Journal Logo

Clinical Sciences: Clinically Relevant

Energy Expenditure from Physical Activity and the Metabolic Risk Profile at Menopause


Author Information
Medicine & Science in Sports & Exercise: February 2005 - Volume 37 - Issue 2 - p 204-212
doi: 10.1249/01.MSS.0000152702.57291.FA
  • Free


Aging and behavioral changes contribute to the increased tendency of women to gain weight and become sedentary after menopause (23). Due to the absence of the protective effects of endogenous estrogens (24), the age-induced weight gain in postmenopausal women is characterized by an increase in visceral adipose tissue (AT) (16). The accumulation of visceral AT is associated with an increased risk of metabolic dysfunctions in pre- and postmenopausal women, such as hypertriglyceridemia, elevated plasma apolipoprotein B (apoB) concentration, insulin resistance, impaired glucose tolerance (IGT), and Type 2 diabetes (14,18,22). These metabolic dysfunctions are associated with an increased risk of cardiovascular disease (CVD), the leading cause of death in women in Canada (12).

In addition to the aging and postmenopausal-induced decrease in body fat-free mass and resting metabolic rate (RMR) (23), physical inactivity also represents an important risk factor for body weight gain, visceral AT accumulation, and CVD (12). Poehlman et al.(23) have shown in a longitudinal study that women entering menopause reduced their engagement in physical activity and had decreased body fat-free mass and an increased waist-to-hip ratio and fat mass compared with women who had not entered menopause. It has also been recognized that the lack of sufficient physical activity can be responsible for weight gain more than any other environmental factors in healthy middle-aged women (27).

Although visceral AT accumulation during menopause increases the risk of metabolic dysfunctions and CVD (22), engagement in physical activity seems to counteract it. In fact, it has been shown that physical activity is negatively associated with visceral AT accumulation in postmenopausal women (16,17). Lifestyle interventions with physical activity have also been associated with an improvement in insulin sensitivity and a reduced risk of Type 2 diabetes development in obese and nonobese women (18). Finally, it has been reported that the risk of cardiovascular disease was approximately 40% lower in physically active compared with nonactive diabetic women (15).

Physically active women are characterized by lower body mass index (BMI) and visceral AT levels (14,17) than nonactive women. The hypothesis that the beneficial effects of physical activity on the metabolic profile and CVD risk could be attributable to the decrease in visceral AT, rather than to physical activity per se, has thus been studied. It seems that in pre- and postmenopausal women, the beneficial effect of physical activity on some of the CVD risk factors is modulated through variations in visceral fat (16). However, it is still unclear whether postmenopausal women with high levels of visceral AT would have a more favorable metabolic risk profile when physically active compared with sedentary postmenopausal women with similar visceral AT accumulation. Accordingly, the objectives of our study were to assess the relationships between daily energy expenditure from moderate to intense physical activity and parameters of the metabolic risk profile in visceral obese and nonobese postmenopausal women not receiving hormone therapy (HT) and to verify whether these associations are independent of visceral AT accumulation. Our hypothesis was that daily energy expenditure from moderate to intense physical activity would independently contribute to the determination of the metabolic profile in our sample of postmenopausal women.



Subjects of this study were recruited through the local newspapers of the Quebec City metropolitan area. A total of 386 women responded to the advertisements. Women were individually interviewed to evaluate whether they satisfied this study’s inclusion criteria for age, postmenopausal status (confirmed by absence of menses for at least 1 yr and levels of follicle-stimulating hormone between 28 and 127 IU·L−1), absence of any HT and other medication, except a stable dose of thyroxine for known well-controlled hypothyroidism. One hundred ninety-six women were not eligible: 9 were older than 70 yr, 36 were not postmenopausal, 52 were using HT, 93 were taking medication, and 6 had unstable body weight. One hundred ninety women were found to be eligible. Among them, 69 dropped out of the study for personal reasons after having received a complete description of the research protocol. Data from three women who did not complete the physical activity record are not included in the present study. Therefore, data from 118 eligible postmenopausal women were used in this study. All participants signed a written consent form, approved by the Laval University Medical Ethics Committee. All but five women were nonsmokers, and none of the women included in the study was receiving treatment for CHD, diabetes, dyslipidemias, or endocrine disorders. Systolic (SBP) and diastolic blood pressures (DBP) were measured in the right arm of seated participants. None of the participants had received a diagnosis of Type 2 diabetes before the study. All women involved in the study were white.

Physical activity record and food record.

