Relationships between the Visceral Fat Area on CT and Coronary Risk Factor Markers

Transcription

1 ORIGINAL ARTICLE Relationships between the Visceral Fat Area on CT and Coronary Risk Factor Markers Yutaka Shiina and Yasuhiko Homma Abstract Objective The visceral fat area (VFA) was measured, and the relationships between the VFA and the body mass index (BMI), waist circumference, blood pressure, and indices of lipid and sugar metabolism were evaluated. Methods The subjects included 607 consecutive patients who underwent VFA examinations using computed tomography (CT) scans. In addition to the routine examination parameters, the levels of adiponectin and homeostasis model assessment as an index of insulin resistance () were measured in all subjects, and the levels of malondialdehyde-modified low-density lipoprotein (MDA-LDL), remnant-like particles (RLP), lipoprotein (a) (Lp(a)), apolipoprotein (Apo) AI, ApoB and ApoE were measured in 270 subjects. Results In both men and women, the VFA showed significant positive correlations with the age, BMI, waist circumference, subcutaneous fat area, visceral fat area/subcutaneous fat area (v/s) ratio, systolic blood pressure, diastolic blood pressure, the fasting blood sugar (FBS), the hemoglobin A1c (HbA1c), high-density lipoprotein cholesterol (HDLC), triglyceride (TG), uric acid, and ApoB and the ApoB/LDLC ratio and significant negative correlations with the levels of HDLC and adiponectin. The levels of the total cholesterol (TC), low-density lipoprotein cholesterol (LDLC), non-hdlc, MDA-LDL and Lp(a) and the ApoB/ ApoAI ratio were not correlated with the VFA in either men or women. The RLP exhibited a significant positive correlation with the VFA in women. Conclusion The VFA exhibited high positive correlations with the waist circumference, blood pressure and TG level and a negative correlation with the HDLC level, regardless of gender, supporting the validity of the present diagnostic method for evaluating metabolic syndrome (MS). Although the LDLC level is not included in the diagnostic criteria for MS, the positive correlations between the VFA and the ApoB level and ApoB/ LDLC ratio observed in both men and women indicate qualitative abnormalities of lipoproteins, such as an increase in the amount of small dense LDL. Measuring the levels of apolipoproteins in addition to lipoproteins during health screening is therefore useful for evaluating of atherogenicity. Key words: visceral fat area, adiponectin, small dense LDL, MDA-LDL () () Introduction Metabolic syndrome (MS) is believed to underlie atherosclerotic diseases. It has also been reported that the concurrence of insulin resistance, abnormal glucose tolerance, abnormal lipoprotein levels and hypertension is essential for the development of MS and that the accumulation of visceral fat underlies all of these conditions. However, the vague basis of the diagnostic criteria (1-3) and reference values for MS has been raised as a problem (4). In this study, we evaluated the relationships between the visceral fat area and the body mass index (BMI), waist circumference, blood pressure, and indices of lipid and sugar metabolism in patients who voluntarily underwent measurement of the visceral fat area using computed tomography (CT) scans during a complete health screening (ningen dock). Department of Clinical Health Science, Tokai University School of Medicine, Japan Received for publication October 24, 2012; Accepted for publication March 28, 2013 Correspondence to Dr. Yutaka Shiina, shiina@is.icc.u-tokai.ac.jp 1775

