Atherosclerosis 191 (2007) 182–190
Determinants and definition of abdominal obesity as related to risk of diabetes, metabolic syndrome and coronary disease in Turkish men: A prospective cohort study Altan Onat a,b,∗ , H¨useyin Uyarel c , G¨ulay Hergenc¸ d , Ahmet Karabulut c , Sinan Albayrak e , G¨unay Can b a
Turkish Society of Cardiology, Istanbul University, Istanbul, Turkey Cerrahpa¸sa Medical Faculty, Istanbul University, Istanbul, Turkey c S. Ersek Cardiovascular Surgery Center, Istanbul, Turkey Biology Department, Yildiz Technical University, Istanbul University, Istanbul, Turkey e Cardiology Department of I. Baysal U. D¨ uzce Medical Faculty, D¨uzce, Turkey b
d
Received 19 December 2005; received in revised form 6 March 2006; accepted 9 March 2006 Available online 6 May 2006
Abstract We aimed to investigate determinants of abdominal obesity and its clinical impact on metabolic syndrome (MS), diabetes (DM) and coronary heart disease (CHD) in men. Methods: Prospective evaluation of 1638 male participants (aged 48.5 ± 12.3), representative of Turkey’s men who have a high prevalence of MS. For components of MS, criteria of NCEP guidelines were adopted, modified for abdominal obesity. Follow-up constituted 9650 person-years. Results: Insulin level (relative risk [RR] 1.40 for doubling), C-reactive protein (CRP) and heavy smoking (protective) were independent predictors of newly developing abdominal obesity. High triglyceride and low HDL-cholesterol were significantly associated already with waist girth quartile II, apolipoprotein B with quartile III. Waist girth significantly predicted future MS from quartile II on, independent of insulin resistance (IR) by homeostatic model assessment, whereby its hazard ratio (HR, 2.6) exceeded double that of HOMA. CRP independently predicted MS. Age-adjusted HR of waist girth (1.59) was significant in predicting DM. Age- and smoking-adjusted top waist quartile conferred significant risk for incident CHD (RR 1.71) but not for overall mortality. As judged by sensitivity and specificity rates for future CHD, DM and MS, abdominal obesity was most appropriately defined with a waist girth of ≥95 cm, and an action level 1 of ≥87 cm was proposed for MS in this population. Conclusions: Serum insulin, CRP levels and (inversely) heavy smoking are predictors for abdominal obesity in Turkish men. Atherogenic dyslipidemia and elevated blood pressure are associated significantly already with modest rises in waist girth adjusted for age and smoking. Abdominal obesity shows substantial independence of IR in the development of MS. Increasing waist girth was predictive of MS, more strongly than of DM. Risk for CHD imparted by abdominal obesity is essentially mediated by risk factors it induces. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Abdominal obesity; Coronary heart disease; Diabetes mellitus; Hypertension; Metabolic syndrome; Proinflammatory state; Cigarette smoking
Abbreviations: Apo, apolipoprotein; CHD, coronary heart disease; CRP, C-reactive protein; MS, metabolic syndrome; NCEP, National Cholesterol Education Program; IDF, International Diabetes Federation; ROC, receiver operating characteristics; RR, relative risk; S.D., standard deviation; HOMA, homeostatic model assessment ∗ Corresponding author at: Nisbetiye cad. 37/24, Etiler 34335, Istanbul, Turkey. Tel.: +90 212 351 6217; fax: +90 212 351 4235. E-mail addresses:
[email protected], alt
[email protected] (A. Onat). 0021-9150/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2006.03.012
A. Onat et al. / Atherosclerosis 191 (2007) 182–190
1. Introduction With the almost universal increase in its prevalence, obesity in general and abdominal obesity, in particular, continues to attain greater significance in the context of metabolic and cardiovascular disease risk. It has been recognized not only that different measures of obesity and the risk of different health outcomes may vary, but also that further research is needed to understand the ethnic and racial differences in waist circumference and its association with health risks [1]. For developing appropriate guidelines for clinical practice further evidence in related issues is required. Though cardiometabolic risks are largely a function of the severity of abdominal obesity, appropriate cutpoints for “action level”, is of great importance. Such cutpoints, too, are recognized to be ethnicity-specific. Among men, a cutpoint of >102 cm was adopted for Western populations which has been utilized to define