Relationship between visceral fat,steroid hormones and insulin sensitivity in premenopausal obese women

Abstract: Objectives. The relationships between visceral fat distribution, steroid hormones and peripheral insulin sensitivity were studied. Setting. All subjects were hospitalized in the Institute of Internal Medicine of the University of Verona, Italy. Subjects. Nineteen fertile obese women were studied with ages ranging from 18 to 53 years and body mass indexes ranging from 27.3 to 48.4. Intervention. Body fat distribution was evaluated by waist-to-hip circumference ratio and by computed tomography. The insulin tolerance test was used to evaluate peripheral insulin sensitivity. Glucose, insulin and C-peptide were measured in fasting conditions and during glucose load; total and free plasma testosterone and urinary cortisol excretion were also determined. Results. Significant correlations emerged between visceral adipose tissue and fasting glucose, insulin, and C-peptide. but not between visceral adipose tissue and total testosterone, free testosterone or urinary cortisol excretion. A negative correlation emerged between visceral adipose tissue and insulin sensitivity (r = ?0.70; P < 0.01). No significant correlations were found between insulin sensitivity and age, body weight, body mass index, total adipose tissue, subcutaneous adipose tissue or waist-to-hip ratio. Total testosterone correlated with body weight, subcutaneous adipose tissue and total adipose tissue. Free testosterone and urinary cortisol excretion correlated positively with body weight, and negatively with age. No correlation was found between insulin sensitivity and total testosterone, free testosterone or urinary cortisol excretion. The correlation between visceral adipose tissue and insulin sensitivity remained significant even after adjusting for both age and the body mass index. Conclusions. Our study shows that visceral fat is more closely associated with aberrations of insulin sensitivity than with obesity itself. Total testosterone, free testosterone and urinary cortisol excretion in our subjects do not seem to be associated with such aberrations.     

Insulin sensitivity and plasma homocysteine concentrations in non-diabetic obese and normal weight subjects

OBJECTIVE: To determine whether plasma homocysteine (tHcy) levels were related to insulin resistance and obesity in subjects without diabetes or vascular disease. RESEARCH DESIGN AND METHODS: We studied correlates of plasma tHcy in 26 subjects covering a wide spectrum of obesity and insulin sensitivity (S(I)). The measurement of in vivo insulin sensitivity was performed using the minimal model analysis of the frequently sampled intravenous glucose tolerance test (FSIVGTT). RESULTS: There was no relationship between tHcy and body mass index. There was a significant relationship between plasma tHcy and S(I) (r=0.53, P=0.006), demonstrating that the more insulin sensitive subjects had higher levels of tHcy. On log transformation of the plasma insulin values, log insulin correlated negatively with plasma tHcy (r=-0.47; P=0.02). None of the subjects were deficient in vitamin B(12) and folate. Plasma vitamin B(12) was significantly related to plasma tHcy (r=-0.44, P=0.017), although we found no significant relationship between plasma folate and tHcy (r=-0.21, P=0.27). S(I) correlated significantly with vitamin B(12) (r=0.4, P=0.045) whereas, we found no significant relationship between S(I) and plasma folate (r=0.27, P=0.2). On multiple linear regression using tHcy as the dependent variable, S(I) and vitamin B(12) remained significant predictors of plasma tHcy, whereas, age and plasma folate were not predictors of tHcy. CONCLUSIONS: We conclude that in vitamin replete lean and obese individuals, insulin sensitivity correlates significantly with plasma tHcy. This relationship may need to be considered when evaluating the role of plasma homocysteine as a risk factor in patients with obesity and type 2 diabetes.     

Early carotid atherosclerosis in normotensive severe obese premenopausal women with low DHEA(S)