2 Table 1. Clinical Characteristics of the Subjects Men Women n=607 Age(y/o) Visceral Fat Area (cm 2 ) BMI(kg/m 2 ) Waist Circumference(cm) Subcutaneous Fat Area(cm 2 ) BPs(mmHg) BPd(mmHg) FBS(mg/dL) HbA1c(%)(NGSP) TC(mg/dL) LDLC(mg/dL) HDLC(mg/dL) TG(mg/dL) non-hdlc(mg/dl) Uric acid(mg/dl) Hypertension(%) Diabetes mellitus(%) Dyslipidemia(%) Smoking(%) Drinking(%) n=270 MDA-LDL(U/L) RLP(mg/dL) Lp(a)(mg/dL) ApoAI(mg/dL) ApoB(mg/dL) ApoE(mg/dL) ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± * 0.012* * * * 0.019* Means±SE. *s<0.05 were considered statistically significant. BMI: body mass index, : Visceral Fat Area/ Subcutaneous Fat Area ratio, BPs: systolic blood pressure, BPd: diastolic blood pressure, FBS: fasting blood sugar, HbA1c (NGSP): hemoglobin A1c (National Glycohemoglobin Standardization Program), TC: total choresterol, LDLC: low density lipoprotein choresterol, HDLC: high density lipoprotein choresterol, TG: triglyceride, non-hdlc: TC-HDLC, :homeostasis model assessment as an index of insulin resistance, MDA-LDL: malondialdehyde-modified low density lipoprotein, RLP: remnant like particle, Lp (a): lipoprotein (a), ApoAI: ApolipoproteinAI, ApoB: ApolipoproteinB, ApoE: ApolipoproteinE Subjects Materials and Methods The subjects included 607 consecutive individuals (385 men and 222 women) who underwent visceral fat examinations using CT scans for the first time on complete health screenings (ningen dock) performed during the three years and nine months between February, 2008 and October, 2011 at Tokai University Hospital, Health Screening Center. Methods Blood tests were performed by collecting blood samples after a fast of 12 hours or longer. In addition to routine examinations of health screening, the levels of adiponectin and homeostasis model assessment as an index of insulin resistance () were measured in all subjects. In 270 (162 men and 108 women) of these subjects, the levels of malondialdehyde-modified low-density lipoprotein (MDA- LDL), remnant-like particles (RLP), lipoprotein (a) (Lp(a)), apolipoprotein AI (ApoAI), apolipoprotein B (ApoB) and apolipoprotein E (ApoE) were also measured. The adiponectin level was measured using latex immunoturbidimetry (5), and the was calculated as the fasting blood insulin level (μu/ml) fasting blood sugar level (mg/ dl)/405. The waist circumference was measured using the method proposed by the Japan Society for the Study of Obesity (6). Abdominal CT scans were performed at the end of the expiratory phase using a Siemens CT system (SO- MATOM Emotion Duo, Siemens, Munich, Germany). While the slice position of scanning was at the umbilical level, modified measurement levels were employed in subjects with a clearly low umbilical body type. The visceral and subcutaneous fat areas were calculated using FAT Scan Ver.4.0 (7), a PC software program designed by the N2 System Co., Ltd. (Hyogo, Japan) to measure the amount of visceral fat. The MDA-LDL level was measured using ELISA (8), the RLP level was measured using an immunoabsorption method (9), the Lp(a) level was measured using latex aggregation turbidimetry and the levels of ApoAI, ApoB and ApoE were measured using a turbidimetric immunoassay (10). The non-high density lipoprotein cholesterol (non-hdlc) level was calculated using the formula TC-HDLC. The presence or absence of hypertension, diabetes, dyslipidemia and smoking and drinking habits were evaluated using medical interview forms. Statistical analyses Fundamental statistics were calculated, and the unpaired t- test was employed. Comparisons of the visceral fat area were made by dividing the subjects into four groups based on quartiles and performing the Kruskal-Wallis test. Since various factors were interrelated, a multiple regression analysis was also performed. The level of significance was p <0.05. This study was carried out with approval by the Institutional Review Board of the Tokai University School of Medicine. Informed consent was obtained from the persons undergoing health screening who were prospective subjects of this study after explaining in writing that the data would be processed in a manner that prohibits the identification of individuals, that the data would not be used for purposes other than scientific research and that there would be no disadvantage should the person refuse to cooperate in the study. 1776

