Altered serum short-chain fatty acids in sarcopenia among Chinese elderly women: a case–control study

Abstract

Background

Sarcopenia, characterized by the progressive loss of muscle mass and function, is increasingly prevalent among the elderly in China and globally. Emerging evidence suggests that short-chain fatty acids, key metabolites produced by gut microbiota, may influence muscle health. This study aimed to investigate the association between serum short-chain fatty acids and sarcopenia in elderly Chinese women, and to explore potential metabolic biomarkers using a targeted metabolomics approach.

Methods

A case–control study was conducted involving 100 community-dwelling women aged 65 to 75 years in Shanghai, with 50 diagnosed with sarcopenia and 50 age-matched healthy controls. Sarcopenia was defined according to the Asian Working Group for Sarcopenia 2019 criteria. Fasting blood samples were collected, and serum short-chain fatty acid levels were measured using Gas Chromatography-Mass Spectrometry. Dietary intake and demographic data were obtained through structured questionnaires and food frequency assessments. Statistical analyses, including independent sample t-tests and partial correlation analysis, were performed using SPSS version 21.0. Metabolic pathway enrichment was analyzed using MetaboAnalyst.

Results

Compared to the control group, the sarcopenia group exhibited significantly lower serum levels of propionic acid (P = 0.004) and isovaleric acid (P = 0.001). Pathway analysis identified 19 significantly enriched metabolic pathways, three of which, carbohydrate digestion and absorption, protein digestion and absorption, and degradation of aromatic compounds, were highly associated with propionic and isovaleric acids. Dietary assessment revealed that individuals with sarcopenia had lower intake of energy, total protein, and high-quality protein, but higher sodium intake (all P < 0.05).

Conclusions

Altered profiles of serum short-chain fatty acids, particularly reduced propionic acid and isovaleric acid, are associated with sarcopenia in elderly women. These metabolites may serve as potential biomarkers for early identification and risk assessment. The findings highlight the relevance of gut microbiota-derived metabolites and dietary factors in sarcopenia pathophysiology and support future development of nutritional and metabolic interventions for prevention.

Trial registration

The trial protocol was filed with the Chinese Clinical Trial Registry (registration number ChiCTR2100048874) on July 19, 2021.

Introduction

The global population is aging at an unprecedented rate. According to the international standard established by the United Nations in 1956, a country or region is considered to have an aging population when individuals aged 65 years and older comprise more than 7% of the total population [1]. As of 2021, approximately 761 million people worldwide were aged 65 years or older, accounting for 9.8% of the global population. This number is projected to rise to 1.6 billion by 2050 [2, 3]. The demographic shift is particularly evident in China, where the population aged 65 and above reached 216.76 million by the end of 2023, representing 15.4% of the total population [4]. This proportion substantially exceeds the global average, underscoring the severity of population aging in China.

The growing aging population in China is closely linked to an increased prevalence of age-related diseases. Among these, sarcopenia is particularly significant, a progressive and generalized skeletal muscle disorder characterized by the loss of muscle mass and function. First defined by Dr. Irwin H. Rosenberg in 1989, sarcopenia has since been recognized as a major public health concern in older adults. The European Working Group on Sarcopenia in Older People defines it as a “progressive, systemic decline in muscle mass and/or muscle strength or physical function related to aging” [5]. Clinically, sarcopenia presents with reduced muscle strength, decreased skeletal muscle mass, impaired physical performance, and an elevated risk of disability and mortality. These manifestations can ultimately lead to loss of independence, increased reliance on long-term care, higher rates of hospitalization, and greater healthcare expenditures.

The prevalence of sarcopenia among older adults varies considerably across regions, ranging from 6 to 33%, depending on the diagnostic criteria applied. Reported prevalence rates are 14% in North America, 21% in South America, 22% in Europe, and 8.7% in Japan [6, 7]. In China, the prevalence is estimated at 17.4%, which is relatively high [8], and in Shanghai, the rate is particularly concerning, reaching 19.37% in community-dwelling elder [9]. This high prevalence suggests that a substantial proportion of older adults may require increased medical support and long-term care services, thereby intensifying the burden on the healthcare system [10].