abdominal obesity as a central component of the metabolic syndrome (MS) [2]. Asia-Pacific criteria for waist girth defining abdominal obesity were newly developed and applied [3] in the frame of MS criteria in Korean adults. Also newly, the International Diabetes Federation defined central obesity as waist circumference ≥94 cm for Europid men [4]. For Eastern Mediterranean populations, it was recommended that European data be used until more specific data become available. Abdominal obesity was identified in the case-control INTERHEART study as one of the main modifiable risk factors associated with myocardial infarction in 52 countries in all regions of the world [5]. Prospective studies assessing risk for cardiometabolic disorders are scarce among Turks [6] or populations of the Eastern Mediterranean and Middle East. Due to low sensitivity of the ATP III definition of abdominal obesity in Turkish men, related risk is poorly captured. A better understanding of the relationships between abdominal obesity and hypertension, atherogenic dyslipidemia, type 2 diabetes (DM) or MS might lead to a more suitable definition of abdominal obesity for this population, might improve the recognition and prevention of its health risks. The present study, based on a representative sample of Turkish middle-aged and elderly men in whom MS is highly prevalent [5], aims at following points in a cohort comprising 1638 men of the Turkish Adult Risk Factor Study, followed up for a mean 6 years: (1) to assess the main determinants of abdominal obesity, (2) to delineate the risk abdominal obesity confers in men to the risk of DM, MS and coronary heart disease (CHD), and (3) to discuss its most appropriate cutpoint regarding the stated three outcomes.
2. Methods 2.1. Population sample Participants of the nationwide survey 1997/1998 of the Turkish Adult Risk Factor Study and followed up thereafter
183
till 2004/2005, numbered 3392, among whom 1674 were men. This is a prospective survey on the prevalence of cardiac disease and risk factors in adults in Turkey carried out periodically almost biennially since 1990 in 59 communities scattered throughout all geographical regions of the country [7]. The survey involves a random sample of the Turkish adult population, representatively stratified for sex, age, geographical regions and for rural–urban distribution [7]. Since combined measurements of waist circumference, HDL-cholesterol and apolipoprotein (apo) B were first performed at the follow-up visit in 1997/1998, the latter examination formed the baseline. Participants were 28 years of age or older at baseline examination. Of the survivors, 8% were examined up to the survey 2001/2002, 14% up to 2003, the remainder having been examined lastly in the survey 2004/2005. Exclusion due to missing values for waist circumference at baseline in 14 men, and age 80 or over in 22 men limited the study sample to 1638 men. In prospective analyses regarding incident CHD and incident DM, 80 and 77 men with baseline CHD and DM, respectively, were excluded. Individuals of the cohort were visited in their addresses on the eve of the examination and were requested to give written consent for participation after having read an explanatory note, which manifested by their voluntary participation the next morning. The survey conformed to the principles embodied in the Declaration of Helsinki. Data were obtained by history of the past years via a questionnaire, physical examination of the cardiovascular system, sampling of blood and recording of a resting electrocardiogram. 2.2. Measurements of risk variables Blood pressure (BP) was measured in the sitting position on the right arm, and the mean of two recordings 3 min apart was recorded. Weight was measured without shoes in light indoor clothes using a scale. Waist circumference was measured with a tape (Roche LI95 63B 00), the subject standing and wearing only underwear, at the level midway between the lower rib margin and the iliac crest. Body mass index was calculated by the computer as weight divided by height squared (kg/m2 ). Physical activity was graded by the participant himself into four grades of increasing order with the aid of the following scheme: Grade 1: white-collar worker, sewingknitting, walking ≤1 km daily; Grade 2: repair worker, house work, walking 1–2 km daily; Grade 3: mason, carpenter, truck driver, cleaning floors and windows, walking 4 km daily; Grade 4: heavy labour, farming, regular sports activity [7]. Plasma concentrations of cholesterol, fasting triglycerides, HDL-cholesterol and glucose were determined at baseline examination by the enzymatic dry chemistry method using a Reflotron apparatus. LDL-cholesterol values were computed according to the Friedewald formula. In the final three surveys, the stated parameters, as well as apo B, insulin and CRP values were assayed in a single central laboratory. Blood samples were spun at 1000 g for 10 min and shipped within a few hours on cooled gel packs at 2–5 ◦ C to
