Exploring Causal Effects of Sarcopenia on Chronic Obstructive Pulmonary Disease and Hospitalization Risk: A Bi-Directional Mendelian Randomization Study

Abstract

Aims/Background:

As the global population ages, age-related chronic obstructive pulmonary disease (COPD) and sarcopenia present significant clinical and economic concerns. Emerging evidence suggests that these conditions are interrelated; however, potential confounding factors may impact observational investigations. Therefore, this study employed a bi-directional Mendelian randomization analysis to systematically investigate the causal relationship—rather than merely conducting a correlation analysis, between sarcopenia-related traits and COPD and hospitalization risk.

Methods:

Univariate, two-sample, and bi-directional Mendelian randomization (MR) analyses were performed using data from genome-wide association studies. The data on Sarcopenia features, including appendicular lean mass (ALM), hand grip strength (HGS), usual walking pace (UWP), and moderate-to-vigorous physical activity (MVPA) were acquired from the UK Biobank. However, data on COPD and hospitalization risk were sourced from the FinnGen consortium. Inverse-variance weighted (IVW) MR and sensitivity analyses were performed to evaluate causal relationships. Additionally, the observed findings were validated using an independent COPD dataset sourced from the UK Biobank.

Results:

The forward MR analysis using IVW revealed a significant negative causality between ALM, HGS, UWP, MVPA, and COPD (all p < 0.05, all p_FDR < 0.05) (False Discovery Rate, FDR), as well as with COPD-related hospitalization (all p < 0.05, all p_FDR < 0.05). These findings were corroborated by validation analyses. Furthermore, reverse MR assessment demonstrated that COPD alleviates UWP (p = 0.001, p_FDR = 0.015); however, validation analysis did not confirm this result. Moreover, additional MR analyses yielded similar trends in causal relationships as evidenced by robust sensitivity tests.

Conclusion:

Our study supported a unidirectional, negative causality between sarcopenia-related traits and COPD and hospitalization risk. This provides possible evidence that sarcopenia increases the risk of COPD and hospitalization at the genetic level. Our findings suggest that improving sarcopenia may serve as a promising strategy for minimizing the incidence of COPD and hospitalization risk, thereby reducing the health burden on these patients.

Graphical abstract

Keywords

Mendelian randomization / sarcopenia / COPD / hospitalization

1. Introduction

Chronic obstructive pulmonary disease (COPD) and sarcopenia, present significant clinical and socioeconomic challenges within geriatric populations [1, 2, 3]. COPD is the third leading cause of death worldwide, with the estimated global financial impact of International Dollars (INT$) 4.3 trillion between 2020 and 2050 [3]. At the population level, a study on the economic burden of sarcopenia assessed its direct healthcare costs in the United States at $18.5 billion for the year 2000 [1].

COPD is a complex and heterogeneous condition characterized by persistent inflammation and various extrapulmonary manifestations [4, 5, 6, 7, 8]. Patients with COPD often exhibit comorbidities, including cardiovascular disease, metabolic dysfunctions, osteoporosis, muscular disorders, cachexia, and mental health conditions, all of which detrimentally affect patient outcomes [5, 7, 8]. Sarcopenia, characterized by age-related reductions in skeletal muscle mass, strength, and function, frequently coexists with COPD [9, 10]. The extrapulmonary manifestations of COPD can severely influence performance by reducing physical activity and diminishing muscle mass and strength, thereby significantly compromising the quality of life [11, 12, 13]. These physical factors are strongly associated with the onset and progression of sarcopenia [14, 15]. A recent systematic meta-analysis reported a higher prevalence of sarcopenia (27.5%) among COPD patients [16]. Furthermore, sarcopenia is related to a range of adverse outcomes, including extended hospitalization, increased risk of fractures, acute exacerbations, falls, disability, and mortality—serving as a strong prognostic indicator for worse disease progression [11, 15]. While observational studies consistently suggest a close association between COPD and sarcopenia, the presence of confounding variables limits the capability to ascertain a direct causal relationship. Therefore, more effective approaches are required to determine their causal relationship, informing targeted management and intervention strategies for these interrelated conditions.

Given the current research-based evidence, we hypothesized that sarcopenia has a causal relationship with both COPD and hospitalization risk, and we tested this assumption using a Mendelian randomization (MR) approach. MR analysis includes genetic variants as instrumental variables (IVs) to evaluate causal links between exposures and outcomes, effectively minimizing the risk of confounding factors and reversing causation [17]. Despite the clinical significance causal links between COPD and sarcopenia, limited studies have explored this association using MR approaches. To date, no randomized controlled trials (RCTs) have been performed to evaluate the bidirectional interactions between these two interrelated conditions. Therefore, we conducted a bi-directional two-sample MR analysis to assess causal associations between sarcopenia, COPD, and hospitalization risk, employing appendicular lean mass (ALM), hand grip strength (HGS), usual walking pace (UWP), and moderate-to-vigorous physical activity (MVPA).