Women filled out a 3-d activity diary (2) including two weekdays and one weekend day. In the activity record, a day (24 h) was divided into 96 periods of 15 min each. Subjects used a list of categorized activities to fill out their diary. Activities were classified according to mean energy expenditure on a 1–9 scale. Therefore, during the 3 d the activity diary was filled, women recorded the dominant activity that they were engaged in for each 15-min period, using the gradation 1–9. For example, category 1 indicated activities of very low energy expenditure such as sleeping or resting in bed. Category 6 indicated activities of moderate energy expenditure such as light manual work or light exercise such as playing golf or riding a bike. Finally, category 9 indicated activities of very high energy expenditure such as running or intense manual work (2). Our study focused on mean daily energy expenditure and frequency of participation for categories of energy expenditure 6, 7, 8, and 9 (EE6–9), which have an energy cost ≥1.2 kcal·kg−1·15 min−1 (≥4.8 METs). EE6–9, expressed as kcal·kg−1·d−1, was calculated by multiplying the number of 15-min periods of categories 6 to 9 by the approximate median energy cost of each category previously determined by the authors of the activity diary (2) (i.e., 1.2, 1.4, 1.5, and 2 kcal·kg−1·15 min−1 for categories 6, 7, 8, and 9, respectively). The following formula was used: EE6–9 = (number of 15-min periods of category 6 × 1.2) + (number of 15-min periods of category 7 × 1.4) + (number of 15-min periods of category 8 × 1.5) + (number of 15-min periods of category 9 × 2). The mean EE6–9 value of the 3 d was then calculated (kcal·kg−1·d−1) and was used for analyses. The physical activity record used in this study has some limitations as described by Bouchard et al. (2). Indeed, the procedure of this physical activity diary relies on approximation because the energy cost used in conjunction with each categorical value is the approximate median amount of energy expended when engaged in the activities of the relevant category (2). Nevertheless, this physical activity record has been validated (2) and is appropriate to estimate a subject’s average time and frequency of participation in activities of different physical intensities (2), which was also this study’s objective. Dietary intakes were collected using a 3-d food record, which was completed during two weekdays and one weekend day. Guidelines for completing the food record were explained to subjects by the study nutritionist. Women were asked to weigh foods with a scale provided by the nutritionist. The evaluation of nutrient intakes derived from the food record was performed using the Nutrition Analysis Software Food Processor version 7.2 (ESHA Research, Salem, OR).

Anthropometric measurements.

Body density was estimated with the hydrostatic weighing technique (1). The mean of six valid measurements was used to calculate the percentage of body fat from body density with the equation of Siri (28). Body fat mass was obtained by multiplying the percentage body fat by body weight. Height, body weight, and waist and hip circumferences were determined following the procedures recommended at the Airlie Conference (20), and the BMI and waist-to-hip ratio were calculated. Waist and hip circumferences were measured to the nearest millimeter with a measuring tape. Height was measured to the nearest 0.1 cm with a stadiometer, and body weight was measured to the nearest 0.1 kg on a calibrated scale. Participants wore swimming suits and were asked to remove their shoes for these last measurements.

Computed tomography.

Total, visceral AT, and subcutaneous AT areas were assessed by computed tomography, which was performed on a GE High Speed Advantage CT scanner as previously described (8). Briefly, subjects were examined in the supine position with both arms stretched above the head. The computed tomography scan was performed at the abdominal level between the L4 and L5 vertebrae using a radiograph of the skeleton as a reference to establish the position of the scan to the nearest millimeter. Total abdominal AT area was calculated by delineating the abdominal scan with a graph pen and then computing the AT surface using an attenuation range of −190 to −30 Hounsfield units (19,29). The abdominal visceral AT area was measured by drawing a line within the muscle wall surrounding the abdominal cavity. The abdominal subcutaneous AT area was calculated by subtracting the amount of abdominal visceral AT from total abdominal AT area.

Oral glucose tolerance test (OGTT).

A 75-g OGTT was performed in the morning after 12-h overnight fast. Blood samples were collected in EDTA-containing tubes (Becton Dickinson, Franklin Lakes, NJ) through a venous catheter from an antecubital vein at −15, 0, 15, 30, 45, 60, 90, 120, 150, and 180 min for the determination of plasma glucose, insulin, and C-peptide concentrations. Plasma glucose was measured enzymatically (25), whereas plasma insulin was measured by radioimmunoassay with polyethylene glycol separation (5). Plasma concentration of glucose at 120 min of the OGTT was measured (2-h postload glycemia). Plasma C-peptide levels were measured by a modification of the method of Heding (13) with polyclonal antibody A-4741 from Ventrex (Portland, ME) and polyethylene glycol precipitation (5). The interassay coefficient of variation was 1.0% for a basal glucose value set at 5.0 mmol·L−1.