There is growing recognition within the academic community that the development and progression of sarcopenia are closely linked to nutritional status [11]. Over the past decade, short-chain fatty acids (SCFAs) – principally acetate, propionate and butyrate – have been identified as microbiome-derived metabolites that directly influence skeletal muscle physiology through at least three, non-mutually exclusive mechanisms:(i) Butyrate and propionate have been shown to inhibit NF-κB activation and expand colonic Foxp3⁺ regulatory T cells [12, 13]. These short-chain fatty acids also reduce systemic levels of IL-6 and TNF-α [14], cytokines known to activate NF-κB and promote muscle protein degradation via the ubiquitin–proteasome pathway [15]. (ii) Mitochondrial biogenesis & oxidative efficiency: In mouse skeletal muscle and L6 myotubes, sodium butyrate increases PGC-1α expression, mitochondrial gene expression (e.g., CPT1b, COX-I), and fatty acid oxidation, associated with AMPK and p38 phosphorylation [16]. (iii) Insulin–IGF-1 sensitisation: Current evidence indicates that oral administration of SCFAs, such as acetate or butyrate, enhances insulin sensitivity and stimulates glucose uptake in skeletal muscle of high-fat-fed mice via activation of the G protein-coupled receptors GPR41 and GPR43 [17, 18]. Collectively, these data position SCFAs as key transducers of the gut–muscle axis, although population-level metabolomic evidence remains scarce [19–21]. Animal studies have demonstrated that exogenous supplementation with these metabolites can reverse muscle mass and strength loss in mice [22]. However, the mechanisms through which SCFAs influence skeletal muscle metabolism and function remain incompletely understood. This knowledge gap underscores the need for further research to clarify their role in the pathophysiology of sarcopenia.

Metabolomics is an emerging analytical approach increasingly employed to investigate the associations between metabolic pathways and various disease processes [23]. Despite its promise, the use of metabolomics to explore the relationship between SCFAs and sarcopenia remains limited, particularly in real-world population-based studies. To address this gap, we conducted a case–control study involving elderly individuals with sarcopenia, applying a targeted metabolomics strategy using Gas Chromatography–Mass Spectrometry to quantify serum SCFAs. This approach aims to elucidate the metabolic alterations and potential mechanisms underlying sarcopenia in the Chinese elderly population. A deeper understanding of these mechanisms may offer novel insights into nutritional interventions and therapeutic strategies for sarcopenia prevention and management.

Methods

Participants

Considering that there were significant differences in muscle mass and metabolic characteristics between genders, this study only investigates females. Males will be included in the next step if meaningful results were obtained.We selected 50 women with sarcopenia as the case group using the Asian Working Group Standard for Sarcopenia (AWGS) [24], and then matched them with another 50 healthy women, aged within ± 1 year and having the same number of chronic diseases, as the control group, in Huangpu and Putuo districts of Shanghai. Since we adjusted for chronic diseases, we did not match for BMI to avoid over-matching. Firstly, all participants were able to walk independently. In addition, the following situations were also excluded: having cognitive impairment; suffering from heart, liver, kidney, and other important organ diseases and infectious diseases; experiencing pain or dysfunction of the hip and knee joints; having a recent history of trauma, fractures, or surgery, and other diseases that were not suitable or unable to exercise; taking drugs that affect the study (thyroid function drugs, hormone drugs, long-term steroid drugs, weight loss drugs, etc.).

According to the 2019 consensus of the Asian Working Group for Sarcopenia (AWGS 2019), the operational diagnosis of sarcopenia was established through a standardized three-step algorithm. First, appendicular skeletal muscle mass (ASM) was quantified by multi-frequency bioelectrical impedance analysis (BIA); the cut-offs defining low muscle mass was < 5.7 kg·m⁻2 (BIA) in women. Second, maximal isometric hand-grip strength is assessed with a calibrated Jamar-type dynamometer in the dominant arm after 2–3 familiarisation trials; the highest of two consecutive readings is recorded and classified as reduced when < 18 kg for women. Third, physical performance is evaluated using at least one validated test: habitual gait speed over a 6-m course (< 1.0 m·s⁻1 indicates impairment), the five-time chair-stand test (≥ 12 s indicates impairment), or the Short Physical Performance Battery (total score ≤ 9). Sarcopenia is confirmed when low muscle mass coexists with either low grip strength or impaired physical performance.