2. Methods

2.1 Study Design

As shown in Fig. 1, a two-sample bidirectional MR analysis was conducted to examine the likelihood of sarcopenia-related traits causally impacting COPD and hospitalization risk, and vice versa. Subsequently, replication analyses were performed to validate the results from the initial MR analysis. The MR analysis is based on 3 key assumptions: (1) Independence, where the IVs must not be linked to any known confounders; (2) Relevance, where IVs must be associated with the exposures; (3) Exclusivity, where IVs must influence the outcomes exclusively through the exposures [18]. Reporting procedures adhered strictly to Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization (STROBE-MR) guidelines [19]. Since this study reanalyzed previously published data, no additional ethical approval was required.

2.2 Data Sources

Since published genome-wide association studies (GWAS) data (https://www.ebi.ac.uk/gwas/) targeting sarcopenia are lacking, we explored causal relationships using sarcopenia-related traits. Based on the European Working Group on Sarcopenia in Older People 2 (EWGSOP2) consensus, sarcopenia diagnosis involves three key domains: muscle quantity/quality, physical performance, and muscle strength [20]. Therefore, in this MR analysis, we selected sarcopenia-related traits that reflect these three domains: ALM representing low muscle quantity, HGS indicating low muscle strength, and UWP and MVPA denoting low physical performance. These criteria were used as proxies for sarcopenia in this study. This approach enhances the accuracy and comprehensiveness of MR analyses assessing sarcopenia [21].

For this analysis, IVs were identified using publicly available GWAS summary statistics: ALM (n = 244,730) and MVPA (n = 377,234) were obtained from European Bioinformatics Institute, while right HGS (n = 461,089), left HGS (n = 461,026) and UWP (n = 459,915) were from UK Biobank, a prospective cohort comprising >500,000 cases of European ethnicity [22]. Detailed protocols for measuring these parameters are available in the UK Biobank Assessment Centre manual, with most observations adjusted for age, gender, and other associated covariates. GWAS data for COPD (n = 193,638) and COPD-related hospitalizations (n = 218,792), utilized in primary MR analysis, were acquired from the FinnGen consortium (https://www.finngen.fi/en) [23]. COPD was defined based on the International Classification of Diseases (ICD) codes obtained from Finland’s national registries. For mixed-model logistic regression analyses, the Scalable and Accurate Implementation of the Generalized mixed model (SAIGE) approach was used in FinnGen data, with adjustments for sex, age, genotype batch, and 10 principal components [23].

Additionally, another GWAS dataset for COPD (n = 361,194) sourced from the UK Biobank was utilized for MR validation analysis. As the exposure and outcome datasets in the primary MR analysis were sourced from different sample repositories (with no sample geographic or temporal overlap between UK Biobank and FinnGen, and FinnGen officially confirming that its samples are independent of other international cohorts), there was no sample overlap, ensuring the independence of our analyses. All GWAS data were based on populations of European descent. The sources of the specific phenotypic data are summarized in Table 1.

2.3 Instrumental Variables (IVs) Selection

The IVs were selected following the three core assumptions of MR analysis [18]: the IV must be substantially related to exposure, must be independent of confounding factors, and should exclusively affect the outcome through exposure, thereby minimizing horizontal pleiotropy. The criteria used for IV selection were as follows:

(1) Single nucleotide polymorphisms (SNPs) exhibiting genome-wide significant association (p 5 \times 10^-8) with the exposure were selected, a threshold widely accepted in the GWAS to minimize false positives. In cases with limited SNP availability, a less stringent criterion of p 1 \times 10^-5 was applied.

(2) A linkage disequilibrium correlation coefficient of r² 0.001 with a clumping window >10,000 kb was applied to ensure the independence of selected SNPs [24].

(3) The strength of each IV was evaluated using an F statistic >10, computed as F = \beta²(exposure)/SE²(exposure) [25].

(4) SNPs linked to outcome factors were excluded, and proxy SNPs were absent.

(5) Harmonization of SNPs between outcome and exposure datasets was conducted, removing palindromic SNPs and incompatible alleles.

(6) IVs associated with known confounding factors, including smoking, air pollution, blood lipid levels, alcohol consumption, body mass index, were excluded using PhenoScanner v2. Comprehensive data on selected IVs are given in Supplementary Table 1. All of our analyses in PhenoScanner v2 were completed by March 8, 2024, after which date the site was discontinued, but its data source (GWAS Catalog) is still separately accessible even when PhenoScanner is not available. In this case, we directly used the GWAS Catalog (https://www.ebi.ac.uk/gwas/home) as an alternative.

2.4 Univariable, Two-Sample, and Bi-Directional MR Analyses

A random effects inverse-variance weighted (IVW) analysis was used to examine causal links between the exposures and outcomes, as this approach assumes that the selected SNPs represent valid IVs and offer greater precision under these conditions [26]. Supplementary analysis includes the weighted mode, simple mode, weighted median, and MR-Egger approaches [27]. Using multiple MR approaches improves the robustness of the causal estimates across various analytical scenarios.