Criteria used to define glucose tolerance states.

Subjects were classified according to their fasting plasma glucose (FPG) concentration and their 2-h postload glycemia. Normal glucose tolerance (NGT) was defined as a FPG concentration <6.1 mmol·L−1 and a 2-h postload glycemia <7.8 mmol·L−1. IGT was defined as FPG concentration <7.0 mmol·L−1 and a 2-h postload glycemia ≥7.8 and <11.1 mmol·L−1. Type 2 diabetes was defined as a FPG ≥7.0 mmol·L−1 or a 2-h postload glycemia ≥11.1 mmol·L−1 (9).

Insulin sensitivity.

Insulin sensitivity was measured by the hyperinsulinemic-euglycemic clamp technique as described by DeFronzo et al. (4). Subjects arrived at the Diabetes Research Unit after a 12-h overnight fast. Indwelling catheters were inserted in the antecubital veins bilaterally. One catheter was used for administration of insulin (Humulin 40 U·m−2·min−1) and glucose (20% dextrose), and the other was used for drawing blood samples. Insulin was administrated intravenously as a primed continuous infusion during 2-h and glucose was administrated simultaneously. Blood glucose was monitored every 5 min during the insulin infusion, and euglycemia was maintained during the clamp by infusing 20% dextrose at a variable rate. The duration of the insulin infusion was such that the rate of infused glucose reached a constant value during the last hour of the clamp. Blood samples were collected from time −15 min and then every 5 min during the test to measure the blood glucose concentration by using a Glucometer-Elite (number 3903-E; Bayer Corporation Inc., Tarrytown, NY) (25). Plasma insulin concentrations were monitored from blood samples collected every 10 min and stored at −20°C for later analyses using radioimmunoassay with polyethylene glycol separation (5). M value corresponds to the glucose infusion rate per kilogram of body weight during the last 30 min of the clamp. Insulin sensitivity (M/I) was calculated by dividing the M value by the mean insulin concentration during the last 30 min of the clamp. Free fatty acids (FFA) were measured in the fasting state and every 30 min of the clamp spectrophotometrically (Wako Chemicals GmbH, Neuss, Germany). The FFA suppression rate was calculated with the following formula: [(fasting FFA concentration − mean of 90-min and 120-min FFA concentration) × fasting FFA concentration].

Plasma lipid-lipoprotein profile.

On the morning of the hyperinsulinemic-euglycemic clamp, blood samples were collected in the fasting state to measure a complete plasma lipid–lipoprotein profile. After a 12-h overnight fast, blood samples were collected from an antecubital vein into Vacutainer tubes containing EDTA. Chol and triglyceride (TG) concentrations were determined enzymatically in plasma and lipoprotein fractions with a Technicon RA-500 analyzer (Bayer Corporation Inc.), and enzymatic reagents were obtained from Randox Laboratories (Crumlin, UK). Plasma lipoprotein fractions (VLDL, LDL, and HDL) were isolated by ultracentrifugation (11). The HDL-chol fraction was obtained after precipitation of LDL-chol in the infranatant (d >1.006 g·mL−1) with heparin and MnCl2 (3). The chol and TG content of the infranatant were measured before and after the precipitation step. HDL2-chol was precipitated from the HDL-chol fraction with a 4% solution or low molecular weight dextran sulfate (15–20 kd) obtained from SOCHIBO (Boulogne, France). The chol content of the supernatant fraction (HDL3-chol) was determined, and HDL2-chol levels were derived by subtracting HDL3-chol from total HDL-chol concentrations (10). The chol and TG content of the infranatant were measured before and after the precipitation step. ApoB was measured by nephelometry (BN ProSpec; Dade Behring Inc., Newark, NJ) with reagents provided by this company (N antisera to human ApoB).

Statistical analyses.

