The relationship between lean mass, muscle strength and physical ability in independent healthy elderly women from the community

Abstract

Background/Objectives

The association between muscle mass, strength and physical performance has been established in the elderly with co-morbidities. In this study, lean and fat mass, bone mineral density, knee extension and flexion strength and physical ability tests in healthy independent elderly women were investigated. Main determinants of lean mass, strength and physical ability were determined searching for predictors of healthy aging.

Methods

A total of 100 healthy women aged ≥ 65 years considered independent and active were invited. Bone mass and body composition were assessed by DXA. The strength of the lower limb was assessed by isokinetic dynamometry, and physical ability was measured by: Timed Up and Go (TUG), Berg Balance Test (BBT) and Dynamic Gait Index (DGI).

Results

Women were on average 70.8±4.92 years old, had BMI of 27.38±5.11 kg/m2 and fat mass of 26.96±9.62 kg or 40.65±8.06%. Total lean mass and appendicular lean mass (ALM) were 35.38±4.83 kg and 15.32±2.26 kg, respectively, while relative skeletal mass index (RSMI) was 6.51±0.77 kg/m2. Age did not correlate significantly with ALM. Age and ALM were the main determinants of the strength of the lower limb (p<0.001) while age and strength of the lower limb were significantly associated with the performance on the physical tests (p<0.001).

Conclusion

Age has a negative impact on the strength and the physical performance in independent healthy women without co-morbidities. Physical ability tests are positively influenced by the strength of the lower limb. These relationships suggest that muscle strength should be the parameter to be prioritized when preparing for healthy aging.

Introduction

The proportion of people aged 60 and older is growing faster than any other age group (1, 2). Aging is accompanied by significant changes in body composition. Studies have shown that fat mass increases while lean body mass decreases with age, even when body weight and physical activity are maintained constant (3, 4). Decrements found in muscle (sarcopenia) and fat mass have also been related to significant reduction in physical capacity (5, 6).

Sarcopenia prevalence has been reported in different studies to range from 10 to 40% in older people (7). While loss of muscle mass and strength can lead to increased incidence of disability among the elderly and might affect mortality (8), is not yet well established whether any gain in these parameters eventually translates into an improvement in general health status (9). Similarly, the relationship between muscular strength and physical ability does not appear to be linear and the association between these parameters and motor function is still unclear (10).

The isolated effect of age on muscle mass, strength and performance is still controversial since most studies on the issue have included individuals with multiple co-morbidities and that could have an impact on the interpretation of their results. There is still controversy regarding the importance of appendicular lean mass for the determination of the physical capacity of older women. Other parameters beyond muscle mass seem to influence the impact of the strength of the lower limbs on physical capacity. Our working hypothesis is that the physical ability is determined, at least partially, by the strength of the lower limbs, which might be modulated by the amount of muscle mass and that age is the main modulator of those relationships. The association between muscle mass, muscle strength and physical functioning in independent healthy older women was evaluated in an attempt to determine the main determinants for successful aging in a group of individual with no known co-morbidity.

Methods

A convenience sample of women aged 65 years or older from the community was invited to participate in the present study. Women were selected between caregivers and relatives of patients seen at the Outpatient Clinics in the Federal University of São Paulo/Escola Paulista de Medicina or employees working at the Institution. Volunteers answered a detailed questionnaire designed to determine their eligibility for the study. A total of 815 women were interviewed and 100 healthy women met the inclusion criteria and volunteered to participate. Clinical assessment was performed to report diseases and regular medications, as well as to assess Barthel index (activities of daily living, ADL), Lawton scale (instrumental activities of daily living, I-ADL); IPAQ (assessment of physical activity) and quality of life questionnaire (SF-36) (11).

The UNIFESP's Ethics and Research committee approved the work, and all participants signed an informed consent form before any study procedure.

Inclusion Criteria

The inclusion criteria were: women from 65 years and older, not institutionalized, considered independent to perform activities of daily living (Barthel index ≥ 75) (12) and instrumental activities of daily living (Lawton scale score ≥ 5) (13) and physically active according to the International Physical Activity Questionnaire criteria (IPAQ score ≥ 150 minutes/ week) (14).

Exclusion Criteria

Women with hypo or hyperthyroidism or dyslipidemia without regular medication, current or prior corticosteroid treatment (for at least 3 months), autoimmune disorders, cardiovascular disease, diabetes mellitus, neurological or musculoskeletal disease or medication that interferes with ambulation (including sedatives and antidepressant drugs), malignancies, use of sex steroids and renal or hepatic impairment were excluded from the present study. Participants were also excluded if they had a systolic blood pressure greater than 199 mmHg or diastolic blood pressure greater than 110 mmHg or significant pain in the knees.

Physical ability tests

A translated version of the dinamic gait index (DGI) validated for our culture and language was used in the present study (15, 16). The test evaluates the individual's ability to modify gait in response to changing demands during certain physical tasks. Participants received instructions for each item and a demonstration, if necessary. Two opportunities were offered to complete each task. The Timed Up and Go Test (TUG) was also performed to evaluate the participants’ mobility and balance (17). The researcher explained verbally about the test procedure and the execution was carried out in a usual self-selected speed, safely. The time was recorded with a stopwatch accurate to 0.01s. Three attempts were performed, and the best value was recorded. Physical capacity was also assessed by using the Brazilian version of the Berg Balance Test (BBT) (18). The participants received verbal instructions and a demonstration of the items was performed. A maximum of two opportunities was given for completion of tasks.

Bone densitometry and body composition analyses

Bone densitometry was performed using dual-emission X-ray (DXA) densitometer (DPX MD +, GE - Lunar Radiation Corporation, Madison, WI, USA). BMD at the proximal femur and lumbar spine (L1-L4) was measured using a standardized protocol with daily calibration as instructed by the manufacturer. Body composition analyses were performed with the same DXA method. The following body composition parameters were assessed: total lean mass (TLM), appendicular lean mass (ALM) obtained by the sum of lean mass of arms and legs, total body fat (BF) in kilograms and percentage, as well as the percentage of body fat (BF%). From the raw data for ALM, relative skeletal muscle index (RSMI) was calculated (RSMI = appendicular lean mass/height²) (19). The precision of DXA was evaluated as coefficient of variation (CV), as recommended by the International Society of Clinical Densitometry (ISCD) (20). Precision expressed as the root mean square of the CV for body fat in percentage, fat in grams, BMD, Bone Mineral Content, TLM and ALM were 1.62%, 1.53%, 0.67%, 1.72%, 1.14% and 1.64%, respectively. Osteoporosis and osteopenia were defined according to the WHO criteria (21) while sarcopenia was defined as a RSMI lower than 5.45 for women (19, 22).

Isokinetic dynamometry

Isokinetic dynamometer (Cybex 125 dynamometer AP, Chattanooga, TN) was used to evaluate the strength of the lower limbs. Peak torque (Nm) of both lower limbs measured separately for the movements of extension and flexion of the knee at 60 ° per second were recorded with the participant seated.

Participants were instructed to push and pull the lever as hard as possible until the conclusion of the test. Test procedures began with a five minutes warm-up on a cycle ergometer. Immediately after the warm-up, five submaximal replicates were completed at 60° per second for the recognition of the apparatus. Three attempts of five repetitions with verbal encouragement were performed on each leg and the best value of the three was recorded. The strength was calculated as the average maximum torque output (Nm) in the three attempts.

Statistical Analysis

A descriptive analysis was performed and the study variables were expressed as mean, standard deviation, minimum and maximum values. The Komolgorov-Smirnov test was used to determine whether the parameters had normal distribution.

Correlations between anthropometric parameters, BMD, lean mass, fat mass, physical tests and strength were investigated by Spearman and Pearson's correlation tests. Variables with significant correlation were used for multiple linear regression models to identify the main outcomes in the study: ALM, knee extension and flexion strength and the physical tests (DGI, TUG and BBT).

P values lower than 0.05 were considered significant. Data were analyzed using SPSS for Windows version 17.

Results

Anthropometric characteristics, BMD, body composition parameters, physical ability performance and lower limb strength are shown in Table 1. The average relative skeletal mass index (RSMI) for the population was 6.51 kg/m2 and about 13% of the participants were considered sarcopenic (19, 22). According to the WHO criteria for the diagnosis of osteoporosis (21), the prevalence of osteoporosis and osteopenia was 26% and 53%, respectively. Nonetheless, physical ability tests confirmed that the study population consisted of independent and physically active women.

Table 1.