The aim of this study was to investigate the direct involvement of hyperinsulinaemia, DHEA and DHEA-S [DHEA(S)] in severe obesity in early carotid atherosclerosis, measured as intima-media thickness (IMT). Seventeen normotensive premenopausal women with very high BMI (43.5 +/- 1.6 kg/m2) were recruited for the study. Six women were also evaluated 12 months after laparoscopic adjustable silicone gastric banding (LASGB). Dietary intake, fasting plasma lipid profile, glycemic and insulinemic response to the OGTT, adrenal secretion, at baseline and after ACTH stimulation test, were measured. IMT, common carotid diameter (CD) and left ventricular mass index (LVMi) were measured by B-mode echotomography. All obese subjects showed higher fasting and stimulated insulin levels, but lower DHEA(S) levels than controls, showing a negative correlation between both fasting and stimulated insulin and DHEA(S), either at baseline or after ACTH testing. IMT was higher (p < 0.05) than controls, with a positive correlation with stimulated insulin (p < 0.05) and a strong negative correlation with DHEA(S) (p < 0.001). In a multiple linear regression analysis, insulin response to OGTT maintained an association with DHEA(S) independent of fasting insulin, while DHEA maintained the association with IMT independent of stimulated insulin (p < 0.0001). In the six patients evaluated 12 months after LASGB, fasting insulin levels decreased, while DHEA(S) levels increased (p < 0.05). In conclusion, an early cardiovascular involvement was detected in this group of severe obese with hyperinsulinaemia and low DHEA(S), even in the absence of other well known CVD risk factors.     

Relative impact of low-density lipoprotein-cholesterol concentration and insulin resistance on carotid wall thickening in nondiabetic, normotensive volunteers

The relative effect of an increase in low-density lipoprotein-cholesterol (LDL-C) concentration, as compared with insulin resistance and its manifestations, on intimal medial thickening (IMT) of the common carotid artery was defined in 72 healthy men and women. Insulin-mediated glucose disposal was quantified by the insulin suppression tests, in which the height of the steady-state plasma glucose (SSPG) concentration during the last 30 minutes of a 180-minute infusion of octreotide, insulin, and glucose provides an estimate of insulin resistance. IMT was determined by high-resolution B-mode ultrasonography. Univariate analyses defined statistically significant correlation coefficients between IMT and LDL-C concentration (r =.25, P <.05), SSPG concentration (r =.32, P <.01), triglycerides (TG) (r =.25, P <.05), and high-density lipoprotein-cholesterol (HDL-C) (r = -.28, P <.05) concentrations (changes associated with insulin resistance) and ratio of waist-to-hip girth (r =.29, P <.05). When forward step-wise linear regression analysis was used, concentrations of SSPG, LDL-C and HDL-C all emerged as independent predictors of IMT (P <.05). Furthermore, the magnitude of their relationship to IMT values was comparable. These results provide evidence that insulin resistance is as significant a predictor of degree of atherogenesis (estimated by IMT) of the common carotid artery as a high LDL-C concentration.     

Unchanged asymmetric dimethylarginine levels in non-diabetic, premenopausal obese women who have common risk factors for cardiovascular disease

This study was performed to test whether plasma asymmetric dimethylarginine (ADMA) concentrations are related to obesity and obesity complications including decrement in insulin sensitivity and adiponectin levels, dyslipidemia and low-grade inflammation. Asymmetric dimethylarginine (ADMA) and symmetric dimethylarginine (SDMA) concentrations were analyzed by HPLC in 17 overweight (BMI ≥ 25 kg/m²) and 40 obese (BMI ≥ 30 kg/m²) premenopausal women. Age-matched healthy women were studied as controls. Obesity did not give rise to a significant change in circulating ADMA levels but reduced in SDMA levels. As compared with control subjects (0.441 ± 0.102 μM), ADMA values in overweight and obese subjects were found to be as 0.412 ± 0.102 and 0.436 ± 0.093, respectively. No Pearson’s association of ADMA with relevant risk variables for cardiovascular disease, including blood pressure, insulin sensitivity, inflammatory markers, lipid and adiponectin levels. However, in linear regression analysis, BMI, diastolic blood pressure, glucose, insulin, and IL-8 emerged as significant predictors of ADMA. In spite of obese women have elevated hs-CRP, triglyceride levels and decreased insulin sensitivity, adiponectin and HDL-cholesterol levels, all of which is closely linked risk factors for cardiovascular disease, circulating ADMA levels remained unchanged in obese individuals as compared with controls.     

Postprandial plasma glucose is an independent risk factor for increased carotid intima-media thickness in non-diabetic individuals.