Statistical analyses were performed using software from the SAS Institute (8.02 version) (Cary, NC). For the first series of analyses, simple Pearson correlations were computed within the total sample of women. P values were considered significant at <0.05. Partial correlations were sought between EE6–9 and metabolic parameters with adjustment for visceral AT and between visceral AT and metabolic parameters with adjustment for EE6–9. According to their visceral AT accumulation and EE6–9 level, the total group of subjects was further divided into four groups on the basis of median values of visceral AT (< or ≥134 cm2) and EE6–9 (> or ≤1.6 kcal·kg−1·d−1): 1) low VAT-A (active): women with low visceral AT (<134 cm2) and high EE6–9 (>1.6 kcal·kg−1·d−1); 2) low VAT-NA (nonactive): women with low visceral AT (<134 cm2) and low EE6–9 (≤1.6 kcal·kg−1·d−1); 3) high VAT-A: women with high visceral AT (≥134 cm2) and high EE6–9 (>1.6 kcal·kg−1·d−1) and 4) high VAT-NA: women with high visceral AT (≥134 cm2) and low EE6–9 (≤1.6 kcal·kg−1·d−1). Comparisons between these groups were made by ANOVA using the general linear model, and the Duncan post hoc test was used in situations in which a statistically significant group effect was observed. In the four groups of women separated on the basis of visceral AT and EE6–9, the chi-square test was used to compare the prevalence of women with NGT, IGT, and Type 2 diabetes. A Student’s t-test was also performed to compare a subgroup of our sample composed of women with either IGT or Type 2 diabetes (IGT/T2D), which was further subdivided on the basis of the total group median value for EE6–9 (> or ≤1.6 kcal·kg−1·d−1): 1) IGT/T2D-A (active): women with IGT or Type 2 diabetes and high EE6–9 (>1.6 kcal·kg−1·d−1) and 2) IGT/T2D-NA (nonactive): women with IGT or type 2 diabetes and low EE6–9 (≤1.6 kcal·kg−1·d−1). Analyses were performed on log transformed values for variables that were not normally distributed. For EE6–9, the log(x + 1) value was used to normalize the distribution.


Anthropometric and metabolic characteristics of the subjects are shown in Table 1. Simple correlations between anthropometric and metabolic variables are shown in Table 2. Anthropometric variables (BMI, body fat mass, visceral AT, subcutaneous AT, and waist circumference) were all significantly and positively associated with SBP, DBP, TG, FPG, 2-h postload glycemia, and fasting C-peptide and negatively associated with insulin sensitivity (Table 2). With the exception of waist circumference, all anthropometric variables were significantly associated with plasma fasting FFA concentrations. BMI was significantly and negatively associated with HDL-chol, HDL2-chol, and FFA suppression rate, whereas body fat mass was not significantly associated with any of the lipid-lipoprotein variables studied. Visceral AT and waist circumference were positively associated with total chol/HDL-chol ratio and were negatively associated with HDL-chol, HDL2-chol, and FFA suppression rate, whereas subcutaneous AT area was significantly correlated with HDL2-chol and FFA suppression rate.

Characteristics of postmenopausal women into study (N = 118).
Correlation coefficients for the associations between anthropometric and metabolic parameters studied.

The daily participation in the nine categories used in the physical activity record was such that the numbers of 15-min periods spent in every category decreased as the category’s energy expenditure increased (35.0 vs 0.1 15-min periods for categories 1 and 9, respectively). Women spent an average of 31.3 ± 40.7 min·d−1 at activities in categories 6–9.

Among the total group of women, there was a negative and significant relationship between BMI and EE6–9 (r = −0.22, P < 0.05) and between visceral AT and EE6–9 (r = −0.18, P < 0.05). It was also found that EE6–9 was significantly associated with the percentage of body fat mass (r = −0.27, P < 0.01), subcutaneous AT area (r = −0.23, P < 0.05), and waist circumference (r = −0.25, P < 0.01). No significant association was found between EE6–9 and age or between EE6–9 and the macronutrient composition of the diet (percentage of energy derived from proteins, carbohydrates, and fat).

Table 3 shows that EE6–9 was significantly associated with SBP, DBP, TG, HDL-chol, HDL2-chol, total chol/HDL-chol ratio, FPG, 2-h postload glycemia, fasting C-peptide, and insulin sensitivity. When correlation analyses included adjustment for visceral AT, it was found that DBP, TG, total chol/HDL-chol ratio, and 2-h postload glycemia lost their significant association with EE6–9. However, SBP, HDL-chol, HDL2-chol, FPG, fasting C-peptide, and insulin sensitivity remained significantly associated with EE6–9 after adjustment for visceral AT. Adjustment for BMI resulted in the same pattern of partial correlations between EE6–9 and the metabolic variables as those observed when associations were adjusted for visceral AT. All associations between visceral AT and metabolic variables presented in Table 2 remained significant after adjustment for EE6–9 (results not shown).

Correlation coefficients for the associations between EE6–9 and the metabolic parameters studied.