2.5 Sensitivity Analysis

Three approaches were used for sensitivity analysis to ensure the robustness of our results: evaluation of heterogeneity, assessment of horizontal pleiotropy, and “leave-one-out (LOO)” analysis. Heterogeneity was evaluated by applying Cochran’s Q test; significant heterogeneity (p 0.05) was observed, a random-effects IVW was used for subsequent MR analysis, whereas a fixed-effects IVW was used in cases of no significant heterogeneity [28]. Horizontal pleiotropy was examined based on the MR-Egger intercept test. The Mendelian Randomization Pleiotropy RESidual Sum and Outlier (MR-PRESSO) global test determined potential pleiotropy [29, 30]. Potential outliers SNPs were identified employing the MR-PRESSO outlier test. If an outlier was detected (p 0.05), causal effects were re-estimated after eliminating the outlier SNPs. A p-value 0.05 indicated significant pleiotropic bias. To further evaluate the influences of each SNP on the overall causal estimates, a LOO analysis was conducted sequentially, excluding one SNP at a time [31].

2.6 Statistical Analysis

Two-sample MR analysis was performed through R (version 4.4.1; R Foundation for Statistical Computing, Vienna, Austria), using the ‘TwoSampleMR’ [32] and ‘MR-PRESSO’ [30] packages. These packages are accessible through the Comprehensive R Archive Network (CRAN). For dichotomous variable outcome, results were presented as odds ratios (ORs) with 95% confidence intervals (CIs) to elucidate the causal links between sarcopenia-related traits and both COPD and hospitalization risk. A significance threshold was set at a p-value 0.05. To enhance statistical power, a False Discovery Rate (FDR) was adjusted for multiple comparisons, and the risk of Type I error was minimized [33]. An FDR-adjusted p-value of 0.05 based on IVW results indicated a significant causal relationship. FDR analysis was conducted using the BioLadder platform (https://www.bioladder.cn).

3. Results

3.1 Screening of Instrumental Variables Based on Genome-Wide Significance

The screening process yielded 27–356 IVs that achieved genome-wide significance. All selected SNPs exhibited F statistics exceeding 10, ranging from 19.5 to 772.9, indicating the strength of genetic instruments. Through PhenoScanner, 108 SNPs were identified as potentially related to confounding factors. Specific details on these IVs are provided in Supplementary Table 1.

3.2 The Primary MR Analysis

In the forward MR analysis, sarcopenia-associated traits were used as exposure variables, while COPD and hospitalization risk were outcome variables. The IVW results indicated a causal influence of sarcopenia-associated traits on COPD and hospitalization risk (all p 0.05, all p _FDR 0.05; Table 2, Fig. 2). Specifically, high ALM, HGS, UWP, and MVPA were linked to reduced risks of COPD. A significant negative causal relationship was also observed with COPD-related hospitalization risk, suggesting that sarcopenia-related traits may influence hospitalization rates. Additional MR analyses overwhelmingly supported these causal effect trends (Fig. 2).

In the reverse MR analysis, genetically predicted COPD and hospitalization were examined as exposure variables, with sarcopenia-associated traits as outcomes to elucidate potential reverse causal effects. The IVW analysis suggested a significant negative correlation, indicating that COPD was associated with diminished UWP (OR_UWP = 0.993, 95% CI: 0.989–0.997, p = 0.001, p _FDR = 0.015) (Table 3, Fig. 3). However, no causal effects of COPD or hospitalization on ALM, HGS, and MVPA were observed (p _FDR > 0.05, Table 3). Furthermore, no evidence supported the causal effect of COPD-related hospitalization on UWP. Complete MR findings, including analysis using the weighted median, simple mode, weighted modes, and MR-Egger approaches, are illustrated in Fig. 3.

Following several sensitivity analyses, heterogeneity was assessed using Cochran’s Q-test based on the MR-Egger and IVW frameworks, while MR-PRESSO and MR-Egger assessments evaluated horizontal pleiotropy. Heterogeneity was observed in the MR analysis assessing the effects of COPD and hospitalization on ALM and MVPA, as well as the impact of COPD on left HGS, requiring the utilization of random-effects IVW for these associations (Table 4). No significant heterogeneity was detected in the other two-sample MR analyses (p > 0.05).

Additionally, no evidence of horizontal pleiotropy was found across the two-sample MR analyses (p > 0.05). This suggests that the IVs likely impacted outcomes exclusively through the designated exposures, with no effect through alternate pathways. The LOO sensitivity analyses further confirmed that no single SNP influences the MR results (Supplementary Fig. 1), reinforcing the robustness of our findings.

3.3 The MR Validation Analysis

In the MR validation analysis, the causal influence of ALM, HGS, UWP, and MVPA on COPD was successfully replicated, with effect directions consistent with those observed in the primary MR analysis (Supplementary Tables 2,3). No causal impacts of COPD on sarcopenia-associated traits were observed, and the observed association between COPD and UWP was not replicated in the MR validation analysis (Supplementary Tables 2,3). However, sensitivity analyses validated the robustness of these outcomes (Supplementary Tables 2,3 and Supplementary Fig. 2). According to these findings, it was concluded that sarcopenia negatively impacts COPD causally.

4. Discussion

This study utilized an univariable two-sample bi-directional MR to systematically elucidate the causal links between sarcopenia-associated traits, including ALM, HGS, UWP, and MVPA and the risk of COPD and hospitalization. The findings revealed a significant negative causal relationship, demonstrating that genetically determined reduction in ALM, HGS, UWP, and MVPA increases the risk of COPD and hospitalization. Furthermore, these associations were successfully validated through replication analyses. Although the primary MR analysis suggested a negative correlation between COPD and diminished UWP, this finding was not replicated in the MR validation analysis. To our knowledge, this is the first bi-directional MR study to comprehensively evaluate the causal link between sarcopenia-associated traits and COPD and hospitalization risk, while accounting for potential confounders.

