To our knowledge, this is the first study to compare six different criteria for the diagnosis and sub-classification of sarcopenia within the same sample of Brazilian community-dwelling older women. The results show that the criteria proposed by Baumgartner et al. and Cruz-Jentoft et al. identified fewer older women with a loss of muscle mass (13% and 16%, respectively), whereas the highest frequency of older women diagnosed with sarcopenia was found using the criteria proposed by Studenski et al.
Of note, the cutoff points proposed by the different authors were based on populations with different characteristics, whereas only Studenski et al. used data from nine studies with populations from different countries.This suggests that the studied population, specifically population-specific characteristics and racial factors can influence the criteria for the diagnosis of sarcopenia. Thus, the cutoff points proposed by Studenski et al. in the FNIH [4] study exhibited a higher specificity in the identification of pre-sarcopenia and sarcopenia among elders. Similarly, some authors have promoted studies that advocate cutoff points that identify the highest number of elders with sarcopenia, thus allowing for the development of sarcopenia treatment and prevention strategies [13.20].
By applying the criteria proposed by the EWGSOP, we found a sarcopenia prevalence of 16.6% in older women in this study. These data are similar to those found in a study from Hong Kong that applied the same criteria and found a frequency of 12.4%. However, when the above authors applied the criteria proposed by the Asian Working Group for Sarcopenia on the same sample, the prevalence increased to 67.7% [8].
More recently, a research group also assessed the association of the European criteria of the EWGSOP and FNIH and found that individuals who were diagnosed with sarcopenia according to the criteria proposed by the FNIH [5] exhibited twice as high a risk of negative health outcomes. However, when the authors applied the criteria proposed by the EWGSOP, no association was found between negative health outcomes and the occurrence of falls [21].
Among all the consulted studies, reduced lean mass was related to muscle weakness or impaired physical performance, and all authors promoted the use of the DXA as an instrument to assess ALM [1–5]. However, other authors rejected the above assumption and sustained that lean mass is not always directly associated with low strength and function among older people. Scott et al. argued that studies using muscle mass alone for the screening of sarcopenia could be considered faulty because prospective population-based studies examining the associations of low total muscle mass exhibited no predictive linearity of functional loss in the older person [22].
The use of the SMI for the diagnosis of sarcopenia was reported by four research groups [1–3, 5], whereas only one group recommends the use of ALM alone [4] or ALM adjusted for BMI (ALM/BMI) [4]. This shows the lack of consensus regarding whether BMI should be used in the diagnosis of sarcopenia.
The proposed sub-classification of sarcopenia has also been a point of debate in the literature. Preserving functionality in the process of aging is the focus of geriatricians and gerontologists. Thus, although there is no consensus regarding the cutoff points for the diagnosis of sarcopenia, there is agreement in considering both reduced muscle mass and worsening of physical function, which is assessed by functional tests [2, 4, 7, 20]. Thus, one might think that criteria that include the evaluation of physical function are more indicative of identifying sarcopenia.
The present work shows that the criteria suggested by the EWGSOP and FNIH rendered differences regarding the prevalence of sarcopenia of approximately 43%, with significant variability in the number of older women sub-classified with any of the stages, thus confirming the great variability obtained by using different diagnostic and functional criteria. Further reinforcing this context of disagreement, Scott et al. have reported that the proposed criteria to assess the prevalence of sarcopenia and its sub-classification into stages detected fewer individuals with a loss of muscle mass [22]. Bishoff-Ferrari et al. have made similar observations and state that the sub-classification of elders into stages of sarcopenia seems to identify fewer community-dwelling elders with sarcopenia [11]. Conversely, Kim et al. have reported that criteria that consider the functionality of individuals could be more relevant for the assessment of the long-term effects of sarcopenia [23]. Notably, sarcopenia is characterized by quantitative and qualitative losses in skeletal muscle that progress with aging [20] and, as such, both dimensions should be considered for evaluation [24].
Furthermore, muscle strength seems to diminish at a faster rate than muscle mass among older people [22]. The physiological mechanisms that explain muscle weakness with aging are multifactorial and result from impaired neural activation, reduced intrinsic capacity to generate muscle strength and the loss of muscle mass [25, 26]. In light of the results of our study, we hypothesize that subjects with low muscle mass might have previously suffered from reduced strength, thus reinforcing the relevance of considering individuals with pre-sarcopenia and of determining adequate cutoff points for the early detection of sarcopenia.
Thus, regarding measurements of physical and muscle function, although HGS is widely used in the literature as an indirect measure of global and lower limb muscle strength, some studies do not support these statements [24–28]. Specifically, studies on older Brazilian people comparing the results of HGS measurements with those of isokinetic performance found no agreement between both types of assessment [15].This discrepancy of the results might be due to the cutoff points adopted for HGS in different populations [15, 24]. The literature shows that cutoff values per age-range and sex are well above those found in the population of the present study [25, 27], thus suggesting the need to establish specific values for the Brazilian population [25].
Finally, the criteria proposed by Lera et al., who defined specific cutoff points for the Chilean population [3, 19], provide a different approach to the diagnosis of sarcopenia by considering population-specific and cultural variations. A robust comparison of the prevalence of sarcopenia among different populations poses a challenge given the variability of factors that influence its development, progression and diagnosis [9]. Characteristics, such as socioeconomic and cultural aspects, lifestyle habits of each country and the individuality of senescence and senility can affect the diagnosis and sub-classification of sarcopenia stages [9].
We stress that the sample assessed with six different criteria in the present work was the same, thus excluding any interference by clinical, sociodemographic, cultural, functional or lifestyle habit-related variables found in the several previously studied populations. In addition, the use of DXA, which is considered the gold standard for measuring lean muscle mass, the randomization of functional tests, the blinding of researchers with respect to the DXA results and the functional tests and the strict inclusion and exclusion criteria were strengths of this study. However, further studies should be performed comparing the new criteria proposed by the EWGSOP and whether it determines any differences in the prevalence of sarcopenia.