Sarcopenia and sarcopenic obesity among community-dwelling Peruvian adults: A cross-sectional study

Background Sarcopenia and sarcopenic obesity (SO) have emerged as significant contributors to negative health outcomes in the past decade. However, there remains a lack of consensus on the criteria and cut-off thresholds for assessing sarcopenia and SO. Moreover, limited data are available on the prevalence of these conditions in Latin American countries. To address this evidence gap, we aimed to estimate the prevalence of probable sarcopenia, sarcopenia, and SO in a community-dwelling population of 1151 adults aged ≥ 55 years in Lima, Peru. Methods Data collection for this cross-sectional study was conducted between 2018 and 2020 in two urban low-resource settings in Lima, Peru. Sarcopenia was defined as the presence of low muscle strength (LMS) and low muscle mass (LMM) according to European (EWGSOP2), US (FNIH) and Asian (AWGS) guidelines. We measured muscle strength by maximum handgrip strength; muscle mass using a whole-body single-frequency bioelectrical impedance analyzer, and physical performance using the Short Physical Performance Battery and 4-meter gait speed. SO was defined as a body mass index ≥ 30 kg/m2 and sarcopenia. Results The study participants had a mean age of 66.2 years (SD 7.1), of which 621 (53.9%) were men, and 41.7% were classified as obese (BMI ≥ 30.0 kg/m2). The prevalence of probable sarcopenia was estimated to be 22.7% (95%CI: 20.3–25.1) using the EWGSOP2 criteria and 27.8% (95%CI:25.2–30.4) using the AWGS criteria. Sarcopenia prevalence, assessed using skeletal muscle index (SMI), was 5.7% (95%CI: 4.4–7.1) according to EWGSOP2 and 8.3% (95%CI: 6.7–9.9) using AWGS criteria. The prevalence of sarcopenia based on the FNIH criteria was 18.1% (95%CI: 15.8–20.3). The prevalence of SO, considering different sarcopenia definitions, ranged from 0.8% (95%CI: 0.3–1.3) to 5.0% (95%CI: 3.8–6.3). Conclusions Our findings reveal substantial variation in the prevalence of sarcopenia and SO when using different guidelines, underscoring the necessity for context-specific cut-off values. Nevertheless, regardless of the chosen guideline, the prevalence of probable sarcopenia and sarcopenia among community-dwelling older adults in Peru remains noteworthy.


Background
Sarcopenia is a complex syndrome de ned as the pathological decrease of muscle quantity and quality (1,2). Older adults are particularly at risk as physiological muscle loss through aging starts at approximately 40 years of age (3). Sarcopenia is associated with several adverse health outcomes including falls, disability, and death (4,5). Similar negative impacts are associated with obesity, the prevalence of which has increased, particularly in low-middle income settings, posing social, economic and healthcare challenges (6). Obesity and sarcopenia are closely linked and might interact both pathologically and functionally. Obesity can independently lead to loss of muscle mass and function, due to metabolic derangements, sedentarism, and high co-occurence of noncommunicable diseases (7). On the other hand, sarcopenia might promote fat accumulation due to reduced total energy expenditure. In ltration of fat has been described in sarcopenia via in ammation and adipokine-related mechanisms (8). Thus, both might act synergistically to produce a vicious cycle of fat gain and muscle loss (7).
Unfortunately, assessment of both sarcopenia and sarcopenic obesity (SO) has methodological challenges. For sarcopenia, several working international groups (9,10) have developed guidelines such as the European Working Group on Sarcopenia in Older People (EWGSOP2), the Foundation for the National Institutes of Health (FNIH), and the Asian Working Group for Sarcopenia (AWGS2). These guidelines focus on three main aspects: low muscle strength (LMS), low muscle mass (LMM), and low muscle performance (LMP) to classify older adults as sarcopenic and in some cases, severely sarcopenic. In each case (LMS, LMM and LMP), the guidelines use different parameters and cut-off points. For SO, ESPEN (European Society for Clinical Nutrition and Metabolism) and the European Association for the Study of Obesity (EASO), proposed that SO is de ned as the co-existence of excess adiposity and LMS/function. Latin America has one of the highest growth rates for adults aged ≥ 60 years (11), with a faster rise in obesity prevalence than the rest of the world (12). Unfortunately, there are limited estimates of the prevalence of sarcopenia and SO in this population. Studies that included application of the European (EWGSOP2) cut-offs for muscle strength demand careful consideration of different morphological and nutritional aspects that might in uence muscle mass and strength (13)(14)(15)), yet included calculation of muscle mass using formulas validated only in Caucasian populations (16), or used indicators of muscle mass that are no longer recommended e.g., calf circumference (17,18).
Thus, our aim was to estimate the prevalence of probable sarcopenia (refering only to LMS, also called dynapenia), sarcopenia and SO in a representative community-based sample of adults ≥ 55 years old from Lima, Peru. Due to the methodologcal challenges in de ning these conditions, we compared sarcopenia prevalence according to EWGSOP2, FNIH and AWGS2 criteria in our sample. Finally, to make useful and fair prevalence comparisons with similar settings, we compared the prevalence of probable sarcopenia and sarcopenia in selected Latin American countries applying the same criteria and de nition of sarcopenia in our sample.

