Expert working groups1,2 have recommended older adult muscle health is evaluated in terms of mass, strength and functional performance. Laboratory measures of muscle mass, strength and quality which are considered to be below ‘normal’ are associated with functional impairment in the elderly.3–5 Low relative muscle mass, knee extensor strength and muscle quality (strength per unit tissue) have been associated with self-reported and objectively measured functional limitation in older adults.3–7 These associations have been studied extensively in older adults >65y6,8–10 and as a result tend to use simple tasks related to activities of daily living (ADL) such as gait speed tests <6m.11–15

However, it is well documented that age-related change in muscle characteristics which precede functional limitation become noticeably different from a young adult aged ∼50y.16,17 In contrast to age groups >65y, comparatively little is known about the time course and transition to functional limitation after the age of ∼50y. We previously reported age-related difference in lean tissue mass (LTM), muscle strength and quality in healthy older women.18 Furthermore, our research group19 and others20 identified short gait speed tests (<10m) as unsuitable measures of functional performance for use in healthy older adults. This is primarily due to short performance tests inability to detect age-related difference where expected in healthy adults. Instead, we proposed that extended gait speed and chair rise tests are used due to their greater construct validity in detecting age-related difference and greater association with maximal voluntary knee extensor strength.19

Ideally, laboratory measures of muscle mass, strength and quality which can distinguish gradations of capacity with aging would also reflect functional performance. Lower body strength is thought to undergo a preferential age-related decline relative to the upper body. Specifically, quadriceps strength declines at a greater rate than grip strength in healthy volunteers.21 This preferential decline is thought to be due to a change in activity patterns22,23 and is greatest in women.24 Hicks et al.,11 have suggested that establishing cut points from strength at one time point is predictive of future mobility decline in a population based sample of older (∼73y) men and women. Researchers have recently12,25 classified healthy older (65–84y) adults into tertiles based on body composition, muscle function and functional performance. This approach may be more appropriate for attempting to distinguish gradations of capacity in healthy adults whereby cut-points below ‘normal’ in muscle health may not be able to identify individuals of lower physical capability.24

At present, there are two indices of lower extremity muscle function most commonly reported in the literature. The first is knee extensor strength which is normally assessed by a commercially available dynamometer and the second is muscle quality normally expressed as knee extensor strength per unit muscle or LTM mass. Both indices have been associated with functional performance in older (>65y) adults. Hairi et al.5 reported that muscle quality had a stronger association than muscle mass with both self-reported and objectively measured physical disability in community dwelling older (≥70y) men (n=1612). However, the authors also identified muscle strength (grip and quadriceps strength) as having stronger associations with all outcome measures than either muscle mass or muscle quality. Similarly, Visser et al.26 reported lower muscle strength and greater fatty infiltration as stronger predictors of self-reported mobility limitation than muscle mass in well-functioning older (70–79y) adults (n=2631). More recently, Hayashida et al.27 reported knee extension strength but not appendicular LTM to be associated with maximal gait speed (11m) in 318 community dwelling Japanese men and women (≥65y).

Whilst muscle strength can be determined using a variety of cost effective techniques which are often portable, muscle quality requires imaging technology, often at significant time, financial and training cost. The association between both indices and functional performance has received considerably less attention in adults <70y of age. Furthermore, it is not yet known whether one index is superior to the other in distinguishing gradations of capacity in healthy middle aged adults.

The aim of this study was to examine the association between upper leg LTM, knee extensor strength, muscle quality and functional performance in healthy 50–70y women. A secondary aim of this investigation was to determine the utility of tertiles of muscle strength and muscle quality to distinguish gradations of functional capacity in healthy 50–70y women.

Physical characteristics, body composition and functional performance of 50–70y women are displayed in Table 1. Functional performance classified per tertile of muscle strength and muscle quality are displayed in Table 2. Classified by muscle strength or muscle quality, those in the highest tertile (T1) had a higher functional performance across all tests than those in lowest tertile (T3) (p<0.05). Classified by tertile of muscle strength extended gait speed was different between those in the highest tertile, middle tertile and lowest tertile. Classified by tertile of muscle quality, those in the lowest tertile were different to those in the middle and highest tertile for extended gait speed and time taken to complete 5 chair rises. Those in the middle and highest tertile of muscle quality did not differ in functional capability.

There was no association between upper leg LTM and functional performance (r=0.007–0.061, p>0.05). Muscle strength demonstrated a moderate association with extended gait speed (Fig. 1) and the number of chair rises undertaken in 30s and a weak association with the time taken to complete 5 chair rises (Table 3). Muscle strength was associated with all functional performance measures in the highest tertile and with gait speed and the number of chair rises completed in 30s in the lowest tertile but was not associated with functional performance in the middle tertile.

Figure 1.

The association between knee extensor torque and 900m gait speed in healthy older women.

(0.08MB).

Muscle quality demonstrated a moderate association with extended gait speed and a weak association with both chair rise tests. Muscle quality was associated with the time taken to complete 5 chair rises in the highest tertile but was not associated with any other functional performance measure in any other tertile. Compared to muscle quality, muscle strength explained a greater proportion of the variance in extended gait speed (36% vs. 16%), the number of chair rises completed in 30s (22% vs. 15%) and the time taken to complete 5 chair rises (13% vs. 11%).

Compared to muscle quality, knee extensor strength per kg body mass had a stronger association with functional performance and could better differentiate between those of high, intermediate and lower functional performance in healthy 50–70y women.