184
A. Onat et al. / Atherosclerosis 191 (2007) 182–190
Istanbul to be stored in deep-freeze at −75 ◦ C, until analyzed at a central laboratory in the same city. Concentrations of insulin were determined by the chemiluminescent immunometric method using Roche kits and Elecsys 1010 immunautoanalyzer (Roche Diagnostics, Mannheim, Germany). Concentrations of C-reactive protein (CRP) were measured by the Behring nephelometry using an N Latex CRP mono reagent (Behring Diagnostics, Marburg, Germany), as were serum apo B values. Within run and day to day coefficients of variation (CV) for CRP were 1.3–2.9%. Within run CVs for insulin regarding normal and pathological control samples were 4.2% and 4.9%, respectively, and day to day CVs 4.9% and 3.2%, respectively. External quality control was performed with a reference laboratory in a random selection of 5–6% of participants. Data on CRP and insulin were available from the survey 2000/2001 on in 80% and 50% of participants, respectively, and tests were made again in 2004/2005. 2.3. Definitions and outcomes Never smokers, past smokers and current smokers formed the quartiles in cigarette smoking. Current smokers of >10 cigarettes daily (most of smokers), designated as heavy smokers, were analyzed separately only in the linear regression model. Anyone consuming alcohol once a week or more were considered as alcohol users. Elevated BP was defined, in agreement with NCEP guidelines [2] as being under antihypertensive treatment or having a BP ≥130 mmHg systolic and/or ≥85 mmHg diastolic. Individuals with diabetes and prediabetes were diagnosed with criteria of the American Diabetes Association [8], namely by self report or when plasma fasting glucose was ≥126 mg/dl or when 2-h postprandial glucose was >200 mg/dl. Impaired fasting glucose denoted fasting glucose values of 100–125 mg/dl. Homeostatic model assessment (HOMA) was calculated with the following formula [9]: insulin (mIU/l) × glucose (in mmol/l)/22.5. The 70 percentile of HOMA (>2.18) was considered to represent IR for this sample. Abdominal obesity was defined in this study in terms of waist circumference in agreement with this anthropometric measure emerging as the most appropriate one to reflect visceral adiposity among Turks [10]; a cutpoint was not set since it constituted an aim of the study. Results of receiver operating characteristics (ROC) related to CHD, DM and MS were utilized for this purpose to detect an optimal cutpoint. MS was defined with criteria of the NCEP [2] modified for prediabetes (fasting glucose ≥100 mg/dl [8]; these criteria were further modified for abdominal obesity, to be determined in this study. Adoption of 2.18), which halved the sample size, attenuated the RR of the top waist girth quartile to a borderline significant 2.31 (p < 0.09).
3.4. Prediction of subsequent risk for CHD and overall mortality Age- and smoking-adjusted RR of waist quartile IV in predicting 164 incident cases of CHD was significant (with an RR 1.71 [95% CI 1.03; 2.85]) versus the bottom quartile (Table 4). Addition of three further risk factors (total cholesterol, physical activity grade and alcohol usage) attenuated the significance of quartile IV to a borderline level. RRs relative to the bottom quartile were, however, elevated in the top two quartiles with 1.25 and 1.49, respectively. Age- and smoking-adjusted predictors of all-cause death in 128 out of 1638 men were analyzed in a logistic regression model. Age, and current smoking [RR 1.75 (95% CI 1.06; 2.87) opposed to never smokers] were significant predictors but waist circumference quartiles were not: RRs for quartiles II–IV were all below 1, namely, 0.80, 0.64 and 0.70, respectively. Findings remained virtually unchanged when elevated BP, alcohol use, presence of diabetes, total cholesterol and physical activity grade were added to the model. Among these, only elevated BP was an additional significant predictor, implying that risk of abdominal obesity for overall mortality was mediated mainly by elevated BP.
