Deterioration of the Metabolic Risk Profile in Women: Respective contributions of impaired glucose tolerance and visceral fat accumulation

A total of 249 women were recruited by solicitation through the media from the Québec City metropolitan area between 1987 and 1998. Subjects were between 31 and 74 years of age. All subjects were healthy nonsmoking volunteers and were not under treatment for coronary heart disease, diabetes, dyslipidemia, or endocrine disorders. IGT was determined as a plasma glucose level between 7.8 and 11.1 mmol/l 2 h after a 75-g glucose load, as recommended by the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (10). Menopause was determined as the absence of menstruation for at least 1 year, and 44% of menopausal women were on hormone replacement therapy (HRT). All participants signed an informed consent document approved by the Laval University Medical Ethics Committee.

The hydrostatic weighing technique was used to measure body density, which was obtained from the mean of six measurements (11). The percentage of body fat was derived from body density using the equation of Siri (12). Height, body weight, and waist and hip circumferences were measured according to the procedures recommended at the Airlie Conference (13), and the waist-to-hip ratio was calculated accordingly. Abdominal adipose tissue (AT) areas were measured by computed tomography with a Siemens Somatom DHR scanner (Erlangen, Germany), as previously described (14). Blood pressures were obtained using a sphygmomanometer and a stethoscope according to the recommendations of the American Heart Association (15).

Blood samples were collected from an antecubital vein into vacutainer tubes containing EDTA after a 12-h overnight fast. Cholesterol and triglyceride concentrations were determined in plasma and lipoprotein fractions using an analyzer Technicon RA-500 (Bayer Corporation, Tarrytown, NY), and enzymatic reagents were obtained from Randox (Randox Laboratories, Crumlin, U.K.). Plasma lipoprotein fractions (VLDL, LDL, and HDL) were isolated using procedures that have been previously described (5). Plasma apolipoprotein B (apoB) and apoA-I concentrations were measured by the rocket immunoelectrophoretic method of Laurell (16). LDL peak particle size was assessed by nondenaturating 2–16% polyacrylamide gel electrophoresis of whole plasma, as previously reported (17).

A 75-g oral glucose tolerance test (OGTT) was performed in the morning after an overnight fast. For measurements of plasma glucose, insulin, and C-peptide concentrations, blood samples were collected at −15, 0, 15, 30, 45, 60, 90, 120, 150, and 180 min in EDTA-containing tubes through a venous catheter placed in an antecubital vein. The method used was described in detail in a previous publication (5).

The presence of components of the metabolic syndrome was assessed using the following referenced cutoff points: triglycerides >2 mmol/l (18), HDL-cholesterol <1.15 mmol/l (19), cholesterol–to–HDL-cholesterol ratio >5 (20), visceral AT area >130 cm² (21), fasting insulin >60 pmol/l (22), apoB >1.22 g/l (23), and blood pressure >160/90 mmHg (24). The cutoff point for LDL peak particle size was determined using the 25th percentile value of nonobese women (BMI <25 kg/m²) with NGT.

Normality of the distribution for each variable was verified, and when variables were not normally distributed, log-transformed values were used for subsequent analyses. An unpaired Student’s t test was performed to compare women with NGT and women with IGT. A χ² test was used to compare the prevalence of metabolic disturbances among women with NGT and IGT. Women with NGT and with IGT were paired on the basis of visceral AT area (±10 cm²), fat mass (±1 kg), and menopausal status, and they were then compared using an unpaired Student’s t test. Pearson or Spearman correlation coefficients were calculated (depending on the normality of the distribution) to quantify the univariate associations among variables. Stepwise multiple regression analyses were computed to sort out the independent contributions of visceral AT accumulation, fat mass, glucose tolerance, and menopausal status to the variation of metabolic variables. Analyses were performed on the SAS statistical package (SAS Institute, Cary, NC).

Table 1 shows that women with IGT were characterized by increased adiposity because they had higher values for BMI, body fat mass, waist circumference, and waist-to-hip ratio as well as higher total abdominal and visceral AT areas than women characterized by NGT (P < 0.0001). Furthermore, 46% of the women with IGT had entered menopause compared with 30% of the women characterized by NGT (P < 0.05). Fasting plasma concentrations of cholesterol (total, VLDL, and LDL), triglycerides, and apoB as well as the cholesterol–to–HDL-cholesterol ratio were significantly higher in women with IGT (P < 0.01), whereas HDL-cholesterol and HDL₂-cholesterol concentrations and the HDL₂-cholesterol–to–HDL₃-cholesterol ratio were significantly lower in women with IGT compared with women characterized by NGT (P < 0.05). Furthermore, LDL peak particle sizes were measured in a subgroup of subjects, and women with IGT were characterized by a reduced LDL peak particle size (P < 0.04) compared with women with NGT. In addition (and as expected), women with IGT were characterized by higher levels of glucose, insulin, and C-peptide in the fasting state and after an oral glucose load. Finally, systolic and diastolic blood pressures were slightly but significantly higher in women with IGT compared with women characterized by NGT (P < 0.05).