Collection of research data

First, all 100 project participants underwent an offline one-on-one questionnaire survey to review and record their general situation, illness, dietary intake, exercise, and anthropometric data. We used a food frequency questionnaire (FFQ) to obtain information on dietary intake, converted each food according to its requirement (raw weight, edible part weight, dry weight, fresh weight, etc.) based on the conversion ratio, and then calculated each nutrient using the Chinese Food Composition Table. High-quality protein was defined as protein from soybeans and their products, livestock, fish and shrimp, eggs, dairy products, preserved animal foods, and dietary supplements. The content of the FFQ and the questionnaire survey was identical to that used in the previously published intervention study [25].

In addition, we collected fasting blood samples from all participants in tubes filled with anticoagulants, centrifuge and aliquot on site, and transported them to the laboratory within 4 h, preservating at −80 °C for subsequent use in metabolomic analysis studies. These steps were taken to minimize degradation and ensure the stability of the samples during collection and storage.

Procedures for targeted metabolomics analysis of serum SCFAs

We employed Thermo- Trace 1300 gas chromatography (Thermo Fisher Scientific, USA) and Thermo- ISQ 7000 Mass Spectrometry (Thermo Fisher Scientific, USA) to investigate the metabolism of SCFAs in the serum samples of participants. Shanghai Weihuan Biotechnology Co., Ltd. (APE × BIO's exclusive agent in China) provided assistance for the targeted measurement of SCFA metabolism in this project.

Pre-test preparations

Chromatographic column: DB-FFAP capillary chromatographic column (30m × 0.25mm × 0.25μm), Chromatographic conditions: The carrier gas was high-purity nitrogen, with a flow rate of 41.1mL/min; injection volume was 5 µL, no split flow; injection port temperature was 200℃; detector temperature was 240℃; tail blowing: nitrogen, flow rate: 25mL/min; temperature rise procedure: hold at 100℃ for 1 min, raise to 180℃ at 8℃/min, and hold for 1 min.

Preparation of standards

We measured appropriate amounts of pure standards of acetic acid, propionic acid, butyric acid, isobutyric acid, valeric acid, isovaleric acid, and caproic acid, and prepared ten mixed standard concentration gradients with diethyl ether: 0.02 μg/mL, 0.1 μg/mL, 0.5 μg/mL, 2 μg/mL, 10 μg/mL, 25 μg/mL, 50 μg/mL, 100 μg/mL, 250 μg/mL, 500 μg/mL. We drew a standard curve with the peak area as the abscissa and the concentration of each compound as the ordinate. Both the mother solution and the working standard solution were stored at −20℃.

Standard curve and quantitative limit

After testing the concentration series of the standard solution, we constructed a calibration curve using the concentration of the standard as the abscissa and the peak area ratio of the standard to the internal standard as the ordinate, and obtained a linear regression equation for each substance (correlation coefficient r > 0.99).

Preparation of sample

We took the appropriate amount of sample in a 2 mL centrifugal tube, added 50 µL of 15% phosphoric acid, 100 µL of 125 µg/mL internal standard (isohexanoic acid) solution, and 400 µL of ether. We homogenated the sample mixure for 1 min, then centrifuged at 12,000 RPM for 10 min at 4 °C, took the supernatant, and analyzed it on the machine. This internal standard was used to account for any variability in sample preparation and injection, thereby improving the accuracy and reproducibility of the quantification.

Measurement and quality control

We measured the mixed standard samples, standard controls, and test samples in sequence. Based on the peak area of the standard samples, we plotted the standard curve, calculated the linear regression equation, and performed qualitative and conversion processing of the results.

We measured the stability of each batch of samples by needle-following QC detection. By calculating the RSD of all QCs through the ratio of peak areas of the standards to the internal standards, we evaluated the stability of the method.

To address potential batch effects, we conducted QC tests at intervals for each batch of samples. By calculating the RSD of all QCs through the ratio of peak areas of the standards to the internal standards, we evaluated the stability of the method. This approach allowed us to identify and correct for any batch-to-batch variations, ensuring the reliability of our results.