To examine further the interaction between participation in physical activity and visceral AT accumulation, the four subgroups of women (low VAT-A, low VAT-NA, high VAT-A, and high VAT-NA women) were compared for metabolic variables that were found to be associated with participation in physical activity independently of visceral AT (SBP, HDL-chol, HDL2-chol, FPG, fasting C-peptide, and insulin sensitivity) (Table 4). The high VAT-NA women had significantly higher SBP and FPG values and lower insulin sensitivity compared with the other three groups (P < 0.05). They also had significantly lower HDL-chol and HDL2-chol values as well as higher fasting C-peptide than low VAT-A and low VAT-NA women. High VAT-A women had significantly lower HDL-chol, HDL2-chol, and insulin sensitivity values and higher fasting C-peptide concentrations than low VAT-A women. The only significant difference between low VAT-NA and low VAT-A women was for fasting C-peptide concentration, which was significantly higher in low VAT-NA women. Because BMI was significantly higher in the high VAT-NA group than in the high VAT-A group, comparisons were made after adjustment for BMI to see the influence of BMI on differences observed between these two groups. After this adjustment, the significant difference in insulin sensitivity that was originally observed between high VAT-NA and high VAT-A groups was eliminated, whereas all other differences remained significant (results not shown). Finally, it was noted that no women with Type 2 diabetes were part of the low VAT-A group but that 67% of them figured among the high VAT-NA group. When the diet compositions of the four groups of women were compared, it was found that the percentage of energy derived from fat was lower in low VAT-A than in high VAT-NA women. Hence, comparisons between the four groups for SBP, HDL-chol, HDL2-chol, FPG, fasting C-peptide, and insulin sensitivity were repeated, with adjustment for the relative content of fat in the diet. After this adjustment procedure, all differences observed previous to adjustment remained significant.

Comparison of selected metabolic parameters between women according to their visceral AT accumulation and EE6–9 values.

Among the total group of postmenopausal women participating in this study, 15% (N = 18) were diagnosed as having Type 2 diabetes and 27% (N = 32) were diagnosed as having IGT. Among the 32 women with IGT, 19% (N = 6) also had impaired fasting glucose, that is, fasting glycemia ≥6.1 and <7.0 mmol·L−1. Women with IGT or Type 2 diabetes (IGT/T2D) had significantly more visceral AT than women with an NGT (162.5 ± 46.9 cm2 vs 122.6 ± 57.3 cm2, P < 0.01). To examine the effect of physical activity on the metabolic profile of women with impaired glucose homeostasis, analyses were conducted while considering only women with either IGT or Type 2 diabetes. These women were then separated into two subgroups (IGT/T2D-A and IGT/T2D-NA) on the basis of their EE6–9 value (Table 5). The proportion of women with IGT or Type 2 diabetes did not differ between these two subgroups (chi-square = 2.97; P = 0.08). In addition, no significant differences in age, BMI, waist circumference, and visceral AT were observed between the two subgroups. Table 5 shows that IGT/T2D-A women had lower SBP (P < 0.05), DBP (P < 0.01), and TG concentrations (P < 0.05), total chol/HDL-chol ratio (P < 0.05), FPG (P < 0.05), 2-h postload glycemia (P < 0.01), and fasting C-peptide values (P < 0.05) than IGT/T2D-NA women. IGT/T2D-A women also had higher HDL-chol, HDL2-chol (P < 0.05), and FFA suppression rates (P < 0.01) as well as a tendency to have a higher insulin sensitivity (P = 0.06) than IGT/T2D-NA women.

Comparison between women with impaired glucose tolerance and/or Type 2 diabetes according to their EE6–9 values.


A comprehensive metabolic risk profile of 118 postmenopausal women not receiving HT was assessed using validated metabolic measurements. This information was related to energy expenditure from women’s daily engagement in moderate to intense physical activity measured with a 3-d activity record (2). To our knowledge, this is the first study to report the influence of physical activity on a comprehensive set of metabolic variables, independently of visceral AT in postmenopausal women.

Physical activity record.

The daily energy expenditure for two weekdays and one weekend day has been assessed for all subjects. Because we were interested in the women’s participation in physical activity, we took into consideration the categories of energy expenditure 6–9, which correspond to physical activity of moderate to intense intensity (2). Our interest in these categories of energy expenditure is explained by previous studies that showed an inverse relationship between the volumes or levels of physical activity (i.e., the amount of energy expended) and the risk of Type 2 diabetes, CVD rates, and all-cause mortality in women (15). For the same reason, we did not consider categories of energy expenditure 1–5, corresponding to activities of low energy expenditure. Our results showed that women spent an average of approximately 30 min·d−1 at activities in categories 6–9. These results do not differ markedly from some data previously obtained in the province of Québec, Canada. In fact, it has been shown that 63.8% of women aged 45–64 yr were moderately to very active, that is, engaged 2–4× wk−1, for a minimum of 25–50 min each time, in moderate to intense physical activities (21).