Study design and setting
This was a cross-sectional study nested in a large multi-national community-based project called the Global Excellence in Chronic Obstructive Pulmonary Disease Outcomes (GECo) study (19). At the Lima-Peru site, GECo enrolled an age and sex-strati ed random community sample of 3,551 individuals aged 40 years and above, from two urban low resource settings of Lima. The inclusion and exclusion criteria for the GECo study is described elsewhere (20). During the study home-visit, participants performed physical tests (hand grip strength, short-physical performance battery), and socioeconomic and health surveys were administered. The study visit was performed at the participant's home, except for a few cases (< 15) that were in a research o ce.

Study sample and participants
For the present analysis, we selected a subset of participants who were ≥ 55 years old, performed handgrip strength testing and underwent bioelectrical impedance analysis (BIA) for body composition. We excluded from the analysis those who could not perform the handgrip test due to current wrist pain in the dominant hand or could not undergo BIA due to lack of adequate space at the participant's home. Additionally, to increase the accuracy of BIA results, we excluded from the analysis those who had exercised in the last 12 hours and/or had consumed alcohol in the last 24 hours. Figure 1 shows a ow diagram of the enrolment of the participants.

Muscle strength
To measure muscle strength, participants performed three handgrip strength trials using a Jamar hydraulic dynamometer while sitting down [Additional le 1-SOP dynamometry SOP]. The dominant hand was determined with the question "Which hand do you use to hold a fork to eat, or sign your name?". To de ne LMS the best trial had to be lower than the cut-off points from these de nitions: EWGSOP2 (< 27 kg for men; <16 kg for women), FNIH (< 26 kg for men; <16 kg for women) and AWGS2 (< 28 kg for men; <18 kg for women). The best trial was used since it is less likely to be affected by the number of trials compared to the mean of the trials (21).

Muscle quantity
Appendicular skeletal muscle mass (ASMM, kg) was estimated using the whole-body single-frequency