Postprandial (pp) hyperglycemia--frequently associated with an increase in cardiovascular risk factors--may be damaging for the endothelium. So far, little information exists how glucose, insulin and lipids may affect atherosclerosis in the pp state. Therefore, we evaluated the relationship of pp hyperglycemia, insulin secretion and coronary risk factors to intima-media thickness (IMT) in a non-diabetic risk population. In 403 subjects (147 males, 256 females), aged 40-70 years, in the majority relatives of index cases with type 2 diabetes--a 75 g oral glucose tolerance test was performed together with measurement of insulin fractions, various risk factors and IMT of the common carotid artery. We found a continuous rise of 2h pp insulin fractions along the quintiles of 2h pp plasma glucose. A significant increase of body mass index, waist to hip ratio, triglycerides and decrease of HDL-cholesterol was observed in the top quintile of 2h pp plasma glucose (8.24 > or = pp plasma glucose < 11.1 mmol/l). Albuminuria was significantly enhanced in the 5th quintile. In parallel, IMT was significantly increased in the 5th quintile versus the bottom quintile of 2 h and maximal glucose (range 11.7-15.3 mmol/l) postprandially. After age and sex adjustment pp glucose and C-peptide, total cholesterol, triglycerides and HDL-cholesterol but not fasting plasma glucose were significantly correlated to IMT. In multivariate analysis age, male sex, pp plasma glucose, total and HDL-cholesterol were found to be independent risk factors for increased IMT. In conclusion, our data in a non-diabetic European risk population show that the two top quintiles of pp plasma glucose are associated with a clustering of standard risk factors. Corresponding to this clustering of risk factors IMT was significantly increased in the top quintile of 2 h and maximal pp plasma glucose. These data show that pp hyperglycemia may exert a noxious impact on the arterial wall together with a cluster of anomalies typical for the metabolic syndrome.     

Impaired insulin sensitivity as an independent risk factor for mortality in patients with stable chronic heart failure.

OBJECTIVES: The aim of this study was to determine the significance of insulin resistance as an independent risk factor for impaired prognosis in patients with chronic heart failure (CHF). BACKGROUND: In CHF, impaired insulin sensitivity (S(I)) indicates abnormal energy metabolism and is related to decreased exercise capacity and muscle fatigue. The relationship between insulin resistance (i.e., low S(I)) and survival in patients with CHF has not been established. METHODS: We prospectively studied 105 male patients with CHF due to ischemic (63%) or non-ischemic (37%) etiology. All patients were in clinically stable condition (age 62 +/- 1 year, New York Heart Association [NYHA] functional class 2.6 +/- 0.1, left ventricular ejection fraction [LVEF] 28 +/- 2%, peak oxygen uptake [Vo(2)] 18.2 +/- 0.7 ml/kg/min). Insulin sensitivity was assessed from glucose and insulin dynamic profiles during an intravenous glucose tolerance test using the minimal model technique. RESULTS: During a mean follow-up period of 44 +/- 4 months, 53 patients (50%) died. Patients with S(I) below the median value (median: 1.82 min(-1) x microU x ml(-1).10(4); n = 52) had worse survival (at two years 61% [range 47% to 74%]) than patients with S(I) above the median value (n = 53; at two years 83% [range 73% to 93%]; risk ratio [RR] 0.38, 95% confidence interval [CI] 0.21 to 0.67; p = 0.001). Both patient groups were similar in terms of age, NYHA functional class, and body composition parameters (dual-energy X-ray absorptiometric scan; p > 0.2), but patients with a lower S(I) had a lower LVEF (24 +/- 2% vs. 33 +/- 3%) and peak Vo(2) (16.8 +/- 1.0 ml/kg/min vs. 19.7 +/- 1.0 ml/kg/min; both p < 0.05). On univariate Cox analysis, higher S(I) predicted better survival (RR 0.56, 95% CI 0.35 to 0.89; p = 0.015). On stepwise multivariate analysis, S(I) predicted mortality independently of other variables. CONCLUSIONS: In patients with CHF, lower S(I) relates to higher mortality, independent of body composition and established prognosticators. Impaired S(I) may have implications in the pathophysiology of CHF disease progression. Therapeutically targeting impaired insulin sensitivity may potentially be beneficial in patients with CHF.     

Inflammatory and prothrombotic parameters in normotensive non-diabetic obese women: effect of weight loss obtained by gastric banding