Correlations between EE6–9 and anthropometric parameters.

EE6–9 was significantly and negatively associated with BMI and visceral AT accumulation. These results are similar to those obtained in previous studies in postmenopausal women (14,17). Because of the cross-sectional nature of our observations, it is not possible to determine whether nonobese women in our study would have had a “natural” tendency to be more active or whether women’s participation in physical activity would have contributed to their lower BMI and visceral AT. We cannot rule out the possibility that obese women may be less involved in physical activity compared with nonobese women because of their heavier body weight. However, it has also been established from longitudinal studies that physical activity plays a role in attenuating the age-related weight gain (6).

Correlations between EE6–9 and metabolic parameters.

We have also demonstrated significant associations between EE6–9 and many metabolic parameters. Associations observed between EE6–9 and metabolic parameters in our study are of similar magnitude compared with relationships between physical activity level and metabolic variables found in other studies (16). Associations between EE6–9 and DBP, TG, total chol/HDL-chol ratio, and 2-h postload glycemia lost their significance after adjustment for visceral AT. These results suggest that the link between EE6–9 and these variables was largely mediated by visceral AT because active subjects were characterized by less visceral AT accumulation. It also highlights the fact that visceral AT seems to be a strong predictor of a deteriorated metabolism, in accordance with previous data from the literature (14,16,22). On the other hand, the association between EE6–9 and SBP, HDL-chol, HDL2-chol, FPG, fasting C-peptide, and insulin sensitivity remained significant after adjustment for visceral AT. Although these results do not exclude the contribution of visceral AT to the determination of these metabolic variables, it does put emphasis on the beneficial and independent effect of physical activity. More specifically, results of this study demonstrated that engagement in physical activity has a determinant influence on SBP, especially among viscerally obese women. Significantly higher values of SBP were reported for the high VAT-NA women compared with the other three groups. This would be in accordance with the lower SPB levels observed in aerobically trained postmenopausal women compared with untrained postmenopausal women (30). These authors suggest that the higher waist-to-hip ratio in untrained postmenopausal women might explain the difference between the two groups (30). However, our results showed similarity between high VAT-A and both groups of low VAT women for SBP values, which would emphasize the importance of physical activity for maintaining normal SBP values among high VAT-A women. Insulin sensitivity values followed the same trend because high VAT-NA women were less insulin sensitive than any other group. It has to be recognized that the lower insulin sensitivity in high VAT-NA compared with high VAT-A could be partly explained by the higher BMI in high VAT-NA women because adjustment for BMI eliminated the difference in insulin sensitivity between the two groups. It was interesting to note that high VAT-A women had values for insulin sensitivity similar to those of low VAT-NA women. These results demonstrated again the strong relationship between physical activity and insulin sensitivity, even in the presence of visceral obesity. This is in accordance with the observed increase in insulin sensitivity induced by physical activity, independently of waist circumference and BMI in previously sedentary adults (7). Nevertheless, our results also showed that for a similar level of EE6–9, low VAT-A women were more sensitive to the action of insulin compared with high VAT-A women. This argues in favor of an additional contribution of visceral AT to insulin sensitivity, which, when combined with a sedentary lifestyle, can lead to the highest level of insulin resistance, as found in the high VAT-NA subgroup.

With regard to FPG, we found that high VAT-A women had significantly lower FPG concentrations than high VAT-NA women and had values similar to those in both groups of women with low visceral AT. This suggests that physical activity in the presence of visceral obesity can lead to normalization of glucose concentration. The impact of EE6–9 on insulin secretion, as determined by fasting C-peptide concentration, was more apparent among women with low VAT accumulation. In fact, according to our results, having low VAT accumulation and being physically active appear to be two essential conditions for lowered insulin secretion. Partial correlation analyses suggest a beneficial influence of physical activity on HDL-chol and HDL2-chol concentrations. Our results are not fully concordant with those of Hunter et al. (16) who reported no significant association between physical activity and HDL-chol and HDL2-chol after adjustment for visceral AT. Nevertheless, the differences between high VAT-A and low VAT-A women and between high VAT-NA and low VAT-NA women for HDL-chol and HDL2-chol concentrations also suggest a significant influence of visceral AT on these variables.

These results altogether suggest that, although participation in physical activity cannot completely compensate for the deleterious effects of visceral AT, it accounts, at least partly, for more favorable blood pressure, insulin sensitivity, FPG concentration, HDL-chol, HDL2-chol, and insulin secretion, which are markers of an improved CVD risk profile in postmenopausal active women.