Statistical analysis

SPSS 21.0 was used to analyze participants' basic information and dietary intake. A T-test was performed on data with a normal distribution, expressed as mean ± standard deviation; Chi-square test was used for grade data, and P < 0.05 was considered statistically significant.

In addition to the data above, we conducted the following four steps of analysis for data from serum samples, which was also the focus of this study. First, we used principal component analysis (PCA) to integrally check the aggregation degree of QC samples (QC samples refer to the qualified samples used by enterprises in quality control). Second, we conducted PCA and orthogonal partial least squares discriminant analysis (OPLS-DA) on the two sets of data to screen for differences between the two groups of substances. In addition, we used an independent sample T-test of the OPLS-DA model (P < 0.05) to further search for differential representative metabolites. Finally, using MetaboAnalyst online analysis software, we mapped the differential metabolites from case group and control group to their KEGG IDs and selected Homo sapiens as the species for pathway analysis. To control the false positive rate in multiple comparisons, we applied the Benjamini–Hochberg procedure to adjust the P-values for the False Discovery Rate (FDR).

Results

Basic Information

Comparison between the two groups showed that body weight, skeletal muscle mass, and body mass index were significantly lower in the case group than in the control group (P < 0.05). There was a significant difference in the distribution of education levels between the groups (P < 0.05). No significant differences were observed in other demographic or clinical characteristics (Table 1).

Table 1.

Basic characteristics of participants in each group

Variable	Sarcopenia group (n = 50)	Control group (n = 50)	t/X2	P
AGE, y(mean ± SD)	68.44 ± 3.47	68.74 ± 3.36	−0.439	0.661
Education level, n (%)			7.349	0.026
High school and above	12(24.0)	25(50.0)
Junior high School	32(64.0)	22(44.0)
Primary school and below	6(12.0)	3(6.0)
Occupation, n (%)			0.198	0.656
Physical work oriented	13(26.0)	15(30.0)
Mainly mental work	37(74.0)	35(70.0)
Marital status, n (%)			0.638	0.424
Divorced or widowed	10(19.2)	7(14.0)
Married	40(76.9)	43(86.0)
Height, m(mean ± SD)	156.79 ± 5.15	157.84 ± 4.93	−1.037	0.302
Body weight, kg(mean ± SD)	55.46 ± 6.67	59.95 ± 6.31	−3.461	0.001
Skeletal muscle mass, kg(mean ± SD)	13.49 ± 1.32	16.92 ± 1.78	−10.957	< 0.001
BMI, kg/m2(mean ± SD)	22.53 ± 2.33	24.08 ± 2.44	−3.238	0.002
Body fat percentage, %(mean ± SD)	35.26 ± 5.38	35.77 ± 5.62	−0.469	0.640
Waist-to-hip ratio(mean ± SD)	0.91 ± 0.06	0.92 ± 0.05	−0.877	0.383
Occurrence of chronic diseases, n (%)
Stroke	12(24.0)	11(22.0)	0.056	0.812
Coronary heart disease	12(24.0)	9(18.0)	0.542	0.461
Hypertension	21(42.0)	28(56.0)	1.961	0.161
Hyperlipidemia	11(22.0)	17(34.0)	1.786	0.181
Diabetes	14(28.0)	12(24.0)	0.208	0.648
Osteoporosis	10(20.0)	8(16.0)	0.271	0.603
Fatty liver	11(22.0)	18(36.0)	2.380	0.123

Dietary intake

Comparison of dietary data between the two groups showed that the sarcopenia group had significantly lower intakes of total energy, protein, high-quality protein, and dietary fiber compared to the control group (P < 0.05). In contrast, sodium intake was significantly higher in the sarcopenia group (P < 0.05). No significant differences were observed in the intake of other nutrients between the groups (Table 2).

Table 2.