Hypertension and diabetes are known risk factors for obesity-related thrombosis, but several studies have shown that obesity is characterised by a potentially prothrombotic inflammatory state because of activated coagulation and impaired fibrinolysis. In order to verify if obese patients—unaffected by hypertension, diabetes, dyslipidemia, cigarette smoking or inflammatory diseases—show increased prothrombotic markers and whether the weight loss induced by gastric banding normalises such parameters. Plasma levels of C reactive protein (CRP), fibrinogen, plasminogen activator inhibitor-1 (PAI-1), von Willebrand factor (VWF) and factor VII (FVII) were measured in 25 women with isolated obesity prior to, as well as 3, 6 and 12 months subsequent to gastric banding. Twenty-five healthy women served as a baseline control group. The obese women had higher CRP (p = 0.0001), fibrinogen (p = 0.014), PAI-1 (p = 0.003), VWF (p = 0.004) and FVII levels (p = 0.0001) than the normal controls, and their body mass index (BMI) positively correlated with CRP (r = 0.462, p = 0.02), fibrinogen (r = 0.426, p = 0.04) and PAI-1 (r = 0.468, p = 0.02). Twelve months after gastric banding, the median BMI had decreased from 40.0 to 34.9 (p = 0.0001); CRP from 4.18 to 1.69 μg/ml (p = 0.01); fibrinogen from 389 to 318 mg/dl (p = 0.0001); PAI-1 from 32.1 to 12.0 UI/ml (p = 0.003); VWF from 144 to 120% (p = 0.0001); and FVII from 134 to 112% (p = 0.002). Even in the absence of major cardiovascular risk factors, obese patients are characterised by a prothrombotic state. The weight loss induced by gastric banding decreases the parameters of inflammation, coagulation and impaired fibrinolysis, thus potentially reducing the thrombotic risk.     

High sensitivity C-reactive protein as an independent risk factor for essential hypertension

BACKGROUND: C-reactive protein (CRP), one of the hepatic acute phase reactants, has been associated with decreased endothelium-dependent relaxation, a potential risk factor for hypertension. However, the relationship between CRP and hypertension has not been well elucidated. The aim of this study is to assess whether circulating levels of CRP are independently related to essential hypertension. METHODS: We evaluated the relationship between high sensitivity CRP with blood pressure (BP) and several cardiovascular risk factors in a cross-sectional survey of 8347 apparently healthy Korean persons. The CRP was measured by nephelometry. RESULTS: The subjects consisted of 4813 men and 3534 women, aged >/=20 years. Mean (SD) age and CRP level of the population were 47.1 (11.5) years and 1.12 (1.72) mg/L. Overall hypertension prevalence was 34%. There was a significant positive association between BP and the CRP level (P <.0001). After adjustment for age, sex, fasting blood sugar, triglyceride, low-density lipoprotein, body mass index, waist-hip ratio, high-density lipoprotein, the prevalence of hypertension by CRP was 1.267 (95% confidence interval [CI] 1.079-1.487, P =.004), 1.253 (95% CI 1.062-1.477, P =.007), and 1.451 (95% CI 1.231-1.711, P <.001) times higher in subjects in the second, third, and fourth quartiles of CRP, as compared to subjects in the first quartile. CONCLUSIONS: Our results suggest that the CRP level may be an independent risk factor for the development of hypertension in Korean persons. However, because of the cross-sectional nature of our study, this finding should be confirmed in prospective cohort studies, aimed at elucidating the role of CRP in the prediction, diagnosis, and management of hypertension.     

B cell secretion and insulin sensitivity in hypertensive and normotensive obese subjects

To test the hypothesis that in obesity hypertension is associated with more pronounced hyperinsulinaemia and insulin resistance we compared plasma insulin levels and insulin sensitivity in a group of 6 obese subjects with untreated hypertension and in a group of 6 obese subjects with normal blood pressure. The two groups were similar for sex, age, body mass index and glucose tolerance. Six nonobese subjects served as controls. The study consisted of a 2-h hyperglycaemic clamp (steady-state plasma glucose = 11 mmol/l) and a 15-min insulin tolerance test (0.1 U/kg body wt). During hyperglycaemic clamp, insulin and C-peptide plasma levels were similar in normotensive and hypertensive obese subjects: the area under the plasma insulin curve was 36,000 +/- 3000 pmol/l X 120 min in the former and 34,000 +/- 1000 pmol/l X 120 min in the latter; the area under the plasma C-peptide curve was 298,000 +/- 26,000 pmol/l X 120 min in the former and 246,000 +/- 26,000 pmol/l X 120 min in the latter (P = n.s.). The ratio M/I between the amount of glucose metabolized (M) and the mean plasma insulin levels (I) during hyperglycaemic clamp was similar in the two groups: 0.59 +/- 0.09 in normotensive and 0.58 +/- 0.08 mg/min X m2 per pmol/l in hypertensive obese subjects (P = n.s.). Also the rate coefficient of glucose disappearance from plasma (K(itt)) after i.v. insulin injection was similar in the two groups (4.08 +/- 0.51 vs. 3.87 +/- 0.53 per cent/min).(ABSTRACT TRUNCATED AT 250 WORDS)     