Anthropometric parameters, bone mineral density (BMD), body composition parameters, physical ability tests and lower limb strength in healthy older women from the community. N=100

Parameters	Mean	SD	Minimum	Maximum
Age (years)	70.76	4.92	65	85
Height (m)	1.53	0.06	1.36	1.68
Weight (kg)	64.37	13.19	35.10	98.70
BMI (kg/m2)	27.38	5.11	16.40	41.76
Fat mass (kg)	26.96	9.62	5.20	53.58
Fat mass (%)	40.65	8.06	14.09	56.52
Total lean mass (kg)	35.38	4.83	26.75	48.36
Appendicular lean mass (kg)	15.32	2.26	10.48	39.07
skeletal lean mass index (kg/m2)	6.51	0.77	4.82	8.38
spine BMD (g/cm2)	0.973	0.162	0.714	1.552
femur BMD (g/cm2)	0.822	0.111	0.573	1.241
time Up Go test (seconds)	9.08	1.30	7.00	13.00
Berg Balance test (0-56)	54.23	1.86	48.00	56.00
dynamic Gait index (0-24)	22.28	1.75	17.00	24.00
Knee extension strength (Nm)	87.80	19.43	43.00	139.00
Knee flexion strength (Nhi)	45.73	13.79	17.00	72.00

Physical ability tests (TUG, BBT and DGI) correlated significantly with age (r=0.28-0.14; p<0.001) and knee extension and flexion strength (r=0.25-0.31; p<0.05). There was a significant positive correlation between age and TUG (r=0.28; p<0.001). Likewise, TUG test was negatively correlated with knee extension and flexion strength (r=-0.27, p<0.001; r=-0.31, p<0.05), suggesting that the strength of the lower limb is an important determinant for the performance in this test. Similar correlations were also observed for the BBT with age and knee extension and flexion strength. As seen for the TUG test and BBT, knee extension and flexion strength also correlated significantly with the performance on DGI (r=0.28; p<0.05; r=0.31; p<0.001). The performance in physical tests did not correlate significantly with ALM.

ALM was significantly associated with knee extension and flexion strength (r=0.50-0.44; p<0.001) and bone mass (r=0.23-0.43; p<0.001). Likewise, knee extension and flexion strength correlated negatively with age (r=-0.27; p<0.001; r=-0.22; p<0.05) while was positively associated with BMD values at the femur (r=0.24-0.26; p<0.05).

Table 2 presents multiple linear regression equations for the dependent variables: ALM, knee extension strength (KES), knee flexion strength (KFS) and physical ability tests TUG, BBT and DGI. Age, weight, height, and KES were the main determinants of ALM. According to the regression model presented for ALM, variations in age, weight, height and KES explain about 72% of the variation in ALM. Note that for each increase of one unit in KES (Nm) there is an increase of 0.02 (Kg) in ALM.

Table 2.

Multiple linear regression models for appendicular lean mass (ALM), knee extension strength (KES) and flexion of the knee (KFS) and performance in the timed up and go test (TUG), berg balance test (BBT) and dynamic gait index (DGI) in healthy older women from the community N=100

	Multiple linear regression	Adjusted R2
ALM	-12.28 + 0.03 x age + 0.09 x weight + 11.38 x height + 0.02 x sKE	0.72
kes	32.06 - 1.15 x age + 0.02 x weight + 52.17 x height + 3.67 x ALM	0.35
KFs	21.85 - 0.78 x age - 0.05 x weight + 28.06 x height + 2.65 x ALM	0.28
tUG	9.50 + 0.08 x age + 0.02 x weight - 4.08 x height - 0.01 x sKE	0.18
BBt	59.13 - 0.11 x age - 0.03 x weight + 2.56 x height + 0.01 x sKE	0.13
DGI	20.31 - 0.03 x age - 0.02 x weight + 2.38 x height + 0.02 x sKE	0.10

Age, weight, height and ALM were important determinants of the strength of the lower limb measured by isokinetic dynamometry. In the regression model for the knee extension strength (KES), variations in age, weight, height and ALM explain about 35% of the variations of the KES. Each one-unit increase in ALM (Kg) is associated with an increase of 3.67 (Nm) in the KES. Similar associations were seen for the knee flexion strength (KFS). Variations in age, weight, height and ALM explain about 28% of the variation in KFS. Each one-unit increase in the amount of ALM (Kg) is associated with increments of about 2.65 (Nm) in the KFS.

Multiple linear regression equations constructed to investigate the determinants of the performance on physical ability tests demonstrated that besides age, weight and height, the strength of the lower limb has a significant impact on the performance in functional tests. In all the physical ability tests evaluated (TUG, BBT and DGI), age, and weight were negatively associated with general performance on the test, while the strength of the lower limb (KES and KFS) showed a positive association with physical ability. Each one-unit increase in KES (Nm) was associated with 0.01 -second improvement in the individual's performance on the TUG test. As seen for the previous models, variation in age, weight, height and KES explains about 18% of the variation in the TUG test. Each one-unit increase in KES was associated with improvements in BBT and DGI (points) values of 0.01 and 0.02, respectively. Variation in age, weight and height of the participants, as well as variation in the KES explains about 13% and 1 0% of the variation observed in BBT and DGI performance, respectively.