Interestingly, classified by tertile of muscle strength or muscle quality those in the highest tertile could be differentiated from those in the lowest tertile in terms of functional performance. It became more challenging to separate those in the middle tertile from those in the highest or lowest; in fact, only gait speed classified by muscle strength could provide this differentiation. Furthermore, knee extensor strength was associated with functional performance in the highest and lowest tertiles of muscle strength but muscle quality was generally not associated with functional performance in any tertile. The absence of an association between strength and functional performance in the middle tertile perhaps suggests a non-linear relationship between strength and functional performance. Buchner et al.40 reported an association between strength and gait speed (15.2m) in weaker participants but not in stronger participants. The authors suggested that small changes in physiological capacity of frail older adults, may lead to large changes in functional performance whereas small changes in physiological capacity of stronger adults, may lead to little or no change in functional performance. Our results suggest that this may be the case in the lowest tertile but also that small changes in strength for healthy adults in the highest tertile may lead to significant changes in functional performance. This has recently been demonstrated in well-trained distance runners who experienced an increase in running economy as a result of a strength training intervention.41 Buchner et al.38 may not have seen this effect at the higher end due to using a population based sample that allowed for participants with musculoskeletal conditions. These explanations whilst plausible need to be interpreted cognisant of the sample size used to derive the associations in each tertile (n=42–43).

As reported by Visser et al.,26 Hairi et al.5 and Hayashida et al.27 in older (≥65y) adults, we report muscle strength and quality to have a greater association with functional performance than muscle mass in healthy 50–70y women. However, compared to muscle quality, we report knee extensor strength corrected for body mass to have a stronger association with all functional performance measures. This suggestion has been supported by Hairi et al.5 when investigating the association between knee extensor strength and 6m gait speed and by Vilaça et al.25 using the 6min walk test in older (≥65y) adults. In our study, it is perhaps expected that as the difficulty of the test increases towards maximum so does the association between performance of the test and maximal voluntary knee extensor strength. Strength per unit LTM is considered to be an accurate representation of a muscles force generating capacity. The lower association between muscle quality and functional performance is perhaps because the upper leg LTM does not represent the body mass required to be elevated from a chair or accelerated during a gait speed test. Furthermore, we previously reported no difference in LTM between women in the 6th and 7th decade of life19 which may suggest that changes in muscle quality are a function of the underlying strength rather than mass.25,26 However, it must be acknowledged that we have used DXA to estimate LTM which has been shown to underestimate age-related change in skeletal mass compared to gold standard methods such as MRI and CT.42

Previously, Buchner et al.40 have reported lower extremity strength to account for ∼17–20% of short (<16m) gait speed tests in older adults (50–96y). To the authors knowledge, the 900m gait speed test is the first gait speed test reported in which >30% of the variance in performance among healthy older women can be explained by knee extensor strength. Our test which requires a similar time (∼6–7min) to complete as the 6min walk test is perhaps a step in the right direction toward the call for the generation of maximal gait speed reference values in healthy adults.25 The fact that lower extremity strength has a strong association with a maximal gait speed test which has a range of 1.3–3.9m/s is encouraging. This suggests that the relative muscular effort for those with the mean gait speed (2.3m/s) is considerably less when performing a habitual gait speed (∼1.4m/s) for extended periods of time. The relative effort required to perform tasks related to ADL could be a key determinant as to how long quality of live and independent living are preserved.

Our study makes a contribution to knowledge about the association between laboratory measures of LTM, muscle strength, muscle quality and functional limitation in healthy late middle aged20 adults who are not yet frail. The applicability of our study is limited by our small (n=128) convenience sample of healthy older women recruited from the University campus community and surrounding area. This may be especially important as education and socioeconomic status have been reported to influence the health status of a population.43 The mean habitual gait speed (1.4m/s) and dietary protein intake (1.2gkg−1BWday−1) of participants in the University of Limerick Healthy Aging Study demonstrate the relative physical and nutritional health of this cohort compared to more representative cohorts (dietary protein intake 0.8–1.1kg−1BWday−1 and habitual gait speed 0.8m/s)44,45 which may introduce a health bias. Furthermore, we did not assess cognitive function, stage of the menopause or control for habitual physical activity and as such it is unknown how these co-founding variables may have affected the associations reported.

Although 900m gait speed had the strongest association with muscle strength and quality, the self-selected mode of maximal gait speed which included walking, jogging or running may have over or underestimated our gait speed and therefore influenced the strength of the associations reported. Furthermore, endurance performance is also dependent on cardiorespiratory and muscle oxidative capacity which we did not control for. The associations reported for all muscle and functional performance parameters must be interpreted cognisant of the cross-sectional study design and therefore survival bias may have influenced the results.

The fact that strength measurement at a single time point has been shown to be predictive of mobility decline and11 that the lower limbs are most affected21,24 should make lower limb strength assessment a priority for researchers generating reference values for muscle health and clinicians in practice. As Dulac et al.46 have recently reported using handgrip strength, we report knee extensor strength normalised to body mass as an index that could provide a cost effective and simplistic measure of muscle health which could be used in a clinical setting and can represent a significant proportion of the variance in functional capacity. This may be useful to clinicians attempting to assess the functional capability of older adults in the community. By contrast our results suggest that muscle quality assessment, requiring expensive imaging technology may be of lesser value in relation to distinguishing functional performance. Further work is required to determine whether these differences remain when using MRI or CT to determine skeletal muscle mass.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent was obtained from all individual participants included in the study.

The authors declare no conflicts of interest.