Physical performance
To evaluate physical performance, we measured the Short Physical Performance Battery (SPPB) score. The SPPB is based on three timed tasks: standing balance, 4-meter gait speed, and chair stand tests (23). The timed results of each subtest are scored according to prede ned cut-points for obtaining a global score ranging from 0 (worst performance) to 12 points (best performance) (23). Gait speed was measured at usual pace at 4-meter length using the mean of two tests.
Low physical performance (LPP) was de ned using the cut-off points for gait speed, SPPB, and chair time. For the EWGSOP2 de nition an SPPB score ≤ 8 or a gait speed ≤ 8 was used. (5) In the AWGS2 de nition, an SPPB score ≤ 9 or a 5-chair stand test ≥12 seconds was used. FNIH does not have a de nition of LPP, since this guideline does not de ne a category of severe sarcopenia.
Probable Sarcopenia, Sarcopenia and Sarcopenic obesity Probable sarcopenia was de ned as LMS based on EWGSOP2 and AWGS2 guidelines.
Sarcopenia was de ned as the presence of LMS and LMM according to the EWGSOP2, FNIH and AWGS2 guidelines. Severe sarcopenia was de ned as the presence of sarcopenia and LMP according to the EWGSOP2 and AWGS2 guidelines.
Height was measured three times with a SECA 213® stadiometer, and weight three times with SECA 803® scale, clothed without shoes. BMI (score and categorized by WHO guidelines; BMI ≤ 24.9 as normal, BMI of 25 to 29.9 as overweight, BMI of 30 to 34.9 as class I obesity, and BMI ≥ 35 as class II obesity or more, all measured in kg/m2). Sarcopenic obesity was de ned using a BMI equal or greater than 30 kg/m2 and sarcopenia as de ned in EWGSOP2 or AWGS2. We additionally included age, sex, and level of education (number of years on education and classi ed).
Finally, we compared probable sarcopenia of our Peruvian sample with other Latin American countries, matching each study criteria: LMS cut-off, grip strength calculation, age and sex. For this, we conducted a convenient review of literature, searching databases (PubMed and Google Scholar) for articles published in English or Spanish, with search tems "sarcopenia", "probable sarcopenia", "dynapenia" performed in Latin American countries. Since EWGSOP2 or AWGS2 were developed in 2019, we mainly focused in studies performed since January 2019. In countries were no EWGSOP2/AWGS2 were used, we admitted EWGSOP1 guidelines using their respective cut-off. Additinally, we sought publications from the references lists of identi ed papers. Using information from each of the studies methods section, we obtained prevalences of probable sarcopenia by sex, and utilized the same parameters for calculating probable sarcopenia in our sample.

Data analysis
We performed a descriptive analysis with means and standard deviation for continuous variables and frequencies and proportions for categorical variables. Prevalence estimates of probable sarcopenia, sarcopenia, and SO were calculated for all included de nitions (EWGSOP2, FNIH and AWGS2) with their 95% con dence intervals (CI). While the GECO multisite study was weighted based on census information, we did not include weighted analysis for the present study. Missing data was handled by complete case analysis. Statistical analysis was performed using STATA 17 statistical software (StataCorp LP).

Ethics
All participants provided written informed consent. Ethics permissions were obtained from the University College   Finally, a range from 0.8 to 5.0% of our sample had sarcopenic obesity. When using EWGSOP2 (ASM de nition), a statistical difference between men and women was identi ed (2.9% vs 7.5%, p < 0.001). When using SMI in both AWGS2 and EWGSOP2, the prevalences were around 1%, without differences between men and women. Table 4 shows the comparison of probable sarcopenia point prevalence with different Latin American countries.

Prevalence comparison in Latin America
We used the same criteria used in the selected papers, and used those criteria in our dataset. In Chilean studies, a custom cut-off hand grip estimated for their population (27 kg men and 15 kg women) was used which produced a lower prevalence compared to our sample. In Colombia, prevalence wa much higher than in the current study population (46.5% vs 34.2%) using the EWGSOP2 cut-off. Unless speci ed, grip strength was measuread as best attempt and EWGSOP2 threshold values were used (< 27 kg for men; <16 kg for women)

