metabolic risk factor clustering an cariovascular mortality stratifie by high bloo glucose an obesity: NIPPON DATA9, Diabetes Care 27; 3: Shiwaku K, Anuura E, Enkhmaa B, Nogi A, Kitajima K, Yamasaki M, et al. Preictive values of anthrometric measurements for multiple metabolic isorers in Asian populations. Diabetes Res Clin Pract 25; 69: Hara K, Matsushita Y, Horikoshi M, Yoshiike N, Yokoyama T, Tanaka H, et al. A proposal for the cutoff point of WC for the iagnosis of metabolic synrome in the Japanese population. Diabetes Care 26; 29: Lee JS, Kawakubo K, Mori K, Akabayashi A. Effective cutoff values of waist circumference etecting the clustering of cariovascular risk factors of metabolic synrome in Japanese men an women. Diabetes Vasc Dis Res 27; 4: Seino Y, Hirose H, Saito I, Itoh H. High-molecular-weight aiponectin is a preictor of progression to metabolic synrome: A population-base 6-year follow-up stuy in Japanese men. Metabolism 29; 58: Mori I, Ishizuka T, Morita H, Matsumoto M, Uno Y, Kajita K, et al. Comparison of biochemical ata, BP an physical activity between longevity an non-longevity istricts in Japan. Circ J 28; 72: