Effect of Sarcopenia on Cardiovascular Disease Risk Factors in Obese Postmenopausal Women

Authors


Research Centre on Aging, 1036 Belvédère Sud, Sherbrooke, Québec, Canada J1H 4C4. E-mail: isabelle.dionne@USherbrooke.ca

Abstract

Objective: To compare sarcopenic-obese and obese postmenopausal women for risk factors predisposing to cardiovascular disease (CVD) and determine whether there may be a relationship between muscle mass and metabolic risk in obese postmenopausal women.

Research Methods and Procedures: In this cross-sectional study, 22 healthy obese postmenopausal women (mean age, 66 ± 5 years; mean BMI, 27 ± 3 kg/m2) were divided into two groups matched for age (±2 years) and fat mass (FM) (±2%). Sarcopenia was defined as a muscle mass index of <14.30 kg fat-free mass (FFM)/m2 (which corresponds to 1 standard deviation below the values of a young reference population), and obesity was defined as an FM of >35% (which corresponds to the World Health Organization guidelines). FM, FFM (measured by DXA), daily energy expenditure (accelerometry), dietary intake (3-day dietary record), and blood biochemical analyses (lipid profile, insulin, glucose, and C-reactive protein) were obtained. Visceral fat mass (VFM) was calculated by the equation of Bertin, which estimates VFM from DXA measurements.

Results: Obese women had more FFM (p = 0.006), abdominal FM (p = 0.047), and VFM (p = 0.041) and a worse lipid profile [p = 0.040 for triglycerides; p = 0.004 for high-density lipoprotein (HDL); p = 0.026 for total cholesterol/HDL] than sarcopenic-obese postmenopausal women. Obese women also ingested significantly more animal (p = 0.001) and less vegetal proteins (p = 0.013), although both groups had a similar total protein intake (p = 0.967).

Discussion: Sarcopenia seems to be associated with lower risk factors predisposing to CVD in obese postmenopausal women. With the increase in the number of aging people, the health implications of being sarcopenic-obese merit more attention.

Introduction

Aging is associated with a decrease in muscle mass, also known as sarcopenia, and a deleterious increase in body fat (1). Moreover, it is recognized that women gain fat mass (FM)1 mainly in the abdominal area during menopause (2). The prevalence of obesity increased by 47% between 1991 and 1998 for women in the early menopausal years (ages 50 to 59 years) (3). Obesity is recognized among the risk factors predisposing women to cardiovascular disease (CVD). Abdominal FM accumulation is a major risk factor for insulin resistance, hypertension, atherosclerosis, and type 2 diabetes (4, 5).

It is noteworthy that not all obese individuals display a clustering of metabolic and cardiovascular risk factors. In fact, preliminary evidence suggests that metabolically healthy obese (MHO) persons may represent up to 20% of the obese population (4, 6). In this regard, Karelis et al. (7) showed that, despite similar levels of total body fatness, MHO persons have less CVD risk factors than obese subjects with the metabolic syndrome. The authors suggest that the difference may pertain to differences in adipocyte metabolism that are related to differences in the size of the adipocytes (7, 8).

However, one factor that was not taken into consideration in the study of Karelis et al. (7) is the impact of muscle mass on cardiovascular risk factors. Muscle mass may decline by ∼25% between the ages of 50 and 75 years (8), which translates into an atrophy or decrement in type II fibers and a tendency toward an increase or the maintenance of type I fibers (9). Hence, because type II fibers are recognized to be glycolytic and insulin-resistant (10), their decrease in number and size may explain how sarcopenia would positively alter glucose metabolism. Thus, it is reasonable to hypothesize that the presence of sarcopenia could partly explain the observed differences between MHO and obese individuals with a metabolic syndrome. In accordance, previous findings confirmed that MHO persons have a greater mass of lean tissue from diverse compartments (11) as well as more visceral fat (12). Altogether, these findings suggest that a greater lean body mass may be a potential modulator of cardiovascular risk factors and insulin resistance in sedentary obese postmenopausal women. These issues are still obscure and obviously need to be further examined.

The aim of this study was to compare risk factors predisposing to CVD in sarcopenic-obese and obese postmenopausal women and to determine whether there is a relationship between muscle mass and metabolic risk in obese postmenopausal women. We hypothesized that sarcopenic-obese individuals may have a lower risk of CVD compared with obese individuals.

Research Methods and Procedures

Subjects

Sixty postmenopausal women ages 55 to 75 years were recruited by means of advertisements in local newspapers. Although this could be considered as a drawback to our study, most clinical studies use a somewhat similar recruitment strategy. From the group of candidates, 11 obese women were matched to 11 sarcopenic-obese women for age (±2 years) and FM (±2%) to compare risk factors predisposing to CVD. In our study, sarcopenia was defined as a muscle mass index (MMI) of <14.30 kg fat-free mass (FFM)/m2, which corresponds to 1 standard deviation lower than the MMI of our reference sample. The participants included in our analyses also had an FM of >35% (determined by DXA), which is a better marker of obesity than BMI in aging people (13). To be included in the study, women had to meet the following criteria: healthy, having no major physical incapacity, not taking hormonal replacement therapy (HRT) (at the time of the study, women had never been on HRT or had been off HRT for at least 1 year), sedentary, weight-stable (±2 kg) for the past 6 months, non-smoker, moderate drinker (15 grams of alcohol/d maximum, the equivalent of one alcoholic beverage/d), taking no medication that could influence metabolism, and having an absence of menses for the past 12 months. A phone interview was conducted to screen for the aforementioned inclusion criteria. After the nature and goals of the study were thoroughly explained, the subjects provided written informed consent. All procedures were approved by the Ethics Committee of the Geriatric Institute of the University of Sherbrooke.

Study Procedures

After screening, subjects were invited for a visit to the Research Centre on Aging at the Geriatric Institute of the University of Sherbrooke. After their arrival, a 12-hour fasting blood sample was obtained, followed by breakfast, body composition measures, and instructions for the dietary record and accelerometry measurements. After this first visit, subjects completed a 3-day dietary record for the assessment of dietary intakes and carried an accelerometer for a 3-day period to measure daily energy expenditure (DEE).

Body Composition Measurements

Overall

Body weight was determined using an electronic scale accurate to ±0.2 kg (Seca707; Vogel & Halke GmbH, Hamburg, Germany). Height was measured using a tape measure affixed to the wall, with the subject in stocking feet. Determination of FM and FFM was assessed with the subject in a supine position, by the use of DXA (GE Prodigy Lunar, Madison, WI). In our laboratory, the coefficients of variation for repeated measures of FM and FFM in 10 adults (measured 1 week apart) are 0.9% and 0.4%, respectively (12). FFM is defined here as the mass of tissue representing soft tissue exclusively (mineral body mass excluded).

Visceral FM (VFM)

VFM was estimated using the equation developed and validated by Bertin et al. (14): (79.6 × {SD − [(TED − TID)/2] × TID}/H) − 149, where SD is the sagittal diameter (cm), TED is the transverse external diameter (cm), TID is the transverse internal diameter (cm), and H is the height (cm). This estimation of VFM by DXA gives a correlation of r = 0.94 (p < 0.0001) with computed tomographic scan data obtained in overweight women (14).

MMI

Sarcopenia was defined by the use of the MMI. This index was calculated as total FFM (kg)/height (m)2. We intentionally used total FFM in the calculation of MMI (15), because it has been demonstrated that trunk muscle is highly important to prevent the loss of functional capacity with aging (16).

Class I sarcopenia is defined as an MMI value of 1 to 2 standard deviations below the values for young adults from a reference population, whereas Class II sarcopenia represents an MMI of 2 standard deviations or more below the same value (15, 17). In our laboratory, women were considered Class I sarcopenic when their MMI values were <14.30 kg FFM/m2 and Class II sarcopenic when their MMI values were <12.72 kg FFM/m2. These criteria were established on the basis of a reference sample of 30 healthy women, ages 20 to 35 years (18), who were representative of the population and had an average BMI of 23.62 ± 2.11. In the present analysis, only Class I sarcopenic women were studied. Moreover, all women included in our analyses were obese (>35% total FM), as defined by Baumgartner et al. (13).

DEE

DEE was measured by the use of an accelerometer (Caltrac, Torrance, CA). The accelerometer was attached to the waist, according to the manufacturer's instructions. DEE was calculated based on the frequency and the velocity of the movements carried out during a given period. Participants carried the accelerometer for a period of 3 days that was representative of everyday life (19). The measurement of energy expenditure, which was carried out by the use of constant values according to sex, age, body weight, and height, includes both permanent work performed and peak loads (20). Accelerometry is generally considered to be representative of a general physical activity lifestyle, with no specific information about the type of activity performed (20), and it has been validated as a measure of DEE in older adults (21).

Dietary Intake

Each subject was instructed to maintain normal dietary habits throughout the period of data collection, as previously described (22). Subjects were provided with a 5-kg (11-lb) food scale and instructed on how to complete a 3-day dietary record. Diets were recorded during 3 consecutive days. It has been demonstrated that a 3-day dietary record is valid to estimate dietary intakes in older adults without cognitive impairments (22). Dietary analyses were completed by using Candat System software (version 6.0; Candat, London, Ontario, Canada) to determine daily intake of energy, protein, carbohydrate, and lipid. Analyses were performed with only nine pairs of subjects. Afterward, protein intake was subdivided based on its source: animal or vegetal. In addition, lipids were categorized as mono-, poly-, and unsaturated fatty acids.

Biochemical Analyses

Blood samples were obtained in the morning, after a 12-hour fast. Venipuncture was done while the participants were in a sitting position. Venous blood was withdrawn and placed in Vacutainer tubes (Becton-Dickinson, Franklin Lakes, NJ). The plasma lipid profile [high-density lipoprotein (HDL), low-density lipoprotein (LDL), total cholesterol (TC), TC/HDL ratio, and triglycerides (TG)], plasma glucose, and C-reactive protein were analyzed in the clinical laboratory of the Geriatric Institute, and plasma insulin was analyzed at the Sherbrooke University Hospital Center.

Insulin Sensitivity

Basal values for plasma insulin and glucose levels were used in the updated homeostasis model assessment (HOMA2) developed by Wallace et al. (23) to assess insulin sensitivity. HOMA2 has been validated to determine insulin sensitivity. A normal insulin sensitivity is represented as a value of 100% or above.

Statistical Analysis

Results are presented as means ± standard deviation. The Mann-Whitney test was used to compare groups for all body composition, biochemical, dietary intake, and energy expenditure variables. Moreover, vegetal and animal protein intakes were compared between groups by the use of an analysis of covariance using total energy intake as a covariable. Furthermore, a general linear model was used to verify the effect of sarcopenia on lipid profile with VFM and animal protein intake as covariables. p Values of ≤0.05 were considered statistically significant. Analyses were performed using the SPSS software program (version 11.0; SPSS, Inc., Chicago, IL).

Results

As shown in Table 1, the groups were similar with regard to age and total FM but significantly different for FFM and MMI. These results were to be expected, because the groups were matched for FM and age and divided on the basis of MMI status (sarcopenic vs. non-sarcopenic) (Table 2). It should be noted that, although subjects were defined as overweight based on the BMI criteria of the World Health Organization, they were considered obese because of their high percentage of FM (>35%) (13).

Table 1.  Body composition of study participants
VariableSarcopenic-obeseObesep*
  • FM, fat mass; VFM, visceral fat mass; MMI, muscle mass index; FFM, fat-free mass. Values are means ± standard deviation.

  • *

    p values were obtained by Mann-Whitney test.