Influence of physical activity on the metabolic parameters in women with IGT or T2D.

When compared with NGT women, the IGT/T2D women had significantly higher visceral AT accumulations, which is in accordance with previous results showing that accumulation of visceral AT was associated with an increased risk of metabolic dysfunction such as IGT and Type 2 diabetes (14,22). Our results once more suggest that, despite a high level of visceral AT and an impaired glucose homeostasis, women still have a more favorable risk profile when physically active. In fact, we observed that IGT/T2D women who reported being physically active had significantly lower values of DBP and SBP, fasting C-peptide, FPG, 2-h postload glycemia, TG, and lower total chol/HDL-chol ratio. They also had higher HDL-chol and HDL2-chol values as well as higher FFA suppression rates compared with the nonactive IGT/T2D women. Therefore, the physical activity–related beneficial influence on the metabolic parameters observed in our study might result in a decreased risk of CVD among women with IGT or Type 2 diabetes (15) as well as a beneficial environment to prevent the development of Type 2 diabetes in women with IGT. This concurs with results from epidemiological studies (15) such as the Diabetes Prevention Trial, which reported that men and women who maintained moderate-intensity physical activity of at least 150 min·wk−1, followed a healthy diet, and lost an average of 5.6 kg had a 58% lower incidence of Type 2 diabetes compared with controls (18).

Physiological mechanisms to explain the favorable influence of participation in physical activity.

The beneficial effects of physical activity on lipid and glucose parameters reported in this study could be explained through its capacity to improve insulin sensitivity and lipase regulation (7). In accordance with this, we found in our study that FFA suppression rates were significantly lower in high VAT-NA compared with low VAT-A women (results not shown) and were also lower in IGT/T2D-NA women compared with their more active counterparts. This suggests that physical activity could increase the antilypolitic effect of insulin. Because increased plasma FFA concentrations have been related to impairment in glucose homeostasis and plasma lipid lipoprotein metabolism (26), the improved suppression of FFA in response to insulin could explain, at least partly, the more favorable metabolic profile associated with the higher engagement in physical activity.


In conclusion, we demonstrated among a group of postmenopausal women that energy expenditure from moderate to intense physical activity was associated with lower BMI and lower visceral AT accumulation as well as with a more favorable metabolic profile. Our results also suggest a beneficial influence of physical activity on the metabolic risk profile in women with Type 2 diabetes or IGT. We have to acknowledge that our sample was not selected to be representative of the whole population of Canadian postmenopausal women. This precludes generalization of our results.

Nevertheless, our results are of relevant interest because the long-term success of weight loss interventions is poor. When appropriate, an option could be to focus on increasing physical activity rather than focusing solely on weight loss, especially because physical activity can also be an important determinant of weight maintenance. Therefore, in women, participation in moderate to intense physical activity may represent a favorable and rewarding approach to coping with the deterioration of anthropometric and metabolic parameters associated with becoming menopausal and with aging.