Table 2 Predictors of incident abdominal obesity (≥95 cm) in men Model 1
Age (years) Heavy smoking Alcohol usage Log C-reactive protein Total cholesterol (mg/dl) Physical activity grade I–IV Log fasting insulin
Model 2
n = 632 RR
95% conf. interval
n = 358 RR
95% conf. interval
1.00 0.60 1.36 1.85 1.005 0.97
0.98; 1.02 0.35; 1.019 0.79; 2.34 1.21; 2.84 1.00; 1.01 0.79; 1.20
1.00 0.38 1.87 2.49 1.003 1.17 5.30
0.98; 1.03 0.19; 0.77 0.91; 3.85 0.37; 4.53 1.00; 1.01 0.87; 1.57 2.24; 12.5
Model 1 comprised 105 cases of new abdominal obesity, 584 heavily smoking and 238 alcohol using men. Model 2 comprised 67 cases of new abdominal obesity, 184 heavily smoking and 65 alcohol using men, past and light smoking not significant.
A. Onat et al. / Atherosclerosis 191 (2007) 182–190
187
Table 3 Multi-adjusted prediction of incident diabetes and MS in nondiabetic men, by quartiles of waist grith Diabetes
Metabolic syndrome
n = 1561
n = 724
n = 936
n = 406
RR
95% CI
RR
95% CI
RR
95% CI
RR
95% CI
WC 87–94.5 cm WC 95–102 cm WC >102 cm
1.32 1.92 3.63
NS 1.004; 3.66 1.94; 6.80
1.28 1.52 2.31
NS 0.59; 3.94 0.88; 6.06
3.73 7.23 16.1
2.36; 5.89 4.38; 11.9 9.08; 28.6
2.84 5.60 14.0
1.43; 5.66 2.65; 11.8 5.96; 33.1
Age (years) Current smoking Total cholesterol (mg/dl) HOMA >2.18
1.034
1.017; 1.05
1.039 0.80 1.004 3.28
1.013; 1.065 NS 0.996; 1.011 1.46; 7.36
1.004 1.09
NS NS
0.984 0.97
NS NS
3.33
1.92; 5.77
Models comprised 109 and 59 incident cases of diabetes, 225 and 115 incident cases of MS, respectively. Models comprised also past smokers (not significant), and diabetes, physical activity grade and alcohol usage (not significant). NS: not significant (p > 0.2).
Table 4 Waist circumference in prediction of incident CHD, controlled for standard risk factors N 1558
1508
RR
95% CI
RR
95% CI
WC 87–94.5 cm WC 95–102 cm WC >102 cm
0.91 1.43 1.71
NS 0.87; 2.34 1.03; 2.85
0.88 1.25 1.49
NS 0.74; 2.12 0.86; 2.55
Age (years) Current smoking Total cholesterol (mg/dl) Physical activity grade I–IV Alcohol usage
1.079 2.07
1.063; 1.095 1.31; 3.26
1.074 2.04 1.01 0.93 0.85
1.057; 1.092 1.27; 3.28 1.006; 1.015 NS 0.71; 1.027
164 and 155 new CHD were comprised in the models, which also included past smoking (not significant). NS: not significant (p > 0.2). Table 5 Sensitivity and specificity of selected cutpoints of waist girth for future metabolic and cardiovascular diseases Incident CHD (162)
Waist at baseline >102 cm >98 cm ≥95 cm >92 cm Area under curve
Incident DM (109)
Metabolic syndrome (225)
Sensitivity
Specificity
Sensitivity
Specificity
Sensitivity
Specificity
0.33 0.50
0.79 0.65
0.40 0.56
0.80 0.65
0.24 0.39
0.94 0.88
0.61 0.69
0.52 0.40
0.70 0.74
0.53 0.43
0.50 0.66
0.82 0.70
0.592, p < 0.001
0.655, p < 0.001
0.748, p < 0.001
3.5. Selection of appropriate cutpoint
4. Discussion
Analyses of ROC curves between waist circumference and new cases of CHD, DM, or of MS revealed ratios of sensitivity and specificity for selected waist girths provided in Table 5. Optimal cutpoint for incident CHD turned out to be represented by >98 cm, for DM by ≥95 cm, for MS by ≥92 cm, and none by >102 cm of waist girth. It was deduced that a cutoff of ≥95 cm for waist circumference was most appropriate, inasmuch as it had an intermediate balance of sensitivity and specificity for incidence of all three disorders. The areas under the ROC curves for incident CHD, DM and MS were 0.59, 0.66 and 0.75, respectively (each p < 0.001).