Dietary intake of participants in each group (mean ± SD)

	Sarcopenia group (n = 50)	Control group (n = 50)	t	P
Energy, kj	7095.52 ± 3234.73	8334.30 ± 2731.95	−2.069	0.041
Protein, g	74.27 ± 36.93	88.91 ± 34.59	−2.045	0.044
High-quality protein, g	40.01 ± 23.50	53.06 ± 26.44	−2.610	0.010
Fat, g	53.30 ± 30.01	57.20 ± 22.49	−0.735	0.464
Carbohydrate, g	238.67 ± 106.21	250.53 ± 94.80	−0.589	0.557
Dietary fiber, g	6.01 + 3.12	6.52 + 3.98	−2.029	0.046
Calcium, mg	691.73 ± 446.24	846.72 ± 502.80	−1.630	0.106
Iron, mg	28.01 ± 21.11	26.39 ± 12.80	0.463	0.644
Iodine, μg	1.36 ± 1.96	0.80 ± 0.74	1.879	0.065
Vitamin A, μg RAE	658.98 ± 324.30	715.51 ± 330.13	−0.864	0.390
Vitamin D, IU	1.05 ± 3.93	1.06 ± 6.02	−0.013	0.990
Vitamin E, mg	18.82 ± 12.14	20.58 ± 11.73	−0.734	0.465
Vitamin K, μg	60.34 ± 45.38	72.24 ± 60.36	−1.114	0.268
Thiamin, mg	0.94 ± 0.38	1.03 ± 0.43	−1.071	0.287
Riboflavin, mg	1.23 ± 0.69	1.36 ± 0.64	−0.970	0.334
Vitamin C, mg	105.24 ± 71.39	106.77 ± 79.56	−0.101	0.919

Metabolomics analysis

OPLS-DA score chart

Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) revealed a separation between samples from the sarcopenia group and the control group, indicating distinct metabolic profiles. The model quality parameters were R2Y = 0.427 (P < 0.05) and Q2 = 0.363 (P < 0.05). Although these values are below the commonly accepted threshold of 0.4 for strong model reliability, they suggest the model has some discriminatory capability and that metabolic differences exist between the groups (Fig. 1).

Fig. 1.

Permutation test of the OPLS-DA model comparing the control group and sarcopenia group

Permutation test diagram

The OPLS-DA models were evaluated using permutation testing to assess model validity. The results demonstrated that R2Y values consistently exceeded Q2 values, and the Q2 values obtained from randomly permuted data were lower than those from the original dataset. The intercept of the Q2 regression line on the Y-axis was consistently below zero (commonly considered acceptable if less than 0.05), suggesting that the model was not overfitted (Fig. 2).

Fig. 2.

Permutation test of the OPLS-DA model comparing the control group and sarcopenia group

Partial correlation analysis of differential SCFAs

After adjusting for age, education level, body fat percentage, and physical activity, partial correlation analysis revealed that higher levels of isovaleric acid (r = –0.643, P = 0.006) and propionic acid (r = –0.384, P = 0.004) were negatively associated with sarcopenia, suggesting a potential protective effect. In contrast, higher levels of caproic acid (r = 0.706, P = 0.001) were positively associated with sarcopenia, indicating a potential risk factor (Table 3).

Table 3.

Filtered serum differential metabolite levels between control and sarcopenia groups

Name	r	P	FDR-P
Isovaleric acid	−0.643	< 0.001	0.006
Propionic acid	−0.384	< 0.001	0.004
Caproic acid	0.706	< 0.001	0.001

Path analysis

Pathway analysis identified 19 metabolic pathways that were significantly enriched (P < 0.05) (Fig. 3). Among these, three pathways demonstrated highly significant enrichment (P < 0.01): carbohydrate digestion and absorption, protein digestion and absorption, and degradation of aromatic compounds (Table 4).

Fig. 3.

Bubble plot of pathway enrichment analysis for differential metabolites between the control and sarcopenia groups

Table 4.

Key metabolic pathways associated with serum SCFAs

Metabolic pathway	Associated SCFAs	P	FDR-P
Carbohydrate digestion and absorption	Propionic acid	< 0.001	< 0.001
Protein digestion and absorption	Propionic acid, Isovaleric acid	< 0.001	< 0.001
Degradation of aromatic compounds	Propionic acid	< 0.001	0.001

Discussion

The internationally recognized causes of sarcopenia include reduced net synthesis of muscle proteins, a decline in the number of skeletal muscle fibers, oxidative cellular damage, and impaired tissue repair processes [24, 26]. In recent years, the role of the gut microbiota in sarcopenia has gained increasing attention. SCFAs have been identified as key intermediates through which the gut microbiota may influence the onset and progression of sarcopenia. However, the precise mechanisms by which short-chain fatty acids affect muscle metabolism and function remain incompletely understood [27, 28].

In this study, metabolomics was employed to investigate SCFAs metabolism in elderly individuals with sarcopenia. Three key differential metabolites were identified: decreased levels of propionic acid and isovaleric acid, and increased levels of caproic acid. These alterations were associated with significant enrichment in three metabolic pathways. Propionic acid was linked to carbohydrate digestion and absorption, protein digestion and absorption, and degradation of aromatic compounds. Isovaleric acid was associated with protein digestion, while caproic acid showed no significant relationship with the enriched pathways. These three pathways—carbohydrate digestion and absorption, protein digestion and absorption, and degradation of aromatic compounds—were prioritized not only for their statistical significance (P < 0.01) but also for their established roles in nutrient metabolism, energy production, and microbial–mammalian co-metabolism, all of which are essential for muscle maintenance and function. Carbohydrate and protein metabolism are central to ATP production and amino acid availability, which directly support muscle protein synthesis and contractile function [29].Additionally, the degradation of aromatic compounds reflects the activity of gut microbial enzymes involved in the metabolism of dietary polyphenols and amino acids, influencing host–microbe metabolic interactions that have been increasingly linked to systemic energy homeostasis and skeletal muscle health [30]. Dysregulation of these pathways has been associated with aging-related metabolic decline and sarcopenia [31].

The impact of SCFAs on muscle health has been confirmed in several studies [11, 31, 32]. Among the three SCFAs identified in our study, propionic acid has been the most extensively studied. Produced by the fermentation of dietary fiber by gut microbiota, propionic acid can reach relatively high concentrations in the colonic lumen [33]. It plays a key role in maintaining intestinal immune homeostasis, enhancing the function of regulatory T cells (Tregs), and alleviating intestinal inflammation [34]. In our study, participants with sarcopenia had a relatively lower intake of dietary fiber compared to healthy controls, which may partly explain the reduced serum levels of propionic acid observed in this group. Propionic acid may serve as a critical mediator linking gut microbiota to skeletal muscle function. Its effects on muscle appear to involve two primary pathways: energy metabolism and immune regulation. In our pathway analysis, all three metabolic pathways associated with propionic acid were related to energy metabolism. Previous studies suggest that SCFAs can activate AMP-activated protein kinase (AMPK) in myotubes and skeletal muscle cells [16, 35], likely due to their ability to increase intracellular AMP levels and the AMP/ATP ratio [35, 36]. AMPK activation induces a metabolic profile characterized by enhanced fatty acid uptake and oxidation, increased glucose uptake and glycogenesis, and reduced lipogenesis and glycolysis [37]. Although direct evidence of propionic acid activating AMPK in muscle cells is limited, its AMPK-activating potential has been demonstrated in other cellular models, moreover, propionate suppresses hepatic gluconeogenesis via GPR43-AMPK signaling, thereby preserving systemic glucose availability for muscle uptake, and its decline may thus deprive muscle of energy substrate while amplifying proteolysis [38], suggesting that AMPK phosphorylation may be a key mechanism through which propionic acid modulates skeletal muscle metabolism. In addition, energy metabolism is closely linked to protein synthesis. Propionic acid may alleviate muscle atrophy by regulating protein synthesis pathways, promoting ATP production, enhancing glucose uptake, improving insulin sensitivity, and reducing inflammation [39, 40]. Furthermore, some gut microbiota species not only ferment dietary fiber to produce propionic acid but also contribute to the degradation of complex compounds such as aromatic compounds [41], representing another potential pathway through which propionic acid supports muscle energy homeostasis.