3 Table 2. Relationships between the Visceral Fat Area and Various Screening Parameters in Men Visceral Fat Area(cm 2 ) n= Age(y/o) 53.1± ± ± ± * BMI(kg/m 2 ) 21.6± ± ± ±0.58 Waist Circumference(cm) 78.4± ± ± ±1.35 Subcutaneous Fat Area(cm 2 ) 102.2± ± ± ± ± ± ± ±0.05 BPs(mmHg) 121.8± ± ± ±2.6 BPd(mmHg) 78.0± ± ± ±1.60 FBS(mg/dL) 100.7± ± ± ±6.0 HbA1c(%)(NGSP) 5.42± ± ± ±0.23 TC(mg/dL) 204.6± ± ± ± LDLC(mg/dL) 124.3± ± ± ± HDLC(mg/dL) 66.2± ± ± ±2.88 TG(mg/dL) 109.2± ± ± ±17.6 non-hdlc(mg/dl) 138.4± ± ± ± Uric Acid(mg/dL) 5.91± ± ± ± * 10.4± ± ± ± ± ± ± ±1.56 MDA-LDL(U/L) 111.4± ± ± ± n= RLP(mg/dL) 4.65± ± ± ± Lp(a)(mg/dL) 17.6± ± ± ± ApoAI(mg/dL) 146.5± ± ± ± ApoB(mg/dL) 93.5± ± ± ± * ApoE(mg/dL) 4.08± ± ± ± ± ± ± ± ApoB/LDLC ratio 0.76± ± ± ±0.01 Means±SE. *s<0.05 were considered statistically significant. Kruskal Wallis test Results Table 1 shows the clinical characteristics of all subjects separately for men and women. Among the men, the mean age was lower, the visceral fat area, BMI, waist circumference, visceral fat area/subcutaneous fat area (v/s) ratio, systolic blood pressure and diastolic blood pressure were higher and the subcutaneous fat area was lower than that observed in the women. Concerning the blood test results, the fasting blood sugar (FBS), triglyceride (TG), uric acid and HOMA- IR levels were significantly higher and the total cholesterol (TC), low-density lipoprotein cholesterol (LDLC), highdensity lipoprotein cholesterol (HDLC) and adiponectin levels were significantly lower in the men than in the women; however, no differences were noted in the hemoglobin A1c (HbA1c) (NGSP: National Glycohemoglobin Standardization Program) or non-hdlc levels. Regular smoking and drinking were observed more frequently in the men than in the women; however, no differences were noted in the prevalance of hypertension, diabetes or dyslipidemia between the men and women. Among the 270 subjects (162 men and 108 women) who underwent measurement of the levels of MDA-LDL, RLP, Lp(a), ApoAI, ApoB and ApoE as optional items, the RLP level, the and the ApoB/LDLC ratio were significantly higher and the Lp(a) and ApoAI levels were significantly lower in the men than in the women; however, no differences were noted in the MDA-LDL, ApoB and ApoE levels. Table 2 compares the data by classifying the men into those with a visceral fat area of 82.2 cm 2 or below (97 subjects), cm 2 (96 subjects), cm 2 (96 subjects) or cm 2 or greater (96 subjects). The visceral fat area exhibited significant positive correlations with the age, BMI, waist circumference, subcutaneous fat area, v/s ratio, systolic blood pressure, diastolic blood pressure, the levels of FBS, HbA1c, TG, uric acid, and ApoB and the ApoB/LDLC ratio and negative correlations with the levels of HDLC and adiponectin. The visceral fat area demonstrated no significant differences based on the TC, LDLC, 1777

4 Table 3. Relationships between the Visceral Fat Area and Various Screening Parameters in Women Visceral Fat Area (cm 2 ) n= Age (y/o) 57.4± ± ± ± * BMI (kg/m 2 ) 20.4± ± ± ±0.62 Waist Circumference (cm) 71.4± ± ± ±1.69 Subcutaneous Fat Area(cm 2 ) 119.1± ± ± ± ± ± ± ±0.04 BPs(mmHg) 118.6± ± ± ±3.1 BPd(mmHg) 71.7± ± ± ± * FBS(mg/dL) 94.0± ± ± ±3.0 HbA1c(%)(NGSP) 5.41± ± ± ±0.12 TC(mg/dL) 224.8± ± ± ± LDLC(mg/dL) 129.9± ± ± ± HDLC(mg/dL) 87.0± ± ± ±2.93 TG(mg/dL) 69.8± ± ± ±9.0 non-hdlc(mg/dl) 137.7± ± ± ± Uric Acid(mg/dL) 4.36± ± ± ± ± ± ± ± ± ± ± ±0.42 MDA-LDL(U/L) 100.1± ± ± ± n= RLP(mg/dL) 3.57± ± ± ± * Lp(a)(mg/dL) 14.2± ± ± ± ApoAI(mg/dL) 170.4± ± ± ± ApoB(mg/dL) 89.5± ± ± ± * ApoE(mg/dL) 4.50± ± ± ± ± ± ± ± ApoB/LDLC ratio 0.69± ± ± ±0.01 Means±SE. *s<0.05 were considered statistically significant. Krusal Wallis test non-hdlc, MDA-LDL, RLP, Lp(a), ApoAI and ApoE levels or the. The women were similarly divided into those with a visceral fat area of 44.3 cm 2 or below (56 subjects), cm 2 (55 subjects), cm 2 (56 subjects) or 96.5 cm 2 or greater (55 subjects) and the data were compared (Table 3). The visceral fat area exhibited significant positive correlations with the