Because there was a greater prevalence of postmenopausal women in the group of women with IGT, we performed an analysis of covariance for menopause status, and similar results were found for all physical and metabolic variables studied (results not shown).

We also compared the prevalence of deteriorated metabolic parameters among the two groups of women by χ² analyses (Fig. 1). Except for the small dense LDL phenotype, a significantly higher proportion of women with IGT were characterized by a deterioration in all the variables studied compared with women with NGT. We also calculated the number of risk factors found in every subject, and we found that women with IGT had a larger number of deteriorated metabolic parameters compared with women with NGT (3.7 vs. 2.0 risk factors, P < 0.0001).

To investigate the contribution of visceral AT area and fat mass to the differences between the IGT and NGT groups with regard to the lipoprotein-lipid variables, we matched 27 women characterized by NGT with 27 women with IGT. They were matched according to their amount of fat mass and visceral AT area as well as by their menopausal status. Women with NGT and IGT were obviously characterized by similar amounts of fat mass and visceral AT because they were matched on the basis of these variables (Table 2). Differences previously found in fasting plasma concentrations of glucose and C-peptide were eliminated, although women with IGT still displayed higher insulinemic-glycemic responses to an oral glucose load after this pairing procedure. Furthermore, also eliminated were differences initially found in plasma VLDL-cholesterol, LDL-cholesterol, HDL-cholesterol, HDL₂-cholesterol, total cholesterol–to–HDL-cholesterol ratio, and LDL peak particle size as well as in systolic and diastolic blood pressures (Table 2). On the other hand, significant differences remained with regard to concentrations of triglycerides, apoB, and total cholesterol.

We also compared the prevalence of deteriorated metabolic parameters between the two groups of women matched for body fat mass, visceral AT area, and menopausal status. We found that only the difference in the prevalence of triglyceride concentrations ≥2 mmol/l remained significantly different (P < 0.03); 41% of women with IGT were characterized by hypertriglyceridemia compared with only 15% of women with NGT (Fig. 2). When women with NGT were compared with those with IGT and similar body fat mass, visceral AT accumulation, and menopausal status, it was found that the total number of risk factors was not significantly different between the two groups (2.6 vs. 2.0 risk factors, NS).

Furthermore, despite a positive correlation between visceral fat and triglyceride concentrations in women with NGT (r = 0.52, P < 0.0001), no relationship between these two variables was observed in women with IGT (r = 0.18, NS). When we divided the two groups of women into low (<130 cm²) vs. high (≥130 cm²) levels of visceral AT, we found that women with NGT and low visceral AT had significantly lower triglyceride concentrations than women with NGT and high visceral AT accumulations (1.09 ± 0.38 vs. 1.86 ± 0.82 mmol/l; P < 0.0001). However, women with IGT and either low or high levels of visceral AT showed similar concentrations of triglycerides (1.87 ± 0.79 vs. 2.13 ± 0.83 mmol/l; NS). Finally, women with IGT and low levels of visceral AT had similar triglyceride concentrations compared with those with NGT and high visceral AT accumulations (1.87 ± 0.79 vs. 1.86 ± 0.82 mmol/l; NS).

It is well recognized that type 2 diabetes is associated with CVD. This relationship may be explained at least to some extent by the concomitant abnormalities in lipid and lipoprotein metabolism found in diabetic patients (25). Furthermore, type 2 diabetes is a heterogeneous condition from a metabolic standpoint and is invariably preceded by IGT (26). In that sense, IGT may also be a marker of CVD risk (27). Although a dyslipidemic state is frequently found in type 2 diabetic patients and in subjects characterized by IGT, it is also typically found in patients with visceral obesity who may nevertheless display NGT (28); it has been reported that individuals with a high accumulation of visceral AT are characterized by numerous alterations in their plasma lipoprotein-lipid profile (4,29). In this regard, we have recently reported that the higher visceral AT accumulation observed in men with IGT was an important factor involved in the deterioration of their CVD risk profile. In fact, we found that controlling for the variation in visceral AT accumulation eliminated the differences initially found between the two groups of men (NGT versus IGT) in their lipoprotein-lipid profiles, despite remaining differences in their indexes of plasma insulin–glucose homeostasis (5).