Previous observational studies have indicated that sarcopenia elevates the risk of COPD. A prospective, population-based cohort investigation undertaken in Rotterdam, The Netherlands, revealed a higher incidence of COPD among sarcopenic (26.9%) and pre-sarcopenic (29.1%) patients relative to the nonsarcopenic (13.4%) individuals [34]. Similarly, another study involving 469,830 participants from the UK Biobank indicated that gait-muscle group individuals had a 4.16-fold higher risk of COPD than those with the normal physical ability (HR: 4.16, 95% CI: 2.59–6.70), followed by those with severe sarcopenia (HR: 3.85, 95% CI: 2.24–6.62). Sarcopenia in this study was described based on the combination of 3 indicators of physical function: muscle mass, grip strength, and gait speed [35]. The Invecchiare in Chianti (InCHIANTI, Aging in the Chianti Region) study, which longitudinally analyzed 538 participants, revealed that sarcopenia was correlated with a higher hospitalization rate (60% vs. 48%, p = 0.087). Even after adjusting for potential confounders, sarcopenia remained notably associated with hospital admission (HR: 1.57; 95% CI: 1.03–2.41) [36]. Furthermore, a retrospective analysis of 174,808 COPD patients in the United States in 2011 reported associations between muscle loss and increased morbidity and mortality, with sarcopenic patients experiencing longer hospital stays and higher associated costs compared to non-sarcopenic patients. These observations are consistent with our results, indicating a causal association between sarcopenia and increased risks of COPD and hospitalization. According to EWGSOP definitions, the diagnosis of sarcopenia requires meeting both low grip strength and low muscle mass, with severe sarcopenia characterized by reduced gait speed. Our findings demonstrate the substantial negative causal association between ALM, MVPA, UWP, and HGS with COPD and hospitalization, supporting the conclusion that sarcopenia increases the risk of COPD and related hospitalizations.

Several systematic reviews and meta-analyses of prior observational studies have suggested an association between COPD and an elevated risk of sarcopenia [11, 16, 37]. While epidemiological surveys across varied populations have shown different prevalence rates of sarcopenia among COPD patients, meta-analyses consistently underscore its frequent co-occurrence. However, in our study, no genetic evidence supports the causal impact of COPD and hospitalization on sarcopenia-related traits. While the primary MR analysis suggested a negative correlation between COPD and diminished UWP, this finding was not replicated in the MR validation analysis. Given our study’s rigor in applying FDR correction for multiple comparisons and considering the complex interrelations between COPD and sarcopenia, these results warrant cautious interpretation. Further investigation is needed to assess these associations more thoroughly in future studies. Sarcopenia-related traits, like ALM and HGS, serve as key predictors of muscle mass, and low muscle mass has been linked to poor pulmonary function [38, 39]. A cross-sectional study analyzing 452 COPD patients reported a significant association between reduced lean mass and elevated risk of emphysema, a crucial phenotype of COPD [40]. Meanwhile, UWP and MVPA are used as a measure of physical activity performance levels. Evidence suggests that prolonged sedentary behavior may lead to an inadequate physical exercise, adversely affecting both muscle and lung functions, thereby increasing the risk of COPD [41, 42, 43]. Conversely, higher physical activity has been associated with a lower risk of COPD, likely through preserving lean mass and reducing oxidative stress and minimizing chronic airway inflammation [44, 45, 46, 47, 48]. Moreover, MVPA has been reported to have positive effects on lung function at the population level [49]. Taken together, these interrelated clinical factors provided a considerable basis for the observed association between sarcopenia and the higher incidence of COPD and hospitalization risk.

The occurrence of comorbidities in sarcopenia and COPD involves a multidimensional biological interaction, with systemic inflammation playing a crucial role. Increased levels of serum pro-inflammatory factors, such as C-reactive protein (CRP), Interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-\alpha) have been consistently found in COPD patients, where they accelerate myofibrillar protein degradation by upregulating the expression of muscle-specific ubiquitin ligases (MuRF1 and MAFbx) through activation of the Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-\kappaB) pathway [50, 51]. Oxidative stress may be another key mechanism, manifested by the excessive accumulation of reactive oxygen species (ROS) in COPD patients. ROS not only impairs mitochondrial function and muscle energy metabolism but also synergizes with inflammatory factors to activate inflammatory signaling pathways such as p38 Mitogen-Activated Protein Kinase (p38 MAPK), forming a vicious circle that aggravates muscle atrophy and dysfunction [52, 53]. Additionally, an imbalance in actin regulation further elucidates the link between sarcopenia and COPD. For example, irisin promotes muscle growth and is usually elevated during physical activity but is reduced in patients with COPD, whereas levels of myostatin, an inhibitor of muscle regeneration, are elevated [54]. This imbalance skews muscle metabolism towards catabolic processes, accelerating the onset of sarcopenia. However, despite these promising insights, the specific pathophysiological mechanisms linking sarcopenia and COPD remain poorly understood.