  • Appendicular FFM = sum of leg and arm FFM.

Age (years)66.09 ± 5.1866.18 ± 5.910.974
Weight (kg)65.18 ± 9.6572.66 ± 7.650.040
BMI (kg/m2)25.52 ± 2.5528.66 ± 2.970.016
Total FM (kg)27.01 ± 6.7430.35 ± 5.900.193
Abdominal FM (kg)11.85 ± 3.2514.69 ± 3.030.041
Total FM (%)42.16 ± 5.7443.23 ± 4.830.450
Abdominal FM (%)42.02 ± 4.1943.56 ± 5.160.340
VFM (cm2)57.81 ± 23.8981.44 ± 26.750.034
MMI (kg FFM/m2)13.82 ± 0.4315.83 ± 1.010.000
Total FFM (kg)35.24 ± 2.9539.41 ± 3.420.007
Appendicular FFM (kg)15.61 ± 1.8217.67 ± 1.900.028
Table 2.  Energy and nutritional parameters
VariableSarcopenic-obeseObesep*
  • Values are means ± standard deviation.

  • *

    p values were obtained by Mann-Whitney test.

  • n = 9 pairs.

Daily energy expenditure (kcal/d)1726 ± 2931896 ± 2480.171
Total dietary intake (kcal/d)2563 ± 12492438 ± 6900.941
Total protein intake (g/d)97 ± 3397 ± 260.941
Total carbohydrate intake (g/d)315 ± 137272 ± 930.503
Total lipid intake (g/d)106 ± 70106 ± 400.710
Total animal protein intake (g/d)36 ± 1853 ± 180.001
Total vegetal protein intake (g/d)64 ± 1843 ± 160.013
Total saturated fatty acid (g/d)33 ± 1634 ± 140.882
Total monounsaturated fatty acid (g/d)33 ± 1437 ± 150.456
Total monounsaturated fatty acid (g/d)17 ± 917 ± 70.941

On the other hand, although groups were matched for total FM, we observed that the obese women had a greater BMI, abdominal FM, and VFM than the sarcopenic-obese women (Table 1). However, no difference between groups was found for DEE and for intake of energy, protein, lipid, carbohydrate, and saturated, monounsaturated, and polyunsaturated fatty acids. Nevertheless, when we examined more specifically the type of protein ingested, we found a significant difference between groups in animal and vegetal protein intake. In this regard, the sarcopenic-obese women ingested significantly more vegetal and less animal protein than the obese women (Table 2).

In analyzing biochemical parameters, we observed that the groups differed significantly with regard to TG, HDL, and the TC/HDL ratio. The obese women had higher TG and TC/HDL levels and a lower HDL level than the sarcopenic-obese women, although no difference was observed for other parameters, such as LDL, fasting insulin, glucose, insulin sensitivity index, and C-reactive protein (Table 3).

Table 3.  Lipid profile and insulin sensitivity index
VariableSarcopenic-obeseObesep*
  • TC, total cholesterol; LDL, low-density lipoprotein; TG, triglyceride; HDL, high-density lipoprotein; HOMA2, updated homeostasis model assessment; IS, insulin sensitivity (normal values ≥100); CRP, C-reactive protein. Values are means ± standard deviation.

  • *

    p values were obtained by Mann-Whitney test.

  • n = 9 pairs.

  • n = 8 pairs.

  • §

    n = 10 pairs.