Discussion
Main ndings In our community-dweeling sample of adults 55 years and over in Lima Peru, at least one in ve adults had probable sarcopenia, the prevalence of which varied due to different cut-offs across guidelines (22.7-27.8%).
We have determined the prevalence of two important conditions, sarcopenia and sarcopenic obesity in an urban community setting. We had the challenge of not having a local or Latin American guideline for de nitions, similar to European or Asian context. Thus, we found heterogeneity of prevalence estimates due to different cutoff points, and parameters suggested to de ne sarcopenia, with even less consensus about sarcopenic obesity.
In response, to enable fairer comparisons, we elaborated Table 4 using the same cut-off points and age-groups in comparative studies. In spite of this, important differences remain. In Colombia, it was reported that almost half of the sample (46.5%) of healthy older adults had probable sarcopenia (15) compared to 34.2% in our lowresource community sample. Part of the difference might be because the Colombian study used the mean between grip trials in both hands, and not the recommended maximum trial of the dominant hand (21). Other Latin American studies, in Chile, used X-ray absorptiometry (DEXA) which can lead to a more accurate assessment of muscle mass than BIA, but this study used local cut-off points that came from a cross-sectional study that did not evaluate prediction of outcomes (24,25). In a Brazilian study, EWGSOP2 and FNIH cut-off points were evaluated based on the lowest 20th (26). Additionally, the formula used to calculate muscle mass (ASM) was standardized only in an Australian population. In an attempt to consolidate several ndings, a systematic review from 31 studies in Brazil found a sarcopenia prevalence of 17%, combining clinical and community samples, with great heterogeneity, several de nitions, and including measurements using BIA and DEXA (27).

Research Implications
Our study highlights the challenges that underpin measurement of sarcopenia and SO in a Latin American setting where there is no regional or local guideline and cut points. This is a call for research investment and collaboration to pool data from Latin America, with the need to develop longitudinal studies that allows determination of valid local cut-offs for hand grip and muscle mass criteria (with BIA or other methods such as ultrasound) associated with negative health outcomes, value our heterogeinity due to different levels of urbanicity and populations living at high altitude. The second call is for transparency. We included the manuals we used for the procedures of hand grip strength and muscle mass. However, that is not common practice, and creates high variability. For instance, if hand grip was measured standing up or sitting. Finally, although the design of the study did no allow us to recommend which guideline should be used for a Latin American population, there is no clear justi cation to prefer the European (EWGSOP2) over the Asian cut-offs (AWGS). In Peru, due to several factors including height, Asian countries cut-offs may be more appropriate (28).

Strengths and Limitations of our study
This study has several strengths. Our sample was a census-representative sample of community-dwelling adults, which allowed a better approximation of the community prevalence of sarcopenia, although we acknowledge that some potential participants were excluded because they were unable to perform spirometry and these might be at higher risk of sarcopenia. A second strength is the effort to measure the parameters of sarcopenia with accurate methods, i.e., muscle strength measured in three attempts using maximum trial and muscle mass by whole-body bioimpedance analysis with valid equipment. BIA that uses two electrodes in the supine position is more accurate compared to those obtained standing or with only one electrode. Furthermore, accurate results in our population. Further population and tool speci c validation studies in our population would be needed to address this issue. Another limitation is that a considerable number of participants did not have sarcopenia measures, which might lead to some selection bias. Finally, we do not weight analysis to calculate prevalence estimates, which might lead to some innacuracies in the estimates, although they do not invalidate comparison across guidelines.

Conclusion and perspectives
There is a considerable variation in the prevalence of probable sarcopenia, sarcopenia and SO among community-dwelling adults 55 years and older from urban settings in Lima, Peru, as assessed according to several guidelines and parameters. In spite of this variation, there is a substantial burden of probable sarcopenia and sarcopenia, which highlights the need to consider preventive measures and interventions. It is imperative to continue producing high-quality evidence regarding sarcopenia from regions such as Latin America, especially longitudinal studies that allow clinicians and researchers to de ne tailored cut-offs for sarcopenia parameters, which ultimately will promote a better understanding of sarcopenia and its prevention and management in these populations.

Declarations
Ethics approval and consent to participate

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Sup1HandGripStrengthEnglish.docx