The growing understanding of the biological mechanisms linking sarcopenia and COPD carries significant therapeutic implications. Evidence suggested that managing sarcopenia in COPD patients is both feasible and effective through well-structured exercise programs supported by a balanced nutritional intake. Various other events, such as resistance training, aerobic exercise, and pulmonary rehabilitation have been found to improve muscle strength and endurance, while adequate protein intake, vitamin D, omega-3s, and calories help preserve muscle mass [54, 55, 56, 57]. Given the possible negative causal association observed in our study, combining these interventions may maximize therapeutic outcomes and reduce the risk of sarcopenia-related COPD exacerbation and hospitalization. These evidence-based strategies, based on the latest research findings, provide clinicians with a crucial framework for public health decision-making aimed at managing sarcopenia to reduce its impact on COPD patients.

The current study has several significant strengths. Firstly, it comprehensively evaluated critical features related to muscle quality, muscle strength, and physical performance, while also including hospitalization phenotypes alongside COPD, thereby enhancing the rigor and representativeness of the study. Secondly, MR studies, often considered natural randomized controlled trials, offer more robust findings than traditional observational studies. Furthermore, we executed extensive validation and sensitivity analyses to confirm the result’s robustness, especially when heterogeneity was observed: a random-effects IVW was used to minimize its effect. Finally, we effectively excluded confounding factors such as air pollution, smoking, alcohol consumption, body mass index (BMI), and blood lipid levels, mitigating potential biases.

Nonetheless, we acknowledge certain limitations in our study. Specifically, our study population was limited to participants of European ancestry. Due to significant genetic differentiation between East Asian and European populations, with about 12% of SNPs exhibiting significant Fixation Index (FST) differences (indicative of considerable allele frequency differences), which may limit the generalizability of the results across different ethnic groups [58]. Additionally, stratification based on common factors such as age and sex were unattainable due to constraints in the available summarized GWAS data. Moreover, the causal links between COPD-related hospitalization and sarcopenia-associated traits could not be validated due to a lack of relevant data. Lastly, there are inherent limitations in the MR approach. However, the use of MR-PRESSO helped to reduce the risk of horizontal pleiotropy by identifying and excluding outlier SNPs, its validity relies on the assumption of linearity and cannot entirely exclude the possibility of residual pleiotropy arising from nonlinear causation or epigenetic modulation [30].

It should also be noted that, in the partial MR analysis, we employed a relatively lenient threshold (p 1 \times 10^-5) for IV selection due to the limited number of variants meeting the more stringent threshold (p 5 \times 10^-8). Although all selected SNPs had F-values >10, thus meeting the threshold to reduce weak instrumental bias, and pleiotropic SNPs were excluded, the potential for weak instrument bias remains. Additionally, individual leave-one-out plots demonstrated that there are potentially influential SNPs driving the causal link between exposures and outcomes, suggesting the possibility of unrecognized pleiotropy in some SNPs that could influence the stability of our findings. In future studies, using a Polygenic Risk Score, colocalization analyses to prioritize causal SNPs, or the integration of proteomic MR to enhance biological specificity could further validate findings in this area.

5. Conclusion

In conclusion, this study indicates a causal association between sarcopenia and elevated risks of COPD and hospitalization, providing reliable genetic evidence for the adverse impacts of sarcopenia on both outcomes within the European population. These findings underscore the clinical and public health importance of targeting sarcopenia improvement as a promising strategy to reduce the incidence of COPD and minimize hospitalization risk.

Key Points

\bullet This study employed a bi-directional two-sample MR approach to systematically investigate causal relationships between sarcopenia-related traits (ALM, HGS, UWP, MVPA) and both COPD and hospitalization risk, overcoming limitations of observational studies.

\bullet Genetically predicted lower ALM, reduced HGS, slower UWP, and decreased MVPA significantly increased the risk of COPD and COPD-related hospitalization, supporting a unidirectional causal effect of sarcopenia on COPD exacerbation.

\bullet The findings highlight that intervention targeting muscle preservation (e.g., resistance training, nutritional support) may reduce COPD incidence and hospitalization burden, offering practical strategies for mitigating age-related cardiopulmonary morbidity.

\bullet Robust sensitivity analyses (MR-PRESSO, leave-one-out), FDR correction, and replication using independent UK Biobank dataset strengthened causal inferences, with minimal pleiotropy or heterogeneity observed (Cochran’s Q p > 0.05 for most traits).

\bullet While emphasizing sarcopenia’s causal role, this study underscores the need for multi-ethnic cohorts, mechanistic exploration of muscle-lung interactions (e.g., oxidative stress pathways), and trials testing targeted sarcopenia therapies in COPD populations.

Availability of Data and Materials

All data generated or analyzed during this study are available from the corresponding author on reasonable request.

References

[1] Bruyère O, Beaudart C, Ethgen O, Reginster JY, Locquet M. The health economics burden of sarcopenia: a systematic review. Maturitas. 2019; 119: 61–69. https://doi.org/10.1016/j.maturitas.2018.11.003.