age, BMI, waist circumference, subcutaneous fat area,, systolic blood pressure, diastolic blood pressure, the levels of FBS, HbA1c, TG, uric acid,, RLP and ApoB and the ApoB/LDLC ratio and significant negative correlations with the HDLC and adiponectin levels; however, no correlations were observed with the TC, LDLC, non-hdlc, MDA- LDL, Lp(a) and ApoE levels or the. A multiple regression analysis was carried out by examining the correlations among variables and restricting those that were applied. Since various lipid variables exhibited strong multiple correlations with the apolipoprotein variables, only apolipoprotein variables were applied instead of all lipid variables. Since the BMI, abdominal circumference and subcutaneous fat area were also closely correlated with one another, only the subcutaneous fat area was applied. Only the systolic pressure was applied as a blood pressure variable, and only the HbA1c level was applied as a variable of blood sugar. Table 4 shows the results of the calculations. The subcutaneous fat area and the exhibited strong correlations with the visceral fat area in both the men and women. The visceral fat area was also significantly correlated with the age and the uric acid and adiponectin levels in the men and with the and ApoAI levels and the presence of diabetes in the women. Discussion In a separate evaluation of the men and women, the visceral fat area exhibited significant positive correlations with the age, BMI, waist circumference, subcutaneous fat area, v/ s ratio, systolic blood pressure, diastolic blood pressure, the levels of FBS, HbA1c, TG, uric acid, and ApoB and the ApoB/LDLC ratio and negative correlations with the 1778

5 Table 4. s of the Relationships between Various Health Screening Parameters and the Visceral Fat Area in Men and Women Age Subcutaneous Fat Area BPs HbA1c(NGSP) Uric Acid Adiponectin MDA-LDL RLP Lp(a) ApoAI ApoB ApoE Hypertension Diabetes mellitus Dyslipidemia Smoking Drinking Men * 0.026* R 2 (men:0.864, women:0.913) Linear Multiple Regression Analysis *s<0.05 were considered statistically significant Women * * * HDLC and adiponectin levels in both the men and women. In Japanese individuals, visceral fat accumulation is often observed at milder obesity levels than in Western populations and at a BMI within the reference range (11); however, both the BMI and waist circumference are considered to be excellent indices that reflect visceral fat accumulation. The blood adiponectin concentration was significantly lower in the men than in the women, although it exhibited a strong negative correlation with the visceral fat area in both the men and women, in agreement with the findings of previous reports (5, 12). The incidence of MS is thought to increase when the blood adiponectin concentration decreases to a level below 4.0 μg/ml (13). In this study, however, the mean adiponectin level remained relatively high at 6.87± 0.56 μg/ml in the men with a visceral fat area of cm 2 or greater and at 9.36±0.80 μg/ml in the women with a visceral fat area of 96.5 cm 2 or greater. The blood pressure increased in association with the visceral fat area, similar to the results of a study of Japanese subjects conducted by Ohnishi et al. (14). The TG level exhibited a strong positive correlation and the HDLC level exhibited a negative correlation with the visceral fat area in both genders. The RLP level demonstrated a significant positive correlation with the visceral fat area (p=0.006) in the women and a positive correlation, although not significant (p=0.054), in the men. High levels of RLP and ApoB are characteristics of dyslipidemia in patients with metabolic syndrome (15, 16). In this study, triglyceride-rich lipoprotein remnants appeared more often, and the ApoB/LDLC ratio was elevated in association with an increase in the visceral fat area in both the men and women, suggesting an increase in the amount of small dense LDL. Although the TC and LDLC levels alone were not directly related to