Association between activity energy expenditure and peak oxygen consumption with sarcopenia

Sarcopenia or the loss of muscle mass and function may occur in any moment of life, but old age is a risk factor and a condition in which its functional consequences are more striking. There are many pathogenic hypotheses to explain the loss of muscle mass. Among these, sedentariness is always mentioned, but its real association with sarcopenia both in young and old people has not been fully tested.

If sedentariness has a pathogenic role on sarcopenia, this factor should be equally important in young and old people (defined as subjects aged 60 years or more), and there should be an association between usual physical activity and loss of muscle mass. There are previous studies looking for this association. Park et al. found an association between year-long physical activity measured using pedometers and muscle mass in older people [1]. A further study by the same authors following older people for five years showed that the loss muscle mass in the study period was also associated with sedentariness determined using accelerometry [2]. There are several cross sectional studies that confirm this association [3].

An accurate method to measure usual physical activity is to determine activity energy expenditure. Doubly labeled water is the gold standard, however it is extremely expensive. An alternative is the use of actigraphy, whose accuracy increases when heart rate is also measured and is calibrated with individual heart rate/activity curves. There is a good concordance between doubly labeled water measurements and actigraphy with heart rate determination [4]. However some authors suggest that wearable devices lack accuracy [5].

The measurement of peak oxygen consumption by indirect calorimetry gives a precise estimation of the aerobic capacity of an individual. It is proportional to the usual physical activity and a direct indicator of fitness. Although it is influenced by age and genetic factors, usual physical activity is one of its main determinants [6]. Previous reports showed a direct association between muscle mass and peak oxygen consumption, speculating that a lower amount of metabolically active tissue would explain this association. However it is also possible to asseverate that a lower fitness derived from sedentariness, causes a reduction in muscle mass [7]. Therefore this parameter should provide an objective measurement that is inversely associated with long term sedentariness.

With the hypothesis that sedentariness is a main determinant of sarcopenia, the aim of this study was to evaluate whether daily physical activity and maximum oxygen consumption are determinants of muscle mass and function, regardless of age.

We analyzed data obtained from healthy volunteers participating in several studies on body composition, exercise capacity and aging. When the participants signed the informed consent to participate in the aforementioned studies, they allowed us to use their data in secondary analyses. Subjects living in the community aged 22 to 88 years were included. For all studies, we excluded participants with chronic debilitating diseases such as cancer, renal or cardiac failure or chronic infections. Individuals who could not reach the clinic by their own means, were unable to understand and sign the consent form or those who admitted abnormal consumption of alcohol or illicit drugs, also were excluded.

All participants had the following assessments performed and recorded:

Data from 192 assessments performed in participants aged from to 22 to 88 years (106 women) were analyzed. Table 1 shows clinical, body composition, exercise calorimetry and actigraphy features of participants, divided by gender and age range. Women and older people had lower fat free mass, handgrip strength, peak oxygen consumption and workload and activity energy expenditure. Young women reported a total physical activity in the physical activity questionnaire, which was significantly lower than the rest of the groups. The respiratory exchange ratio of 1.3 in all groups is an indication that real peak values were achieved during the exercise calorimetry. There was a significant correlation between peak oxygen uptake and activity energy expenditure measured by actigraphy, both in women and men (r = 0.53 and 0.37 respectively, p < 0.01). There was also a significant correlation between peak oxygen consumption and fat free mass in women and men (r = 032 and 0.62 respectively, p < 0.01). There was no association between total physical activity measured using the International Physical activity questionnaire and peak oxygen uptake or activity energy expenditure.

Table 2 compares men and women according to the presence of sarcopenia, defined using body composition, strength criteria or both. Peak oxygen consumption and work load and 72 h activity energy expenditure were lower in sarcopenic participants, whatever definition used. When sarcopenia was defined using only strength criteria, no differences in activity energy expenditure were observed between sarcopenic women and their non-sarcopenic counterparts.

Figures 1 and 2 show the regression models for appendicular fat free mass index and percentage of muscle mass, expressed as absolute values or z scores, for women and men. In general, the models had a higher r² in men than women. In the former, peak oxygen consumption was always a significant and independent predictor of muscle mass. Among women, beta coefficients of peak oxygen consumption were only significant when fat free mass was expressed as percentage of skeletal muscle mass. Although the regressions are significant, the effect size of the models for men and women is low to intermediate, with r² and ω² values ranging from 0.07 to 0.52. In all models, except for the percentage of skeletal muscle mass expressed as z score in men, age was not a significant predictor of fat free mass.

Figure 3 shows the results of the regression models when handgrip strength was the dependent variable. Again, the r² values were higher in men and peak oxygen consumption was a significant and independent predictor in most models. Gross work efficiency was a significant predictor or handgrip strength in almost all models. Beta coefficients of physical activity level were only significant for the percentage of skeletal muscle mass expressed as z score in men. Total physical activity, calculated using the physical activity questionnaire did not have any predictive value. As in the case of fat free mass, the effect size of the models for men and women is low, with r² and ω² values ranging from 0.16 to 0.45.