This prospective study on a representative sample of middle-aged and elderly Turkish men demonstrated that CRP concentrations and (inversely) heavy smoking were its significant predictors independent of fasting insulin levels. Across waist circumference quartiles, serum HDLcholesterol declined, triglycerides and BP rose significantly, and risk for subsequent MS significantly increased already in quartile II. Top two quartiles predicted risk for diabetes, but HR of waist circumference (with 1.59) was considerably lower than that for MS. CHD risk also was significantly predicted by the top quartile, when controlled for age and
188
A. Onat et al. / Atherosclerosis 191 (2007) 182–190
smoking status. A cutpoint for abdominal obesity of ≥95 cm waist circumference was more appropriate in regard to risk for cardiometabolic disorders than the hitherto utilized >102 cm. Age-adjusted cardiovascular risk parameters across quartiles of increasing waist girth in Turkish men suggested that already in quartile II (87–94.5 cm) significant differences emerged in fasting insulin, triglyceride, HDL- and total cholesterol concentrations and BP. Insulin, apo B and CRP levels exhibited significant rises in quartile III reaching nearly a plateau, triglyceride and BP levels disclosing further rises in the top quartile. The Turkish population, the descendents of a crossbreed of races, is Caucasian. Yet, there appear to be ethnospecific differences compared with West Europeans, or also with South or East Asians regarding determinants and effects of abdominal obesity. As elicited from the present and related studies, these may be outlined as smoking protecting from and CRP predicting abdominal obesity. The latter exhibits substantial independence of insulin resistance (IR), and induces the emergence of atherogenic dyslipidemia and elevated BP at low grades of central obesity, thus resulting in a high prevalence of MS, in addition to, but largely independent of a genetic predisposition to low HDL-C levels. 4.1. Determinants of abdominal obesity Among the examined variables, CRP was a significant predictor of newly developing abdominal obesity contributing moderately (20% per doubling of CRP). Raised CRP concentrations have been found associated with obesity and DM cross-sectionally [12,13], and CRP levels found more elevated in healthy Indian Asians than in European whites were accounted for by greater central obesity and IR in Indian Asians [14]. However, this is the first demonstration, to our knowledge, that baseline levels of CRP predicted prospectively abdominal obesity. The prediction of abdominal obesity by an acute phase protein supports the hypothesis put forward by Yudkin et al. [15] about the antecedence of adipose-tissue generated proinflammatory cytokines and of CRP in IR, based on finding associations between CRP with a measure of IR cross-sectionally in 107 nondiabetic subjects. We found CRP levels to predict also subsequent MS, independent of several confounders. In a prospective study on a Mexican population sample, and MS being defined in the absence of abdominal obesity, CRP was found a predictor in women but not in men [16]. Cifkova et al. [17] newly reported in 1427 adults the prediction of MS as defined by ATP III by rising CRP tertiles in men but not in women. We confirmed such prediction among Turkish men and extended it to CRP concentrations as a continuous variable, when controlled for age, waist circumference, smoking, alcohol use, and physical activity grade. Adding insulin levels into the model suggested that CRP mediated interrelation between IR and abdominal obesity. Among predictors of MS was noteworthy that the magnitude of HR of waist circumference (2.6) was over two-fold