1. Behnke, A. R., and J. H. Wilmore. Evaluation and regulation of body build and composition. Englewood Cliffs, NJ: Prentice-Hall, 1974, pp. 20–37.
2. Bouchard, C., A. Tremblay, C. Leblanc, G. Lortie, R. Savard, and G. Thériault. A method to assess energy expenditure in children and adults. Am. J. Clin. Nutr. 37:461–467, 1983.
3. Burstein, M., and J. Samaille. [On a rapid determination of the cholesterol bound to the serum alpha- and beta-lipoproteins]. Clin. Chim. Acta. 5:609, 1960.
4. DeFronzo, R. A., J.D. Tobin, and R. Andres. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am. J. Physiol. 237:E214–E223, 1979.
5. Desbuquois, B., and G.D. Aurbach. Use of polyethylene glycol to separate free and antibody-bound peptide hormones in radioimmunoassays. J. Clin. Endocrinol. Metab. 33:732–738, 1971.
6. DiPietro, L. Physical activity in the prevention of obesity: current evidence and research issues. Med. Sci. Sports Exerc. 31:S542–S546, 1999.
7. Duncan, G. E., M. G. Perri, D. W. Theriaque, A. D. Hutson, R. H. Eckel, and P. W. Stacpoole. Exercise training, without weight loss, increases insulin sensitivity and postheparin plasma lipase activity in previously sedentary adults. Diabetes Care. 26:557–562, 2003.
8. Ferland, M., J. P. Després, A. Tremblay, et al. Assessment of adipose tissue distribution by computed axial tomography in obese women: association with body density and anthropometric measurements. Br. J. Nutr. 61:139–148, 1989.
9. Gavin, J. R., G. M. Alberti, M. B. Davidson, and R. A. DeFronzo. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 20:1183–1197, 1997.
10. Gidez, L. I., G. J. Miller, M. Burstein, S. Slagle, and H. A. Eder. Separation and quantitation of subclasses of human plasma HDLs by a simple precipitation procedure. J. Lipid Res. 23:1206–1223, 1982.
11. Havel, R. J., H. A. Eder, and J. H. Bragdon. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest. 34:1345–1353, 1955.
12. Heart and Stroke Foundation of Canada. The Growing Burden of Heart Disease and Stroke in Canada 2003. Ottawa, Canada, 2003, pp. 20, 54.
13. Heding, L. G. Radioimmunological determination of human C-peptide in serum. Diabetologia. 11:541–548, 1975.
14. Hernandez-Ono, A., G. Monter-Carreola, J. Zamora-Gonzalez, et al. Association of visceral fat with coronary risk factors in a population-based sample of postmenopausal women. Int. J. Obes. Relat. Metab. Disord. 26:33–39, 2002.
15. Hu, F. B., M. J. Stampfer, C. Solomon, et al. Physical activity and risk for cardiovascular events in diabetic women. Ann. Intern. Med. 134:96–105, 2001.
16. Hunter, G. R., T. Kekes-Szabo, M. S. Treuth, M. J. Williams, M. Goran, and C. Pichon. Intra-abdominal adipose tissue, physical activity and cardiovascular risk in pre- and post-menopausal women. Int. J. Obes. Relat. Metab. Disord. 20:860–865, 1996.
17. Kanaley, J. A., C. Sames, L. Swisher, et al. Abdominal fat distribution in pre- and postmenopausal women: the impact of physical activity, age, and menopausal status. Metabolism. 50:976–982, 2001.
18. Knowler, W. C., E. Barrett-Connor, S. E. Fowler, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N. Engl. J. Med. 346:393–403, 2002.
19. Kvist, H., L. Sjostrom, and U. Tylen. Adipose tissue volume determinations in women by computed tomography: technical considerations. Int. J. Obes. 10:53–67, 1986.
20. Lohman, T. G., A. F. Roche, and R. Martorell. Anthropometric Standardization Reference Manual. Champaign, IL: Human Kinetics Books, 1988, pp. 39–80.
21. Nolin, B., D. Prud’homme, G. Godin, and D. Hamel. Enquête québécoise sur l’Activité physique et la Santé 1998. Québec: Les Publications du Québec, 2002, pp. 21, 28.
22. Pascot, A., S. Lemieux, I. Lemieux, et al. Age-related increase in visceral adipose tissue and body fat and the metabolic risk profile of premenopausal women. Diabetes Care. 22:1471–1478, 1999.
23. Poehlman, E. T., M. J. Toth, P. A. Ades, and C. J. Rosen. Menopause-associated changes in plasma lipids, insulin-like growth factor I and blood pressure: a longitudinal study. Eur. J. Clin. Invest. 27:322–326, 1997.
24. Rebuffe-Scrive, M., J. Eldh, L. O. Hafstrom, and P. Bjorntorp. Metabolism of mammary, abdominal, and femoral adipocytes in women before and after menopause. Metabolism. 35:792–797, 1986.
25. Richterich, R., and H. Dauwalder. [Determination of plasma glucose by hexokinase-glucose-6-phosphate dehydrogenase method]. Schweiz Med. Wochenschr. 101:615–618, 1971.
26. Roden, M., T. B. Price, G. Perseghin, et al. Mechanism of free fatty acid-induced insulin resistance in humans. J. Clin. Invest. 97:2859–2865, 1996.
27. Samaras, K., P. J. Kelly, M. N. Chiano, T. D. Spector, and L. V. Campbell. Genetic and environmental influences on total-body and central abdominal fat: the effect of physical activity in female twins. Ann. Intern. Med. 130:873–882, 1999.
28. Siri, W. E. The gross composition of the body. Adv. Biol. Med. Phys. 4:239–280, 1956.
29. Sjöström L., H. Kvist, A. Cederblad, and U. Tylen. Determination of total adipose tissue and body fat in women by computed tomography, 40K, and tritium. Am. J. Physiol. 250:E736–E745, 1986.
30. Stevenson, E. T., K. P. Davy, P. P. Jones, C. A. Desouza, and D. R. Seals. Blood pressure risks factors in healthy postmenopausal women: physical activity and hormone replacement. J. Appl. Physiol. 82:652–660, 1997.


©2005The American College of Sports Medicine