TC (mM)5.73 ± 0.595.47 ± 0.750.631
LDL cholesterol (mM)3.30 ± 0.733.30 ± 0.590.780
TG (mM)1.02 ± 0.501.49 ± 0.430.049
HDL cholesterol (mM)1.93 ± 0.331.48 ± 0.240.001
TC/HDL (mM)3.02 ± 0.513.77 ± 0.780.041
Plasma fasting glucose (mM)§4.59 ± 0.414.86 ± 0.490.197
Plasma fasting insulin (pM)§36.09 ± 11.4650.40 ± 27.570.251
HOMA2 (% IS)§163 ± 52130 ± 520.185
CRP (mg/L)1.36 ± 1.801.82 ± 2.480.898

The groups also differed significantly with regard to VFM and animal protein intake. Therefore, we verified whether sarcopenia status still affects the lipid profile when the effect of VFM and animal protein intake are taken into account. In fact, we observed a significant effect of sarcopenia on TG (p = 0.046), HDL (p = 0.007), and TC/HDL ratio (p = 0.018) when VFM was used as a covariable. We also found a significant effect of sarcopenia on HDL (p = 0.008) and the TC/HDL ratio (p = 0.021) when animal protein intake was used as a covariable. Finally, we observed a significant effect of sarcopenia on TG (p = 0.032) and HDL (p = 0.043) when animal protein intake and VFM were used together as covariables. In light of these results, we can conclude that sarcopenia influenced lipid profile independently of VFM or animal protein intake.

Discussion

The aim of this study was to compare sarcopenic-obese and obese postmenopausal women for risk factors predisposing to CVD to determine whether there is a relationship between muscle mass and metabolic risks in obese postmenopausal women. We observed that the obese women had more FFM, abdominal FM, and VFM and a worse lipid profile than the sarcopenic-obese postmenopausal women. We also found that the obese women ingested significantly more animal and less vegetal proteins than sarcopenic-obese women.

Our results show that obese women have more risk factors predisposing to CVD than sarcopenic-obese postmenopausal women, which might be partly explained by a greater VFM. In fact, we found that obese women have a worse lipid profile than sarcopenic-obese postmenopausal women. These results were in line with the literature, as other studies have reported that excess trunk FM and VFM were associated with high plasma TG levels (24), low HDL levels (25), and CVD metabolic risks, morbidity, and mortality (26). In this sense, studies have shown that postmenopausal women with abdominal obesity and VFM carry higher risk factors predisposing to CVD than those without abdominal obesity (27, 28). Furthermore, it is recognized that these two factors (higher abdominal FM and worse lipid profile) largely contribute to an increase in atherosclerotic CVD (29). The reason that sarcopenic-obese women have less VFM compared with obese postmenopausal women is unclear and merits further attention. Nevertheless, although this difference between groups may partly explain why sarcopenic-obese women have less risk factors predisposing to CVD than obese postmenopausal women, the effect of sarcopenia persisted even when corrected for the effect of VFM, which implies that other factors are involved.

It is important to note that, in our study, VFM was estimated by the use of an indirect method [the predictive equation validated by Bertin et al. (14)] and not by a computed tomographic scan, which may be a drawback of the study. We suggest that further research examining the relationship between VFM and muscle mass use a direct measure of VFM; this might help highlight the link between muscle mass and visceral fat and their respective roles in the development of risk factors predisposing to CVD.

Interestingly, it must be noted that the type and quantity of protein ingested has the potential to lead to or exacerbate sarcopenia (30). It has been demonstrated that essential amino acids are responsible for anabolic stimulation in the muscle of healthy older people, and all essential amino acids are present in animal proteins (31). Our group showed that the quantity of animal protein intake was significantly correlated with muscle mass (32). In our study, we observed that the obese women ingested more animal and less vegetal proteins than the sarcopenic-obese postmenopausal women, which is in accordance with the findings of Lord et al. (32). On the other hand, these findings could also partly explain the more deteriorated lipid profile. Foods containing animal proteins are also richer in saturated fatty acids, which are known to increase blood cholesterol levels and risk factors predisposing to CVD (33). It is recognized that foods containing vegetal proteins have higher levels of mono- and polyunsaturated fatty acids, which are known to contribute to decreased LDL and increased HDL levels (33). Hence, a lower animal protein intake may lead to a decrease in plasma free fatty acids and consequently to a decrease in TG and an increase in HDL levels. However, in our study, no difference was observed between groups with respect to poly-, mono-, and saturated fatty acids. Moreover, the effect of sarcopenia on CVD remained significant even when we used animal protein intake as a covariable. Thus, the type of proteins and fatty acids does not seem to have influenced the difference between the sarcopenic-obese and obese postmenopausal women regarding CVD risk factors.