that of HOMA-IR; this underlines in this population sample, the major role of abdominal obesity independent of IR in the pathogenesis of MS [18]. Heavy cigarette smoking was an inverse determinant of abdominal obesity with an RR 0.38. This intriguing observation is not apparent but valid for this population sample, is currently under study, and is supported by the finding that smoking is inversely related to HOMA in Turkish males [19]. The inverse relationship is presumably largely due to smoking inducing diminution in appetite and raising the basal metabolic rate [20]. This, in turn, influences insulin sensitivity positively. The elicited observation stands in contrast to the findings in several other populations: for example, in multivariate analyses current smoking was positively related to waist-to-hip ratio in 6705 French men and women [21] and in 617 Canadians of diverse origin [22]; current smoking was also found a risk factor with an approximately two-fold OR for MS among Japanese men and women [23]. 4.2. Some effects of abdominal obesity The emergence already in quartile II of constituents of dyslipidemia of IR type is remarkable in this population sample. It indicates susceptibility to visceral adiposity, contributes substantially in the great majority of men to the presumably genetically-determined low HDL-cholesterol levels [24] and may lead to realize that the risk of dyslipidemic hypertension [25] exists already in men in quartile II. We had recently reported that dyslipidemic hypertension imparted a significant 1.57-fold higher HR for CVD than simple hypertension, after adjustment for sex, age, LDL-C and smoking status [25]. In predicting dyslipidemia of the IR type, our finding (not reported here) that age- and smoking-adjusted HOMA >2.18 (as opposed to 102 cm for DM and an equivocal balance regarding CHD. The proposed level 2 is higher than cutpoints introduced recently in the Asia-Pacific region (≥85 to ≥95 cm) and close to the threshold of male abdominal obesity for Europids [2]. In the context of nonconcordance between MS and IR, as assessed by HOMA, the subset of the male population exhibiting IR alone diminished substantially with the use of the herein adopted criterion for abdominal obesity as a component of MS in Turkish men [19]. Limitation of the study: findings obtained herein may not be generalized for other populations since visceral adiposity and abdominal obesity are known to be largely ethnicityspecific. However, they may be applicable for populations of the Eastern Mediterranean and the Middle East which have a fairly similar risk profile. We conclude that fasting serum insulin and CRP levels, as well as abstinence from smoking are main independent predictors in Turkish men of subsequent abdominal obesity which is most appropriately defined with a waist girth of ≥95 cm. Multi-adjusted waist quartile II significantly predicts MS and elevated BP, and age-adjusted quartiles III and IV predict DM and CHD risk, respectively. Evidence was available that abdominal obesity shows substantial independence of IR in the pathogenesis of MS and its components.
Acknowledgements We thank the Turkish Society of Cardiology and the various pharmaceutical companies that have supported financially the Turkish Adult Risk Factor Survey over the years. The partial logistic support of the Turkish Ministry of Health is acknowledged. We appreciate the dedicated works of Y. Do˘gan, MD, ˙I. Sarı, MD, S. T¨urkmen, MD, M. Yazıcı, MD, ¨ and Mr. M. Ozmay, the coworkers in the survey teams.
References [1] Wang Y, Rimm EB, Stampfer MJ, Willett WC, Hu FB. Comparison of abdominal obesity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr 2005;81:555–63. [2] 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;285:2486–97. [3] Park SH, Lee WY, Rhee EJ, et al. The relative risks of metabolic syndrome according to the degree of insulin resistance in apparently healthy Korean adults. Clin Sci 2005;108:553–9. [4] Alberti KG, Zimmet P, Shaw J, IDF Epidemiology Task Force Consensus Group. The metabolic syndrome – a new worldwide definition. Lancet 2005;366:1059–62. [5] Yusuf S, Hawken S, Ounpuu S, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 2004;364: 937–52. ¨ et al. Metabolic syndrome: major impact [6] Onat A, Ceyhan K, Bas¸ar O, on coronary risk in a population with low cholesterol levels – a prospec-
190
[7]
[8] [9]
[10]
[11] [12] [13] [14]
[15]
[16]
[17]
[18]