Discussion

Our results demonstrated that age is the main determinant of the strength of the lower limb, femoral neck BMD and the performance in physical ability tests. The study was able to establish this relationship in a group of women with no known co-morbidities. As expected, older women had significantly less strength at the knee (extension and flexion), lower femoral neck BMD and lower physical ability levels than younger women. Having a study population consisted only of healthy individual we were able to establish the effect of age itself on muscle mass, strength and performance.

ALM, the strength of the lower limb and physical ability are significantly associated in healthy older women. While age, weight and height were determinants for ALM, lean mass was significantly associated with the strength of the lower limb (strength of the extension and flexion of the knee). In turn, the strength of the lower limb was ultimately a major determinant of the performance on physical ability tests. To our knowledge, this is the first study to demonstrate such associations in a healthy successful group of older women and agrees with our working hypothesis.

Evaluating older individuals in New Mexico, Baumgartner and colleagues have found sarcopenia prevalence of around 13-24%, a rate similar to that reported in our study (19). It is interesting to observe that even with a high prevalence of sarcopenia in our sample, all of the women were considered independent and had good performance in daily living activities and physical tests. These results reinforce the idea that using only lean mass as a criterion to define sarcopenia may not be suitable for all scenarios since the relation between muscle mass and physical performance is not linear.

As suggested in this and other studies, it is likely that there is a non-linear relationship between muscle mass and performance in tests of physical ability. It has been shown that the strength of the legs, and not the ALM, is associated with physical performance (23).

ALM shows no linearity of gain or loss in relation to muscle strength, as previously shown (9). Besides the lack of linearity, these studies have shown that as we age, the decline in muscle strength is greater than the correspondent decline in ALM. Even when no changes are observed in the amount of ALM, the loss of muscle strength that occurs with aging still occurs even in healthy active women. Other components not evaluated in this study (nutritional status, cardiovascular fitness and muscle training) could probably modulate and influence the relationship between muscle mass and strength. On the other hand, recent cross-sectional work has demonstrated that vitamin D intake did not correlate with muscle mass (24).

The strength of the lower limb measured in our population was similar to that reported for the American women (9). Recent data from our group has shown that lean mass assessed by DXA in Brazilian women is consistently lower than that seen for the American women in all ethnic groups (25). These findings corroborate the notion that factors other than muscle mass modulate and influence the strength of the lower limb.

Knee extension and flexion strength was associated with age, body weight, height and ALM and agree with previous studies (26). Besides the negative association between strength and age, our study also noted that knee extension and flexion strength was significantly associated with the performance on physical tests.

Since no correlation was found between ALM and physical tests, the strength of the lower limb seems to have a more important role than muscle mass in physical ability in active healthy older women. Aiming to maintain the physical ability, preserving or gaining muscle strength appears to have greater importance than only preserving muscle mass. In healthy active women in our sample there was a significant association between age and muscle strength, even without significant effect of age on muscle mass. Taking into consideration the cross-sectional design of the present study, these findings also suggest that the maintenance of muscle mass associated with a level of activity of daily living that ranks those women as active does not ensure the maintenance of muscle strength and physical performance. It is possible that training, conditioning and specific exercises (resistance training) may contribute more significantly to the preservation of strength during aging.

Increased fat mass has been associated with the prevalence of osteoporosis and osteopenia, regardless of body weight, physical activity level and age (27). in the present study, we observed a positive correlation between femoral BMD, strength of the lower limb and weight. it is likely that the strength of the flexion and extension of the spine, not measured in this study, also shows association with spine BMD.

some limitations of this study need to be pointed out. The cross-sectional nature of the sample and the fact that it consisted exclusively of healthy and active women restricts extrapolation of our findings to other populations and for men. The absence of data on the nutritional status, vitamin D status, level of physical fitness and muscle training, in addition to biomarkers (28) should also be remembered. other important limitation deals with the fact that DXA and isokinetic dynamometry, in spite of being the gold standard methods for measuring body composition and muscle strength in the clinical scenario, are not totally accurate.

Age negatively influenced muscle strength in a group of healthy active women. The performance on tests of physical ability was also negatively influenced by age and positively affected by the strength of the lower limb. Lean mass is primarily determined by body size, while strength is ultimate and positively influenced by the amount of muscle mass. The performance in physical tests seems to be more affected by the strength of the lower limb than by muscle mass. These relationships suggest that the strength should be the parameter to be prioritized in assessing the physical ability in the elderly and muscle mass could not be used as a sole parameter to define sarcopenia. Healthy aging will require the adoption of strategies aimed at preserving muscle strength to maintain physical ability and ensure independence.