1. Introduction
Sarcopenia has been characterized as age-related low muscle mass (LMM), combined with the decline of muscle strength and physical performance [
1,
2]. Sarcopenia, which is closely associated with disability, poor quality of life, and increased mortality in older people, has gradually attracted considerable attention around the world [
1,
2,
3,
4].
Previous studies indicated that, muscle tissue dysfunction and muscle fiber insufficiency, as the main outcomes of aging-related obesity, may lead to muscle recession and sarcopenia [
5,
6]. As the most commonly used indicator of obesity, body mass index (BMI) has been widely used to detect the association between obesity and sarcopenia [
6,
7]. However, several previous studies reported negative associations between overweight/obesity defined by high BMI and sarcopenia [
8,
9]. A cross-sectional study in Austria found that, compared with their normal weight counterparts, overweight women had a significantly negative association with sarcopenia [
8]. A cohort study in China found high BMI was protective against sarcopenia and increased its reversibility in older people after a four-year follow-up [
9]. Due to the limitation of BMI in differentiating between lean body mass and body fat tissue, some researchers considered that BMI may not be a proper index to discuss the association between obesity and sarcopenia [
10].
Compared with BMI-defined obesity, body fat accumulation during age-related obesity in older people has been considered as a more important risk factor of sarcopenia in older people [
5,
10,
11]. Intermuscular adipocytes hypertrophy and intramyocellular lipid overaccumulation caused by high body fat mass could provoke dysfunction in skeletal muscle cells and inhibit muscle protein synthesis [
12,
13,
14]. Meanwhile, visceral fat increase could induce systematic inflammation and insulin resistance [
15,
16], which in turn could cause LMM and sarcopenia in older people [
17,
18]. Those results indicated that body fat overaccumulation in different tissues may induce muscle decline and sarcopenia through multiple pathways.
A recent study in Japan found diabetes patients with high body fat mass and low BMI may be more likely to develop sarcopenia [
19]. Meanwhile, during the normal progression of obesity, the increasement of lean mass is generally accompanied be a larger increase in fat mass [
20]. This phenomenon suggested that, besides the particular effect of BMI on sarcopenia or the particular effect of body fat mass on sarcopenia, the interaction between BMI and body fat mass should also be considered in the development of sarcopenia. Furthermore, several recent studies indicated that, compared with subcutaneous fat, visceral fat showed more intense association with metabolic dysfunction in older people [
21,
22]. Visceral fat directly is linked to inflammation, liver dysfunction, and insulin resistance, which have been associated with muscle decline and sarcopenia in older people in previous studies [
23,
24]. Several metabolic markers related to inflammation and liver function, such as high sensitivity C-reactive protein (hsCRP), γ-glutamyl transpeptidase (GGT) and alanine aminotransferase (ALT), have been associated with sarcopenia in older people in previous studies [
18,
25,
26]. Those results indicated visceral fat may be a more sensitive predictor of muscle recession and sarcopenia than total body fat in older people. Nevertheless, the interaction between BMI and visceral fat on sarcopenia and the potential mechanisms are still unclear [
23,
27,
28].
Therefore, the present cross-sectional study evaluated the associations of BMI, visceral fat area (VFA), and their interactions with sarcopenia and its components in community-dwelling older Chinese. Relevant mechanisms affecting muscle health, such as body composition and metabolic risk-factors of sarcopenia, were also assessed to explore the potential mechanisms related to the different associations among BMI, VFA, and sarcopenia. Furthermore, we developed screening models for LMM and sarcopenia based on the combination of BMI and VFA in community-dwelling older people.
4. Discussion
Previous studies indicated that underweight older subjects with low BMI were more susceptible to LMM and sarcopenia [
38]. Without the consideration of body fat, older subjects with high BMI tend to have a larger amount of lean body mass, and then they tend to have a sufficient quantity of muscle mass and low risk of LMM [
23]. Meanwhile, increasing accumulation of body fat, especially visceral fat, could increase the risk of muscle protein wasting, systematic inflammation, and insulin resistance, then increasing the risk of muscle mass decline and sarcopenia in older people [
23,
24]. A previous cohort study in Australia found that the increase of body fat mass accompanied with the increase of BMI and the decline of lean body mass, could increase the risk of sarcopenia and fragility in adult males [
39]. Due to the limitation of separating lean body mass and fat mass, BMI could not evaluate the changes in body composition during weight gain [
28]. Therefore, it is necessary to evaluate the relationship between BMI and body fat, especially visceral fat, in the progression of muscle mass decline and sarcopenia.
In the present study, the association of BMI and VFA with sarcopenia and its components was assessed in community-dwelling older people in China. BMI was positively correlated with muscle strength and muscle mass, while VFA showed significantly negative correlations with them. High BMI without visceral obesity was negatively associated with LMM and sarcopenia. While high VFA attenuated the negative association between high BMI and LMM and sarcopenia. Further analysis indicated that the opposite associations of BMI and VFA with LMM and sarcopenia may be partially attributable to their different associations with body composition and metabolic risk factors of sarcopenia. Due to the significant but opposite associations of BMI, VFA with LMM, and sarcopenia, we developed screening tools for LMM and sarcopenia, based on the combination of BMI and VFA for the first time. Our results provided scientific evidence for the diagnosis and prevention of sarcopenia in community-dwelling older people.
As a surrogate of muscle strength, handgrip strength has been measured, and its association with BMI and VFA has been discussed in the present study. A weak positive correlation between handgrip strength and BMI has been shown in
Table 2 (
r = 0.14,
p < 0.001). Meanwhile, the correlation coefficient between handgrip strength and BMI increased after the adjustment of VFA and relevant covariates, this change may partially attribute to the negative correlation between VFA and handgrip strength. The increase of body fat during bodyweight gain may increase the risk of inflammation, insulin resistance, and muscle cell atrophy, and lead to the decline of muscle mass and strength [
12,
13,
17]. A significantly negative association was observed between low handgrip strength and COB, while no significant association was observed between low handgrip strength and other types of obesity in the present study. Meanwhile, the positive association between COB and low handgrip strength were attenuated and disappeared after adjustment. This change reflected the complexity of handgrip strength decline in older subjects. During the aging progress, handgrip strength could be affected by many influencing factors. In addition to BMI and body fat, physical activity, living condition, blood pressure, smoking, stress, chronic disease, and renal function have been associated with handgrip strength in older subjects, and those factors seem to be more important than BMI in handgrip strength decline [
40,
41]. As a surrogate of physical performance, gait speed only showed a weak correlation with VFA. The positive association between COB and low gait speed was also attenuated and disappeared in the adjusted logistic regression model. This change may be partially attributed to the weak correlation between VFA and gait speed, and the numerous influencing factors of gait speed decline in older people, such as knee osteoarthritis, dementia, and balance disorder [
29,
42].
Different with handgrip strength and gait speed, SMI showed a strong correlation with BMI (
r = 0.70,
p < 0.001) in the present study. Compared with the NOB group, SOB was negatively associated with LMM (OR = 0.03, 95% CI = 0.01~0.07) and sarcopenia (OR = 0.07, 95% CI = 0.03~0.18), indicating that high BMI was negatively associated with sarcopenia in older subjects. High BMI showed no significant associations with low handgrip strength and low gait speed, therefore, the negative association between high BMI and sarcopenia mainly attributed to its significantly negative association with LMM. Meanwhile, COB was negatively associated with sarcopenia, while the OR value of sarcopenia in COB group were much higher than SOB group (0.16 vs. 0.07), indicating the negative association between high BMI and sarcopenia could be attenuated by high VFA. Though high VFA could not reverse the negative association between high BMI and sarcopenia, it could increase the risk of muscle protein wasting, systematic inflammation and insulin resistance, then increased the risk of muscle mass decline and sarcopenia in subjects with normal BMI [
6,
23,
43]. Therefore, it is necessary to evaluate the relationship and explore the interplay between BMI and visceral fat in the development of sarcopenia.
The opposite association of BMI and VFA with sarcopenia may be partially mediated by their different associations with body composition. Lean body mass and fat mass both increased with weight gain [
20], and a positive correlation was observed between BMI and body fat mass in older people [
9], suggesting BMI and VFA may have different associations with body composition. Nevertheless, their interactions on body compositions remain unclear. In the present study, high VFA was negatively associated with FFM and ASM, and was positively associated with PBF. Meanwhile, high BMI was positively associated with FFM and ASM, and was negatively associated with PBF. Further analyses indicated that the highest level of total body protein was found in the SOB group, while the lowest level was found in the VOB group. The negative association between total body protein and visceral obesity may be because of the insulin resistance caused by high VFA [
16]. Insulin stimulated muscle protein synthesis by transporting circulating amino acids into skeletal muscle cells [
44]. Meanwhile, insulin also played a vital role in muscle protein breakdown [
45]. The role of insulin in the regulation of protein metabolism indicated the adverse effects of insulin resistance and high VFA on muscle protein metabolism in older subjects. In addition to total body protein, similar negative association was also observed between total body minerals and VOB. It should be noticed that, though high VFA showed a significantly adverse association with body composition, it could not reverse the negative association of high BMI with LMM and sarcopenia in the present study. The negative association between high BMI and sarcopenia was much stronger than the positive association between high VFA and sarcopenia, which may be caused by the intense association between SMI and BMI.