Although the three pathways identified in our study were not directly linked to immune regulation, previous research has demonstrated that propionic acid can inhibit the mammalian target of rapamycin (mTOR) signaling pathway, activate autophagy, and thereby facilitate the clearance of damaged proteins and the recycling of cellular components, ultimately alleviating muscle atrophy [42, 43]. Prior studies [44, 45], including our own findings [46], have shown that amino acids can activate the mTOR pathway to promote protein synthesis. However, excessive or sustained activation of this pathway may impair insulin sensitivity, suppress autophagy, and disrupt mitochondrial function, thereby exerting detrimental effects on skeletal muscle. A balanced combination of propionic acid and amino acids may optimize the regulatory effects of mTOR signaling and enhance its muscle-protective benefits.

Isovaleric acid has been shown to improve osteoporosis by inhibiting the differentiation and activity of osteoclasts (OCs) [47]. Given the interdependence between skeletal and muscle health, a reduction in isovaleric acid levels may contribute to decreased bone mass, potentially exacerbating the development of sarcopenia. Our findings also indicate that isovaleric acid is associated with protein channels, suggesting a role in protein synthesis, which is critical for both muscle and bone maintenance. Moreover, Isovaleric acid decline reflects diminished beneficial proteolysis and lower abundances of Isovaleric acid-producing taxa like Verrucomicrobia [48], suggesting compromised gut–muscle axis signaling. Experimentally, oral animal protein hydrolysate restored Isovaleric acid together with muscle mass in aged mice, and the gain correlated positively with muscle protein and negatively with pro-inflammatory cytokines [49]. Thus, Isovaleric acid reduction could mediate sarcopenia by marking defective protein fermentation, reduced anti-inflammatory metabolites, and subsequent proteolysis acceleration. In contrast, research on caproic acid remains limited, and no definitive evidence has been established linking it to muscle metabolism. Although caproic acid was chemically classified as a medium-chain fatty acid, converging evidence proposes that caproic acid be regarded as “extended” SCFAs which can be generated by specific intestinal bacteria and modulate host metabolism. This may be due to its relatively low abundance in the gut, where it constitutes approximately 1% of total SCFAs [50], potentially resulting in minimal physiological impact. Our study sample was limited, which may have led to a systematic bias resulting in the increase of caproic acid. Nevertheless, it is plausible that caproic acid, like other SCFAs, may be involved in energy metabolism and thereby influence muscle function. Although current evidence on isovaleric acid and caproic acid is limited, prior studies on total SCFAs suggest that their effects on muscle are primarily mediated through energy metabolism and immune regulation pathways. Future in vitro and animal studies are needed to elucidate the specific mechanisms of these fatty acids in gut-muscle interactions. Industrial-scale production of isovaleric acid may help address endogenous deficiencies, while targeted removal of excess caproic acid could potentially support muscle health in individuals with sarcopenia.

In China, research on the relationship between sarcopenia and SCFAs metabolism remains limited, and the use of metabolomics in sarcopenia studies is relatively uncommon. While elderly individuals in different regions may vary in diet and lifestyle, metabolic characteristics are similar among people of the same gender and race. Although we don’t recommend directly extrapolating the findings to the entire national population, this study contributes valuable data to the field and provides new insights for the prevention and management of sarcopenia in older adults. However, several limitations should be acknowledged. First, as a case–control study, this research cannot establish a causal relationship between SCFAs and sarcopenia; further experimental studies are required to address this limitation. Second, the study population included only female participants, limiting the generalizability of the findings to the broader elderly population, particularly men.

Conclusions

In this case–control study, elderly Chinese women with sarcopenia exhibited reduced serum propionic acid, alongside elevated isovaleric acid, suggesting altered SCFA metabolism in sarcopenia pathophysiology. More relevant studies can be carried out to verify its clinical effectiveness.

Authors’ contributions

ZW and JZ designed the research. ZW and CS analyzed the data and drafted the manuscript. ZS, QS, WL, and ZS (Zhuo Sun) conducted the research and contributed to data collection. All authors read and approved the final manuscript.

Funding

This work was supported by grants from the Key Project in the Three-year Plan of Shanghai Municipal Public Health System (2023–2025) [grant numbers GWVI-4] and the Shanghai Sailing Program [grant number 23YF1437000].

Data availability

The data can be obtained by contacting the corresponding or first author.

Declarations

Ethics approval and consent to participate

All procedures involving human participants were conducted in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Ethical Review Committee of the Shanghai Municipal Center for Disease Control and Prevention (Approval No. 2019–46).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.