[2] World Health Organization. Chronic obstructive pulmonary disease (COPD). 2020. Available at: https://www.who.int/news-room/fact-sheets/detail/chronic-obstructive-pulmonary-disease-(copd) (Accessed: 20 March 2024).

[3] Chen S, Kuhn M, Prettner K, Yu F, Yang T, Bärnighausen T, et al. The global economic burden of chronic obstructive pulmonary disease for 204 countries and territories in 2020-50: a health-augmented macroeconomic modelling study. The Lancet. Global Health. 2023; 11: e1183–e1193. https://doi.org/10.1016/S2214-109X(23)00217-6.

[4] Pizarro C, Skowasch D. Comorbidities in Chronic Obstructive Pulmonary Disease. Deutsche Medizinische Wochenschrift. 2019; 144: 28–33. https://doi.org/10.1055/a-0659-0999. (In German)

[5] Barnes PJ, Celli BR. Systemic manifestations and comorbidities of COPD. The European Respiratory Journal. 2009; 33: 1165–1185. https://doi.org/10.1183/09031936.00128008.

[6] Global Initiative for Chronic Obstructive Lung Disease. 2020 Global Strategy for Prevention, Diagnosis and Management of COPD. 2019. Available at: https://goldcopd.org/gold-reports/ (Accessed: 15 January 2024).

[7] Divo MJ, Casanova C, Marin JM, Pinto-Plata VM, de-Torres JP, Zulueta JJ, et al. COPD comorbidities network. The European Respiratory Journal. 2015; 46: 640–650. https://doi.org/10.1183/09031936.00171614.

[8] Nussbaumer-Ochsner Y, Rabe KF. Systemic manifestations of COPD. Chest. 2011; 139: 165–173. https://doi.org/10.1378/chest.10-1252.

[9] Chen LK, Woo J, Assantachai P, Auyeung TW, Chou MY, Iijima K, et al. Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment. Journal of the American Medical Directors Association. 2020; 21: 300–307.e2. https://doi.org/10.1016/j.jamda.2019.12.012.

[10] Vogelmeier CF, Criner GJ, Martinez FJ, Anzueto A, Barnes PJ, Bourbeau J, et al. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Lung Disease 2017 Report: GOLD Executive Summary. European Respiratory Journal. 2017; 49: 1700214. https://doi.org/10.1183/13993003.00214-2017.

[11] Jones SE, Maddocks M, Kon SSC, Canavan JL, Nolan CM, Clark AL, et al. Sarcopenia in COPD: prevalence, clinical correlates and response to pulmonary rehabilitation. Thorax. 2015; 70: 213–218. https://doi.org/10.1136/thoraxjnl-2014-206440.

[12] Rolland Y, Czerwinski S, Abellan Van Kan G, Morley JE, Cesari M, Onder G, et al. Sarcopenia: its assessment, etiology, pathogenesis, consequences and future perspectives. The Journal of Nutrition, Health & Aging. 2008; 12: 433–450. https://doi.org/10.1007/BF02982704.

[13] Costa TM, Costa FM, Moreira CA, Rabelo LM, Boguszewski CL, Borba VZ. Sarcopenia in COPD: relationship with COPD severity and prognosis. Jornal Brasileiro de Pneumologia: Publicação Oficial da Sociedade Brasileira de Pneumologia e Tisiologia. 2015; 41: 415–421. https://doi.org/10.1590/S1806-37132015000000040.

[14] Attaway AH, Welch N, Hatipoğlu U, Zein JG, Dasarathy S. Muscle loss contributes to higher morbidity and mortality in COPD: An analysis of national trends. Respirology. 2021; 26: 62–71. https://doi.org/10.1111/resp.13877.

[15] Gómez-Martínez M, Rodríguez-García W, González-Islas D, Orea-Tejeda A, Keirns-Davis C, Salgado-Fernández F, et al. Impact of Body Composition and Sarcopenia on Mortality in Chronic Obstructive Pulmonary Disease Patients. Journal of Clinical Medicine. 2023; 12: 1321. https://doi.org/10.3390/jcm12041321.

[16] Sepúlveda-Loyola W, Osadnik C, Phu S, Morita AA, Duque G, Probst VS. Diagnosis, prevalence, and clinical impact of sarcopenia in COPD: a systematic review and meta-analysis. Journal of Cachexia, Sarcopenia and Muscle. 2020; 11: 1164–1176. https://doi.org/10.1002/jcsm.12600.

[17] Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Human Molecular Genetics. 2014; 23: R89–R98. https://doi.org/10.1093/hmg/ddu328.

[18] Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ (Clinical Research Ed.). 2018; 362: k601. https://doi.org/10.1136/bmj.k601.

[19] Skrivankova VW, Richmond RC, Woolf BAR, Yarmolinsky J, Davies NM, Swanson SA, et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA. 2021; 326: 1614–1621. https://doi.org/10.1001/jama.2021.18236.

[20] Cruz-Jentoft AJ, Bahat G, Bauer J, Boirie Y, Bruyère O, Cederholm T, et al. Sarcopenia: revised European consensus on definition and diagnosis. Age and Ageing. 2019; 48: 16–31. https://doi.org/10.1093/ageing/afy169.