A paraoxonase gene polymorphism,PON 1 (55), as an independent risk factor for increased carotid intima-media thickness in middle-aged women

Paraoxonase (PON) gene polymorphisms have been proposed as genetic markers of risk for cardiovascular disease (CVD). Sporadic results suggest they are correlated with intima-media thickness (IMT), an indicator of preclinical atherosclerotic disease. We have investigated whether polymorphisms PON 1 (M/L) 55, (Q/R) 192, PON 2 (S/C) 311 are related to site-specific carotid plaques in 310 middle-aged women. Subjects were also investigated for physical and biochemical parameters including oxidative markers to evaluate their effect on development of atherosclerotic plaques (IMT>1.2 mm) identified by high resolution B-mode ultrasound. We demonstrate that PON 1 (LL+ML) 55 is associated with plaques both at the bifurcation (OR=2.40; 95% CI 1.00-5.90) and at the common carotid artery (OR=2.75; 95% CI 1.01-7.50), and to the total number of plaques at any site (P<0.05). This polymorphism is an independent parameter with respect to other variables that are significantly associated with plaques, i.e. systolic blood pressure (OR=2.06; 95% CI 1.11-3.81) and oxidized low-density lipoprotein (LDL) antibodies (OR=1.96; 95% CI 1.05-3.69) in cases of common carotid plaques, and lipid peroxides (OR=1.86; 95% CI 1.00-3.50) in cases of bifurcation plaques. In conclusion, PON 1 (LL+ML) 55 but not PON 1 (Q/R) 192 or PON 2 (S/C) 311, appears to be an independent risk factor for increased carotid IMT in middle-aged women.     

Relationship of raised serum total and protein bound sialic acid levels with hyperinsulinemia and indices of insulin sensitivity and insulin resistance in non-diabetic normotensive obese subjects

AimsObesity is known to be associated with cardiovascular disease and interaction between inflammation and insulin resistance is reported to enhance the cardiovascular risk in these subjects. The present study was designed to assess indices of insulin sensitivity, insulin resistance and sialic acid levels and their association in non-diabetic normotensives obese subjects.Materials and methodsThe present study was conducted in 30 obese male subjects and results were compared with 30 subjects with normal body weight. Insulin, total sialic acid and protein bound sialic acid were estimated in all the subjects. Insulin resistance was calculated by using Homeostatic Model Assessment-insulin resistance formula. Insulin sensitivity was assessed by quantitative insulin check index and insulin sensitivity index.ResultsInsulin resistance, serum total and protein bound sialic acid levels were significantly increased in obese cases as compared to non-obese controls. Total sialic acid showed significant positive correlation with HOMA-IR (p < 0.01), BMI (p < 0.01), waist and hip circumference (p < 0.01) and negative correlation with QUICKI (p < 0.01) and insulin sensitivity index (p = 0.018). There was no significant correlation between protein bound sialic acid and indices of insulin resistance and insulin sensitivity.ConclusionSialic acid levels are elevated in obese subjects and its association with insulin resistance and reduced insulin sensitivity may enhance the cardiovascular risk in these subjects.     

The evaluation of carotid stenosis as risk factor for stroke in hypertensive diabetic and non-diabetic patients

The evaluation of intima-media thickening (IMT) of carotid artery is an important parameter in preclinical diagnosis of atherosclerosis and stroke risk in hypertensive patients. Measurements of carotid wall thickness via Doppler were considered to be valuable in prediction of risk of brain damage and future stroke due to hypertension in diabetic and non-diabetic patients.     

Resting metabolic rate in obese and normal weight women

The relationship between resting metabolic rate and different parameters of body size was investigated among 28 female volunteers in the age group of 20--30 years. The resting metabolic rate of the subjects was determined indirectly by measuring the oxygen consumption in a closed circuit, in which the oxygen concentration was stabilised. The fat percentage of the body was determined by densitometry. The population was divided into two groups: the obese, with an average fat percentage of 33.6 and the normal-weight with an average fat percentage of 20.4. Mean values for the resting metabolic rate were 1550 kcal/24 h (6.488 MJ/24 h) for the obese and 1421 kcal/24 h (5.948 MJ/24 h) for the normal-weight group. The resting metabolic rate per kg body weight was lower in the obese than in the normal-weight persons. However, expressed per kg fat-free body mass, energy expenditure under resting conditions in the obese was higher than in the normal-weight. No single body parameter seems to be suitable in the explantation of RMR in women with substantially different fat content. The best prediction of resting metabolic rate in this population of obese and normal-weight women is obtained when both fat-free mass and fat mass are used as independent variables in a linear regression equation.     