the visceral fat area, measuring the levels of apolipoproteins, which allows for the evaluation of qualitative changes in lipoproteins, is useful for health screening. While a high uric acid level is often related to drinking, the uric acid level was significantly correlated with the visceral fat area in both men and women, and this correlation was stronger in women, who are less frequently regular drinkers. One report has indicated that hyperuricemia is related to increased uric acid production in patients with visceral fattype obesity and reduced uric acid excretion in patients with subcutaneous fat-type obesity (17), and the results of this study also suggest a relationship between obesity and the uric acid level. A high blood RLP level induces atherosclerotic diseases (18) and was also identified to be an independent risk factor of atherosclerosis in women in the Framingham Study (19). In the present study, the RLP level was higher in men than in women, although it was more strongly correlated with the visceral fat area in women. The levels of non-hdlc, MDA-LDL and Lp(a), which are considered to be atherosclerogenic factors, exhibited no direct correlations with the visceral fat area in men or women. However, the level of ApoB was increased relative to the LDLC concentration in both men and women, which requires attention since there is a possibility of the appearance of atherosclerogenic small dense LDL. In particular, with respect to visceral fat accumulation, conducting a qualitative evaluation is necessary, even when the cholesterol level is low. MDA-LDL is a substance that results from the oxidation of unsaturated fatty acids of LDL and induces atherosclerosis by activating the transformation of macrophages to foam cells (20). In our study, the visceral fat area exhibited no significant correlations with the MDA-LDL level in either men or women; however, if the level of small dense LDL is increased, these particles are likely to be denatured by oxidation, possibly causing an increase in the production of oxidized LDL (21). Lp(a) is an atherosclerogenic lipoprotein that has been confirmed to be deposited in atherosclerotic foci and is involved in thrombus formation. The concentration of Lp(a) is genetically determined (22), exhibits gender differences, being high in postmenopausal women, and its effects may diminish with age (23). While no relationships between the Lp(a) level and the visceral fat area were noted in this study, it is necessary to regard a high Lp (a) level as an independent risk factor of atherosclerogenesis along with a high MDA-LDL level (24). According to the results of a multiple regression analysis, the visceral fat area exhibits strong correlations with the subcutaneous fat area and, although, among non-ct markers, the waist circumference and BMI are good indices for estimating the visceral fat area. The severity of MS is suggested to be in- 1779

6 creased in association with an older age, a high uric acid level and a low adiponectin level in men and with diabetes, a high level and a low ApoAI level in women, in addition to the above indices. Conclusion The visceral fat area measured on CT exhibits positive correlations with the waist circumference, blood pressure and the TG level and a negative correlation with the HDLC level in both men and women, supporting the validity of the present diagnostic method for assessing MS. While the LDLC level is not included in the diagnostic criteria for MS, qualitative abnormalities in the levels of lipoproteins, such as an increase in the amount of small dense LDL, are suggested by the positive correlations between the visceral fat area and the ApoB level and ApoB/LDLC ratio in both men and women. Therefore, measuring the levels of apolipoproteins in addition to lipoproteins is useful for evaluating atherosclerogenic factors during health screening. The authors state that they have no Conflict of Interest (COI). Acknowledgement The authors express sincere gratitude to Dr. Takeo Shibata, Ph.D., Course of Basic Medical Science and Molecular Medicine, Department of Medical Informatics, Tokai University School of Medicine for providing instructions regarding the statistical analysis. References 1. 