[21] Liu M, Chen P, Yang C, Sun G. Standardizing trait selection in Mendelian randomization studies concerning sarcopenia. Journal of Cachexia, Sarcopenia and Muscle. 2024; 15: 1220–1221. https://doi.org/10.1002/jcsm.13463.

[22] Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Medicine. 2015; 12: e1001779. https://doi.org/10.1371/journal.pmed.1001779.

[23] Kurki MI, Karjalainen J, Palta P, Sipilä TP, Kristiansson K, Donner KM, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023; 613: 508–518. https://doi.org/10.1038/s41586-022-05473-8.

[24] Sved JA, Hill WG. One Hundred Years of Linkage Disequilibrium. Genetics. 2018; 209: 629–636. https://doi.org/10.1534/genetics.118.300642.

[25] Bowden J, Del Greco M F, Minelli C, Zhao Q, Lawlor DA, Sheehan NA, et al. Improving the accuracy of two-sample summary-data Mendelian randomization: moving beyond the NOME assumption. International Journal of Epidemiology. 2019; 48: 728–742. https://doi.org/10.1093/ije/dyy258.

[26] Slob EAW, Burgess S. A comparison of robust Mendelian randomization methods using summary data. Genetic Epidemiology. 2020; 44: 313–329. https://doi.org/10.1002/gepi.22295.

[27] Bowden J, Del Greco M F, Minelli C, Davey Smith G, Sheehan NA, Thompson JR. Assessing the suitability of summary data for two-sample Mendelian randomization analyses using MR-Egger regression: the role of the I² statistic. International Journal of Epidemiology. 2016; 45: 1961–1974. https://doi.org/10.1093/ije/dyw220.

[28] Greco M FD, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in Mendelian randomisation studies with summary data and a continuous outcome. Statistics in Medicine. 2015; 34: 2926–2940. https://doi.org/10.1002/sim.6522.

[29] Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. International Journal of Epidemiology. 2015; 44: 512–525. https://doi.org/10.1093/ije/dyv080.

[30] Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nature Genetics. 2018; 50: 693–698. https://doi.org/10.1038/s41588-018-0099-7.

[31] Burgess S, Bowden J, Fall T, Ingelsson E, Thompson SG. Sensitivity Analyses for Robust Causal Inference from Mendelian Randomization Analyses with Multiple Genetic Variants. Epidemiology. 2017; 28: 30–42. https://doi.org/10.1097/EDE.0000000000000559.

[32] Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife. 2018; 7: e34408. https://doi.org/10.7554/eLife.34408.

[33] Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I. Controlling the false discovery rate in behavior genetics research. Behavioural Brain Research. 2001; 125: 279–284. https://doi.org/10.1016/s0166-4328(01)00297-2.

[34] Trajanoska K, Schoufour JD, Darweesh SK, Benz E, Medina-Gomez C, Alferink LJ, et al. Sarcopenia and Its Clinical Correlates in the General Population: The Rotterdam Study. Journal of Bone and Mineral Research. 2018; 33: 1209–1218. https://doi.org/10.1002/jbmr.3416.

[35] Petermann-Rocha F, Ho FK, Welsh P, Mackay D, Brown R, Gill JMR, et al. Physical capability markers used to define sarcopenia and their association with cardiovascular and respiratory outcomes and all-cause mortality: A prospective study from UK Biobank. Maturitas. 2020; 138: 69–75. https://doi.org/10.1016/j.maturitas.2020.04.017.

[36] Bianchi L, Ferrucci L, Cherubini A, Maggio M, Bandinelli S, Savino E, et al. The Predictive Value of the EWGSOP Definition of Sarcopenia: Results From the InCHIANTI Study. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences. 2016; 71: 259–264. https://doi.org/10.1093/gerona/glv129.

[37] Yu Z, He J, Chen Y, Zhou Z, Wang L. Chronic obstructive pulmonary disease as a risk factor for sarcopenia: A systematic review and meta-analysis. PLoS ONE. 2024; 19: e0300730. https://doi.org/10.1371/journal.pone.0300730.

[38] Graumam RQ, Pinheiro MM, Nery LE, Castro CHM. Increased rate of osteoporosis, low lean mass, and fragility fractures in COPD patients: association with disease severity. Osteoporosis International. 2018; 29: 1457–1468. https://doi.org/10.1007/s00198-018-4483-z.

[39] Machado FVC, Spruit MA, Groenen MTJ, Houben-Wilke S, van Melick PP, Hernandes NA, et al. Frequency and functional translation of low muscle mass in overweight and obese patients with COPD. Respiratory Research. 2021; 22: 93. https://doi.org/10.1186/s12931-021-01689-w.

[40] Rutten EPA, Grydeland TB, Pillai SG, Wagers S, Dirksen A, Coxson HO, et al. Quantitative CT: Associations between Emphysema, Airway Wall Thickness and Body Composition in COPD. Pulmonary Medicine. 2011; 2011: 419328. https://doi.org/10.1155/2011/419328.

[41] Cao Z, Xu C, Zhang P, Wang Y. Associations of sedentary time and physical activity with adverse health conditions: Outcome-wide analyses using isotemporal substitution model. eClinicalMedicine. 2022; 48: 101424. https://doi.org/10.1016/j.eclinm.2022.101424.