In addition to body composition, metabolic dysfunction and systematic inflammation have been proven to be involved in the development of muscle decline and sarcopenia [
46]. Some metabolic markers, such as GGT, ALT, and hsCRP, have been identified as metabolic predictors of sarcopenia in older people [
18,
25,
26]. Evaluated GGT, positively associated with non-alcoholic fatty liver disease, adiposity, and insulin resistance, has been proven to be an independent risk factor of sarcopenia in community-dwelling older subjects [
25]. Partially in line with the previous study, the highest level of GGT was found in the COB group in the present study, which indicated a higher risk of sarcopenia in older subjects. In previous studies, ALT level was consistently associated with BMI, and subjects with lower BMI and lower ALT tended to be a worse nutritional status [
26,
47]. Inversely, elevated ALT, commonly observed in patients with fatty liver disease, may increase the risk of metabolic dysfunction and insulin resistance [
47,
48,
49]. Meanwhile, a previous study found low ALT level was a predictor for pyridoxine deficiency, frailty, and sarcopenia in older people, which may be associated with liver impairment and malnutrition [
26]. Since a low AST/ALT ratio has been associated with fatty liver disease in previous studies [
50,
51], the AST/ALT ratio was also analyzed in the present study, to further discuss the association between BMI, VFA, and sarcopenia. The levels of the AST/ALT ratio in the SOB group and COB group were significantly lower than the NOB group, indicating that both the high BMI and high VFA, with the highest risk of fatty liver disease, may increase the risk of sarcopenia. Different with the AST/ALT ratio, the significant change of hsCRP was only found in the COB group with high BMI and high VFA, and high VFA showed a positive association with serum hsCRP in GLM analysis. This result indicated that, compared with high BMI, high VFA may be a more dangerous indicator for metabolic dysfunction. High VFA could increase the risk of sarcopenia by triggering liver dysfunction and metabolic dysfunction.
In the present study, due to the limitation of BMI on the evaluation of body composition, and the strongly positive association between high VFA and sarcopenia, the combination of BMI and VFA was used to develop screening tools for LMM and sarcopenia. Sarcopenia diagnosis generally requires measurements of ASM, muscle strength, and physical performance in older people [
29,
37]. However, it is inconvenient to conduct generally screening for LMM and sarcopenia based on CT, MRI and dual-energy X-ray absorptiometry, because of the high equipment costs and the requirement for trained person to use the equipment [
29,
52]. Moreover, dementia, gait disorder, and hand disability in older people may also affect the measurements of muscle strength and physical performance [
29]. Because of those limitations, some researchers developed some simple-to-use screening tools for LMM and sarcopenia via BMI, CC, and other anthropometric measurements [
53,
54,
55]. Meanwhile, most of those screening tools, using a single anthropometric measurement, ignored the variation of body builds and nutrition status among individuals [
53]. In the present study, based on the associations among high BMI, high VFA and sarcopenia, BMI, VFA, and their combination were used to develop screening tools for LMM and sarcopenia. Compared with single anthropometric measurement, the combination of BMI and VFA had significantly better diagnostic efficiency in the prediction of LMM (AUC = 0.88, 95% CI = 0.86~0.90) and sarcopenia (AUC = 0.82, 95% CI = 0.78~0.86). Those results indicated that besides BMI, VFA and their interactions should also be seriously considered in sarcopenia prevention and management. Furthermore, simple-to-use nomograms were developed for the identification of LMM (
Supplementary Figure S1) and sarcopenia (
Supplementary Figure S2) in older people in the present study.
The present study was not without any limitations. First, this study was a cross-sectional study, prospective cohort studies will be necessary for more powerful evidence in the future. Second, due to the absence of a uniform criteria of sarcopenia, criteria recommended by AWGS was used in this study. Furthermore, there are multiple types of obesity classification, based on body fat percentage, waistline and different cut-off values of BMI. Meanwhile, only two obesity classification standards were conducted in the present study. More kinds of obesity classification should be conducted in the future study to discuss the relationship between obesity and sarcopenia in older people.