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 285: , Part I: Diagnosis and Classification of Diabetes Mellitus. In: Definition, Diagnosis and Classification of Diabetes Mellitus and Its Complications. World Health Organization, Geneva, Alberti KGMM, Zimmet P, Shaw J, et al. The metabolic syndrome: a new worldwide definition. Lancet 366: , Alberti KG, Eckel RH, Grundy SM, et al. Harmonizing the metabolic syndrome: a joint interim statement of International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 120: , Arita Y, Kihara S, Ouchi N, et al. Paradoxical decresed of an adipo-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun 257: 79-83, The Examination Committee of Criteria for Obesity Disease in Japan, Japan Society for the Study of Obesity. New criteria for obesity disease in Japan. Circulation J 66: , Yoshizumi T, Nakamura T, Yamane M, et al. Abdominal fat: standardized technique for measurement at CT. Radiology 211: , Kotani K, Maekawa M, Kanno T, et al. Distribution of immunoreactive malondialdehyde-modified low-density lipoprotein in human serum. Biochim Biophys Acta 1215: , Nakajima K, Saito T, Tamura A, et al. Choresterol in remnant-like lipoproteins in human serum using monoclonal antiapob-100 and antiapoa-i immunoaffinity mixed gel. Clin Chim Acta 223: 53-71, Noma A, Hara Y, Goto Y. Quantification of serum apolipoprotein A-I, A-II, B, C-II, C-III, and E in healthy Japanese by the turbidimetric immunoassay: reference values, and age-and sex-related differences. Clin Chim Acta 199: , Miyawaki T, Abe M, Yahata K, et al. Contribution of visceral fat accumulation to the risk factors for atherosclerosis in non-obese Japanese. Intern Med 43: , Matsuzawa Y, Funahashi T, Kihara S, et al. Adiponectin and Metabolic Syndrome. Arterioscler Thromb Vasc Biol 24: 29-33, Ryo M, Nakamura T, Kihara S, et al. Adiponectin as a biomarker of the metabolic syndrome. Circulation J 68: , Ohnishi H, Saitoh S, Akasaka H, et al. Incidence of hypertension in individuals with abdominal obesity in a rural Japanese population: The Tanno and Sobetsu study. Hypertens Res 31: , Lamarche B, Tchernof A, Mauriege P, et al. Fasting insulin and Apolipoprotein B levels and low-density lipoprotein particle size as risk factors for ischemic heart disease. JAMA 279: , Kinoshita M, Ohnishi H, Maeda T, et al. Increased serum apolipoprotein B48 concentration in patients with metabolic syndrome. J Atheroscler Thromb 16: , Matsuura F, Yamashita S, Nakamura T, et al. Effect of visceral fat accumulation on uric acid metabolism in male obese subjects: visceral fat obesity is linked more closely to overproduction of uric acid than subcutaneous fat obesity. Metabolism 47: , Twickler TB, Dallinga-Thie GM, Cohn JS, Chapman MJ. Elevated remnant-like particle cholesterol concentration: a characteristic feature of the atherogenic lipoprotein phenotype. Circulation 109: , McNamara JR, Shah PK, Nakajima K, et al. Remnant-like particle (RLP) cholesterol is an independent cardiovascular disease risk factor in women: result from the Framingham Heart Study. Atherosclerosis 154: , Holvoet P, Perez G, Zhao Z, et al. Malondialdehyde-modified low density lipoproteins in patients with atherosclerotic disease. J Clin Invest 95: , de Gtaaf J, Hak-Lemmers HL, Hectors MP, et al. Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein subfraction in healthy subjects. Arterioscler Thromb 11: , Boerwinkle E. Genetics of plasma lipoprotein(a) conentrations. Curr Opinion Lipidol 3: , Simons L, Friedlander Y, Simons J, et al. Lipoprotein(a) is not associated with coronary heart disease in elderly: cross-sectional data from the Dubbo study. Atherosclesosis 99: 87-95, Homma Y. Predictors of atherosclerosis. J Atheroscler Thromb 11: , The Japanese Society of Internal Medicine