From a clinical standpoint, we observed that obese postmenopausal women do not have an increase in risk factors predisposing to CVD compared with sarcopenic-obese postmenopausal women, although their lipid profile is deteriorated to a greater extent. In fact, the HDL level (1.48) and TC/HDL ratio (3.77) in the obese group were borderline when compared with National Cholesterol Education Program normal values (34), as opposed to the sarcopenic-obese group. Thus, the obese group seemed to display a greater number of factors that place them at risk of developing hyperlipidemia and CVD, compared with the sarcopenic-obese postmenopausal women.

Our study was the first to show that sarcopenic-obese women have a metabolic profile closer to normal values than obese postmenopausal women, although we confirmed that this effect was not totally explained by VFM and animal protein intake. Nevertheless, the relationship between a low MMI and risk factors predisposing to CVD remains to be explained. We could speculate that the difference between obesity and sarcopenic obesity is derived from the muscle fiber type. It is recognized that type IIb fibers are positively correlated with BMI and are known to be insulin-resistant (10). However, sarcopenia is characterized by a decrease in type II fibers and a tendency toward an increase or the maintenance of type I fibers, which may favor a greater insulin sensitivity (9). Thus, sarcopenic-obese women could be less insulin-resistant than obese women. Unfortunately, our results were not in line with this assumption. In fact, we found obese women to have more abdominal FM and VFM and a worse lipid profile than sarcopenic-obese women, but no difference was found between the groups with regard to insulin resistance. We suggest that this finding could be explained by the fact that the method used to assess insulin sensitivity (HOMA2) is less sensitive than more direct methods, such as the glucose clamp, intravenous glucose tolerance test, or oral glucose tolerance test. In addition, the small sample size (11 pairs of subjects) could also contribute to the lack of significance between groups for insulin sensitivity. Thus, further research examining this aspect in a larger sample size would be interesting. Nevertheless, this study is the first to suggest a negative relationship between sarcopenia and risk factors predisposing to CVD in obese postmenopausal women.

Our results show that sarcopenia could be associated with a lower risk factor predisposing to CVD in obese postmenopausal women. Sarcopenia, which affects 50% of men and 30% of women age 80 years or older (35, 36), leads to other health complications, such as physical disabilities, functional limitations, falls, and dependency (17, 37). Therefore, with the increasing numbers of aging people, it is highly relevant to study more attentively the implications of sarcopenic obesity on various aspects of health.

In conclusion, our results indicate that sarcopenic-obese women have less risk factors predisposing to CVD compared with obese postmenopausal women, as they have better HDL levels, less VFM and abdominal FM, and lower TG levels and CT/HDL ratio. Nevertheless, further research is needed to better understand the mechanisms of the putative protective effect of sarcopenia on CVD as well as the potential link between muscle mass and visceral fat.

Acknowledgments

We thank the women who participated in the study. We thank Dr. Bertin for helping with the use of the predictive equation of VFM. This study was funded by the Research Centre on Aging of the Geriatric Institute of the University of Sherbrooke and the Canadian Institutes of Health Research. I.J.D., M.A-L., and E.D.B.G. are supported by the Canadian Institutes of Health Research. A.K. is supported by the Fond de la Recherche en Santé du Québec.

Footnotes

  • 1

    Nonstandard abbreviations: FM, fat mass; CVD, cardiovascular disease; MHO, metabolically healthy obese; MMI, muscle mass index; FFM, fat-free mass; HRT, hormonal replacement therapy; DEE, daily energy expenditure; VFM, visceral FM; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TC, total cholesterol; TG, triglycerides; HOMA2, updated homeostasis model assessment.

  • The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Ancillary