A. Onat et al. / Atherosclerosis 191 (2007) 182–190 tive and cross-sectional evaluation. Atherosclerosis 2002;165:285– 92. ¨ Onat A, Avcı GS¸, S¸enocak M, Ornek E, G¨oz¨ukara Y. Plasma lipids and their interrelation in Turkish adults. J Epidem Commun Health 1992;46:470–6. American Diabetes Association. Clinical practice recommendations 1997. Diabetes Care 1997;20(Suppl):S1–S70. Matthews DR, Hosker JP, Rudenski AS, et al. Homeostatic model assessment: insulin resistance and b-cell function from fasting glucose and insulin concentration in man. Diabetologia 1985;28:412–9. Onat A, Avcı GS¸, Barlan MM, et al. Measures of abdominal obesity assessed for visceral adiposity and relation to coronary risk. Int J Obes 2004;28:1018–25. Rose GA, Blackburn H, Gillum RF, Prineas RJ. Cardiovascular survey methods. 2nd ed. Geneva: WHO; 1982. pp. 124–127. Ford ES. Body mass index, diabetes, and C-reactive protein among U.S. adults. Diabetes Care 1999;22:1971–7. Onat A, Sansoy V, Yıldırım B, et al. C-reactive protein and coronary heart disease in Western Turkey. Am J Cardiol 2001;88:601–7. Chambers JC, Eda S, Bassett P, et al. C-reactive protein, insulin resistance, central obesity, and coronary heart disease risk in Indian Asians from the United Kingdom compared with European whites. Circulation 2001;104:145–50. Yudkin JS, Stehouwer CDA, Emeis JJ, Coppack SW. C-reactive protein in healthy subjects: associations with obesity, insulin resistance and endothelial dysfunction. A potential role for cytokines originating from adipose tissue? Arterioscler Thromb Vasc Biol 1999;19: 972–8. Than TS, Sattar N, Williams K, et al. Prospective study of Creactive protein in relation to the development of diabetes and metabolic syndrome in the Mexico City Diabetes Study. Diabetes Care 2002;25:2016–21. Cifkova R, Frohlich J, Skodova Z, et al. C-reactive protein and the risk of developing the metabolic syndrome. A population study. Eur Heart J 2005;26(Suppl):445 (abstr.). Grundy SM, Brewer HB, Cleeman JI, Smith SC, Lenfant C. Definition of metabolic syndrome: report of the National Heart, Lung, and Blood
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
Institute/American Heart Association conference on scientific issues related to definition. Circulation 2004;109:433–8. Onat A, Hergenc¸ G, T¨urkmen S, Yazıcı M, Sarı ˙I, Can G. Discordance between insulin resistance and metabolic syndrome: features and associated cardiovascular risk in adults with normal glucose regulation. Metabolism 2006;55:445–52. Filozof C, Fernandez Pinilla MC, Fernandez-Cruz A. Smoking cessation and weight gain. Obes Rev 2004;5:95–103. Czernichow S, Bertrais S, Preziosi P, et al. Indicators of abdominal adiposity in middle-aged participants of the SU.VI.MAX study: relationships with educational level, smoking status and physical inactivity. Diabetes Metab 2004;30:110–1. Merchant AT, Anand SS, Vuksan V, et al. Protein intake is inversely associated with abdominal obesity in a multi-ethnic population. J Nutr 2005;135:1196–201. Ishizaka N, Ishizaka Y, Toda E-I, Nagai R, Yamakado M. Association between serum uric acid, metabolic syndrome, and carotid atherosclerosis in Japanese individuals. Arterioscler Thromb Vasc Biol 2005;25:1038–44. Mahley RW, Palao˘glu E, Atak Z, et al. Turkish Heart Study: lipids, lipoproteins, and apolipoproteins. J Lipid Res 1995;36:839– 59. Onat A, Hergenc¸ G, Sarı I, et al. Dyslipidemic hypertension: distinctive features and cardiovascular risk in a prospective population-based study. Am J Hypertens 2005;18:409–16. Hayashi T, Boyko EJ, Leonetti DL, et al. Visceral adiposity and the prevalence of hypertension in Japanese Americans. Circulation 2003;108:1718–23. Rexrode KM, Buring JE, Manson JE. Abdominal total adiposity and risk of coronary heart disease in men. Int J Obes 2001;25:1047– 56. Dagenais GR, Yi Q, Mann JFE, et al. Prognostic impact of body weight and abdominal obesity in women and men with cardiovascular disease. Am Heart J 2005;149:54–60. Bigaard J, Tjonneland A, Thomsen BL, et al. Waist circumference, BMI, smoking, and mortality in middle-aged men and women. Obes Res 2003;11:895–903.