Plasma adiponectin and insulin sensitivity in overweight and normal-weight middle-aged premenopausal women

Adiponectin has been reported to regulate systemic insulin sensitivity as a part of a broader control mechanism in energy balance. However, it is not clear whether adiponectin exerts its positive effects on insulin sensitivity equally in a wide range of obesity. We investigated the association of plasma adiponectin concentration with insulin resistance (IR) in a cross-sectional sample of 98 middle-aged premenopausal women with a wide range of obesity. In addition, we studied the relationship between adiponectin, body composition, and blood biochemical and cardiorespiratory fitness variables. Body composition and fat distribution were measured via dual-energy x-ray absorptiometry in normal-weight (NW) (n = 41, body mass index [BMI] <25 kg/m²) and overweight (OW) (n = 57, BMI ≥25 kg/m²) women. Fasting blood samples were obtained; adiponectin, leptin, insulin, glucose, and insulin-like growth factor-I were measured; and IR index was calculated. The IR index from fasting plasma insulin and plasma glucose levels was estimated using the homeostasis model assessment (HOMA), as follows: fasting plasma insulin (in microliter units per milliliter) × fasting plasma glucose (in millimoles per liter)/22.5. Adiponectin was significantly higher (P = .0001) in NW (14.7 ± 4.7 μg/mL) compared with OW (9.9 ± 3.1 μg/mL) women. Significant differences (P < .003) in body mass, BMI, percentage of fat mass, fat mass, trunk fat, trunk fat-leg fat ratio, leptin, insulin, and HOMA were also observed between NW and OW groups. Leptin was independently related to plasma adiponectin (β = −.259, P = .001) in the overall study group. Plasma adiponectin was only related to trunk fat-leg fat ratio (β = −.242, P = .002) among NW subjects, whereas plasma adiponectin was related to fat-free mass (β = .182, P = .0001) and HOMA (β = −.576, P = .002) among OW women. The inverse relationship between adiponectin and leptin concentrations suggests that leptin may be involved in the regulation of adiponectin in middle-aged premenopausal women. Our data also demonstrate that adiponectin may play an important role in sustaining insulin sensitivity only in OW middle-aged premenopausal women.     

Relationships of body fat distribution,insulin sensitivity and cardiovascular risk factors in lean,healthy non-diabetic Thai men and women

In order to study the relationships of body fat distribution, insulin sensitivity and cardiovascular risk factors in lean, healthy non-diabetic Thai men and women, 32 healthy, non-diabetic subjects, 16 men and 16 women, with respective mean age 28.4+/-6.6 (S.D.) and 32.8+/-8.9 years, mean BMI 21.0+/-2.8 and 21.2+/-3.7 kg/m(2), were measured for total body fat and abdominal fat by dual energy X-ray absorptiometry (DEXA), anthropometry and insulin sensitivity by euglycemic hyperinsulinemic clamp. Cardiovascular risk factors included fasting and post-glucose challenge plasma glucose and insulin, blood pressure, lipid profile, fibrinogen and uric acid. For similar age and BMI, men had a lower amount and percent of total body fat, but had a higher proportion of abdominal/total body fat than women. In men, insulin sensitivity, as determined by glucose infusion rate during euglycemic hyperinsulinemic clamp, was inversely correlated with total body fat, abdominal fat, BMI and waist circumference, whereas only total body fat, but not abdominal fat, BW and hip circumference were inversely correlated with insulin sensitivity in women. No cardiovascular risk factors, except area under the curve (AUC), of plasma insulin in women correlated with insulin sensitivity when adjusted for total body fat. After age adjustment, total body fat was better correlated with fasting and AUC of plasma glucose and insulin in men and with systolic blood pressure as well as triglyceride levels in women. Only HDL-C in men was better correlated with abdominal fat. In conclusion, there were sex-differences in body fat distribution and its relationship with insulin sensitivity and cardiovascular risk factors in lean, healthy non-diabetic Thai subjects. Total body fat was a major determinant of insulin sensitivity in both men and women, abdominal fat may play a role in men only. Body fat, not insulin sensitivity, was associated with cardiovascular risk factors in these lean subjects.