[42] Hill K, Gardiner PA, Cavalheri V, Jenkins SC, Healy GN. Physical activity and sedentary behaviour: applying lessons to chronic obstructive pulmonary disease. Internal Medicine Journal. 2015; 45: 474–482. https://doi.org/10.1111/imj.12570.

[43] Lei Y, Zou K, Xin J, Wang Z, Liang K, Zhao L, et al. Sedentary behavior is associated with chronic obstructive pulmonary disease: A generalized propensity score-weighted analysis. Medicine. 2021; 100: e25336. https://doi.org/10.1097/MD.0000000000025336.

[44] Behrens G, Matthews CE, Moore SC, Hollenbeck AR, Leitzmann MF. Body size and physical activity in relation to incidence of chronic obstructive pulmonary disease. Canadian Medical Association Journal. 2014; 186: E457–E469. https://doi.org/10.1503/cmaj.140025.

[45] Edwards MK, Loprinzi PD. Systemic inflammation as a function of the individual and combined associations of sedentary behaviour, physical activity and cardiorespiratory fitness. Clinical Physiology and Functional Imaging. 2018; 38: 93–99. https://doi.org/10.1111/cpf.12388.

[46] Hansen NB, Henriksen M, Dall CH, Vest S, Larsen L, Suppli Ulrik C, et al. Physical activity, physical capacity and sedentary behavior among asthma patients. European Clinical Respiratory Journal. 2022; 9: 2101599. https://doi.org/10.1080/20018525.2022.2101599.

[47] Metsios GS, Moe RH, Kitas GD. Exercise and inflammation. Best Practice & Research. Clinical Rheumatology. 2020; 34: 101504. https://doi.org/10.1016/j.berh.2020.101504.

[48] Miyazaki H, Oh-ishi S, Ookawara T, Kizaki T, Toshinai K, Ha S, et al. Strenuous endurance training in humans reduces oxidative stress following exhausting exercise. European Journal of Applied Physiology. 2001; 84: 1–6. https://doi.org/10.1007/s004210000342.

[49] Li LK, Cassim R, Perret JL, Dharmage SC, Lowe AJ, Lodge CJ, et al. The longitudinal association between physical activity, strength and fitness, and lung function: A UK Biobank cohort study. Respiratory Medicine. 2023; 220: 107476. https://doi.org/10.1016/j.rmed.2023.107476.

[50] Byun MK, Cho EN, Chang J, Ahn CM, Kim HJ. Sarcopenia correlates with systemic inflammation in COPD. International Journal of Chronic Obstructive Pulmonary Disease. 2017; 12: 669–675. https://doi.org/10.2147/COPD.S130790.

[51] Lin B, Bai L, Wang S, Lin H. The Association of Systemic Interleukin 6 and Interleukin 10 Levels with Sarcopenia in Elderly Patients with Chronic Obstructive Pulmonary Disease. International Journal of General Medicine. 2021; 14: 5893–5902. https://doi.org/10.2147/IJGM.S321229.

[52] Ma K, Huang F, Qiao R, Miao L. Pathogenesis of sarcopenia in chronic obstructive pulmonary disease. Frontiers in Physiology. 2022; 13: 850964. https://doi.org/10.3389/fphys.2022.850964.

[53] Mano Y, Tsukamoto M, Wang KY, Nabeshima T, Kosugi K, Tajima T, et al. Oxidative stress causes muscle structural alterations via p38 MAPK signaling in COPD mouse model. Journal of Bone and Mineral Metabolism. 2022; 40: 927–939. https://doi.org/10.1007/s00774-022-01371-1.

[54] Han L, Li P, He Q, Yang C, Jiang M, Wang Y, et al. Revisiting Skeletal Muscle Dysfunction and Exercise in Chronic Obstructive Pulmonary Disease: Emerging Significance of Myokines. Aging and Disease. 2024; 15: 2453–2469. https://doi.org/10.14336/AD.2023.1125.

[55] Nan Y, Zhou Y, Dai Z, Yan T, Zhong P, Zhang F, et al. Role of nutrition in patients with coexisting chronic obstructive pulmonary disease and sarcopenia. Frontiers in Nutrition. 2023; 10: 1214684. https://doi.org/10.3389/fnut.2023.1214684.

[56] Pancera S, Bianchi LNC, Porta R, Villafañe JH, Buraschi R, Lopomo NF. Muscle function and functional performance after pulmonary rehabilitation in patients with chronic obstructive pulmonary disease: a prospective observational study. Scientific Reports. 2022; 12: 16386. https://doi.org/10.1038/s41598-022-20746-y.

[57] Zheng HY, Zhang HY, Wu KH, Cai WJ, Li ZZ, Song XY. Efficacy of nutrient supplements in managing malnutrition and sarcopenia in Chronic Obstructive Pulmonary Disease (COPD) patients: a protocol for systematic review and meta-analysis. Systematic Reviews. 2025; 14: 58. https://doi.org/10.1186/s13643-025-02801-7.

[58] Sirugo G, Williams SM, Tishkoff SA. The Missing Diversity in Human Genetic Studies. Cell. 2019; 177: 26–31. https://doi.org/10.1016/j.cell.2019.02.048.