In the present study, women with IGT were also characterized by greater adiposity, including increased visceral fat accumulation compared with women with NGT, in accordance with previous observations (4,5). Increased risk for CVD in IGT may be partly attributable to elevated plasma glucose concentrations but may also be due to a greater prevalence of other CVD risk factors. Undoubtedly, there is an association between IGT and CVD. However, we have to bear in mind that in most cases, subjects with blood glucose in the upper normal range are also characterized by a cluster of several other unfavorable CVD risk parameters, such as elevated blood pressure and increased plasma lipids (30). In the present study, women with IGT were more frequently carriers of an important cluster of deteriorated metabolic parameters than women with NGT (3.7 vs. 2.0 risk factors).

The objective of the present study was to verify whether the increased visceral AT accumulation found in IGT could explain the altered lipid-lipoprotein profile associated with IGT. When we matched 27 women with NGT with 27 women characterized by IGT according to their body fat mass, visceral AT area, and menopausal status, most of the differences noted in the plasma lipoprotein-lipid profile were eliminated, despite major remaining differences in their plasma insulin-glucose homeostasis. However, differences in plasma triglyceride, apoB, and total cholesterol concentrations remained significant between the two matched groups. These results suggest that visceral AT accumulation contributes to the deterioration of the lipoprotein-lipid profile observed in women with IGT. However, alterations in insulin-glucose homeostasis that are typical of the IGT state may also play a role in the deterioration of the metabolic risk profile, especially for the regulation of circulating triglycerides. Therefore, present results indicate that high visceral AT accumulation (≥130 cm²) has a major impact on triglyceride concentrations in the NGT state. However, in the IGT state, the impact of visceral AT appears to be attenuated. The factors that may explain why women with IGT show high triglyceride concentrations even in the presence of low visceral AT accumulation remain unclear. It is also important to note that the independent effect of IGT per se on triglyceride concentration was not observed in our study performed in men (5). This may suggest that mechanisms that are regulating triglyceride concentrations are, to some extent, sex-specific. Accordingly, it is also interesting to note that the CVD risk associated with high triglyceride levels may also be sex-specific, because the evidence suggesting that hypertriglyceridemia is an independent CVD risk factor appears to be stronger in women than in men (31).

It is well known that diabetes is a significant and powerful risk factor for CVD. However, the relative CVD risk associated with type 2 diabetes appears to be greater in women than in men (2). It has been suggested that this diabetes-related increased CVD risk in women could be explained by the fact that diabetes has more adverse effects on CVD risk factors in women than in men. Mechanisms underlying the stronger relationship between lipids and glycemia in women than in men are not fully understood. Sex hormones may have a possible impact on this relationship. Indeed, lower sex hormone–binding globulin (SHBG) levels were found in postmenopausal women with IGT compared with those with NGT. These women also have higher androgen activity (as reflected by the testosterone-to-SHBG ratio) than postmenopausal women with NGT (32). Because insulin inhibits SHBG secretion in the liver (33), the hyperinsulinemic state of women with IGT may be associated with a dyslipidemic profile through alterations in SHBG and the estrogen/androgen balance (34).

Although normalization of blood glucose in type 2 diabetes may be cost-effective, strong experimental evidence shows that treatment of other risk factors, particularly hypertension and hyperlipidemia, reduces cardiovascular mortality even among persons with diagnosed type 2 diabetes (35). Therefore, clinical interventions to reduce metabolic risk should focus on obesity, hypertension, and the sedentary lifestyle, which promote insulin resistance. Thus, because lipoprotein disorders are not innocent bystanders of the abnormalities that lead to type 2 diabetes (36), we believe that it is important to identify not only individuals with IGT but also those characterized by abdominal obesity, physical inactivity, and lipid disorders to prevent further aggravation of their hyperglycemia and dyslipidemia.

In summary, the deterioration in the lipoprotein-lipid profile observed in women with IGT is largely related to the increased visceral AT depot found in these women, and it is partly related to alterations in glucose-insulin homeostasis, especially the modulation of plasma triglyceride concentrations, that are typical of the IGT state.

This study was supported by the Canadian Diabetes Association and by the Canadian Institutes of Health Research (MT-14014 and MGC-15187). I.L. is a recipient of a studentship from the Heart and Stroke Foundation of Canada; J.B. is a clinical research scholar from the Fonds de la Recherche en Santé du Québec; and S.L. is a research scholar from the Canadian Institutes of Health Research.

The authors express their gratitude to the subjects for their excellent collaboration and to the staff of the Lipid Research Center, the Physical Activity Sciences Laboratory, and the Diabetes Research Unit for their contribution to this study. We especially thank L. Allard, S. Brulotte, L. Bargone, G. Fournier, H. Bessette, and C. Leblanc for their help in the collection and analysis of the data.