We herein show that in young and old men and women, the presence of sarcopenia is associated with lower activity energy expenditure and a lower physical fitness, measured using an exercise calorimetry.

The definition of sarcopenia relies in the loss of muscle mass and function. We measured muscle mass using DEXA, which is accurate and reliable and accepted by most guidelines on sarcopenia [16]. Imaging methods such as magnetic resonance, ultrasound or computed tomography are also used, but their reliability is usually tested against DEXA [17]. The most widely used functional measure is handgrip strength, whose prognostic value is well established [18]. Therefore we used DEXA and handgrip strength to assess muscle mass and strength respectively. We have shown previously that both measures are predictive of mortality in our population of older subjects [19, 20]. The cutoff of muscle mass used to define sarcopenia was set at 1 standard deviation from the normal value, determined in a large set of participants of the same genetic and socioeconomic background reported previously [12]. In that report, we showed that using one instead of two standard deviations provides a better approach to determine sarcopenia in our population. The same cutoff was proposed by a Brazilian group of investigators previously [21]. The cutoff values for handgrip strength were derived from a similar group of participants, to avoid any regional bias in the determination of sarcopenia.

When comparing participants with and without sarcopenia defined using mass, strength or both, it is noteworthy that in all cases, subjects with sarcopenia had a lower peak oxygen uptake and activity energy expenditure, except in women, when sarcopenia was defined only by using handgrip strength. The differences in peak oxygen uptake have been seldom reported. The aforementioned Brazilian authors, performed an exercise test in a subgroup of participants with and without sarcopenic obesity finding differences in absolute but not relative peak oxygen uptake [21]. A Japanese group found an association between estimated oxygen uptake and skeletal muscle mass in men and women [22]. It is well known that training improves peak oxygen uptake, even in older people [23], but we are reporting results from participants who have not been trained. Thus, what we are observing are the effects of usual physical activity on this parameter [24], which is lower in participants with sarcopenia. There are several reports showing a lower activity energy expenditure in subjects with sarcopenia [25], therefore this finding is not surprising, but gives strength and reaffirms the differences found in peak oxygen uptake. On top of this, both peak oxygen consumption and usual physical activity are predictors of mortality healthy people [26, 27]. Thus, it is tempting to speculate if the prognostic value of sarcopenia on mortality is an independent effect or just a reflection of the higher level of sedentariness of people with this condition.

When muscle mass or strength were treated as a continuous variable in the multiple regression models, peak oxygen consumption was accepted as a significant predictor of fat free mass in men. In women the predictive capacity of this parameter was lower. These results reaffirm the concept that fitness is a protective factor against sarcopenia. The effect size of the regression equations was low, but they fulfill our purpose to show that physical fitness is related to the preservation of muscle mass across all ages. A report by Aggio et al. showed an association between sedentary time and the presence of severe sarcopenia, of a similar magnitude than the association herein reported [28]. A similar association was reported previously in older women. Surprisingly, the peak oxygen consumption values obtained by these authors were very similar to those obtained by us. However they interpreted the results, arguing that a lower muscle mass is the cause and not the consequence of a lower aerobic capacity [29]. Other authors also proposed this alternative hypothesis [30]. The only means of resolving is fitness or low muscle mass are the initiating event in the cascade of functional decline, is to perform longitudinal studies of cohorts, measuring the other factors that may affect aerobic capacity such as cardiovascular performance.

The association between handgrip strength and gross work efficiency is intuitively obvious, considering that both measures are a functional assessment of muscle. However, gross work efficiency and handgrip are assessing different muscular groups. This may be an indicator that a lower work efficiency is indicating an overall increase in the ATP cost for muscle contraction, phenomenon that is also associated with sarcopenia [31].

The International Physical Activity Questionnaire had no association with fat free mass. The lack of association of this parameter with peak oxygen uptake or activity energy expenditure, cast doubts about the accuracy of these types of questionnaires to objectively assess physical activity. Other authors report a significant albeit very low concordance between these type of questionnaires and data derived from accelerometry [32, 33]. Therefore, the interpretation of results obtained from these questionnaires should be very cautious and probably reserved only for large scale epidemiological studies. Using this questionnaire, we observed that young women had a lower total physical activity than the other participants. However this observation was not confirmed with Actigraphy, a much more accurate method.

The two main weaknesses of this study is that it is cross sectional and not having used doubly labeled water to assess activity energy expenditure. The latter is unsurmountable since the cost of doubly labeled water is too high. The former should encourage us to follow the participants of this cohort and perform a new assessment in the future to confirm is the actual loss of muscle mass, the essential aspect of sarcopenia, is associated with usual physical activity.

Our main strengths are that we studied both young and older people, showing that sarcopenia effectively occurs in young subjects and that physical activity is a risk factor for it in all age ranges. Also it is important to highlight that all comparisons were made separately by gender, avoiding the confounding effect of the striking differences in body composition and muscle strength between men and women [34]. It should also be pointed out that we used objective and accurate measures of physical activity such as accelerometry and cardiopulmonary exercise testing.

Sedentariness is associated with sarcopenia in this group of participants and should be considered as an important risk factor for this condition.