Association between sarcopenia and locomotive syndrome in rheumatoid arthritis patients: A multicenter observational study (T‐FLAG)

Sarcopenia is defined as a continuous and widespread decline in skeletal muscle mass and strength, and factors contributing to muscle loss are diverse, including aging, inactivity, poor nutrition, hormonal imbalances, cachexia, and inflammatory conditions such as rheumatoid arthritis (RA).1 In RA, a systemic autoimmune inflammatory disease, tumor necrosis factor-alpha and other inflammatory cytokines that have a catabolic impact on skeletal muscle are generated.2 Consequently, individuals with RA are more vulnerable to musculoskeletal disorders due to joint damage and physical impairment. Locomotive syndrome (LS) is defined as decreased mobility resulting from musculoskeletal disorders such as sarcopenia and is associated with an increased need for care and potential immobility.3 Notably, the prevalence of LS among RA patients is higher than that in the general population.4 As this group of patients is also aging,5 measures focusing on LS in RA patients are of paramount importance. Sarcopenia and LS are both musculoskeletal conditions that potentially result in confinement to bed and adversely impact the quality of life and overall health longevity.6 To guide our investigation, we have adopted a conceptual framework that outlines the various factors influencing LS in RA patients (Figure S1). This framework includes sarcopenia, demographic factors, medications, comorbidities, disease activity measures, laboratory parameters, and muscle strength assessment, providing a comprehensive model for understanding the interplay between these conditions. We hypothesize that higher disease activity in RA significantly drives both sarcopenia and LS. This study aims to elucidate the association between sarcopenia and LS in RA patients and to identify other distinguishing factors between these conditions. Between June and August 2023, a cohort of 696 consecutive RA patients visited the Japanese Red Cross Aichi Medical Center Nagoya Daiichi Hospital (Nagoya), Japan Community Health Care Organization Kani Tono Hospital (Kani), and Yokkaichi Municipal Hospital (Yokkaichi). The present study was conducted as part of the Tsurumai-Frailty and Locomotive syndrome of rheumatoid Arthritis for Globalization [T-FLAG] study, an ongoing multicenter observational study designed to analyze frailty and LS in RA patients in clinical practice (start date: June 1, 2020). This study was approved by the Ethics Committees of Nagoya University School of Medicine (2017-0271), Nagoya (2020-451), Kani (20110901), and Yokkaichi (2017-29). Informed consent was obtained from all patients. Among these patients, data on clinical characteristics, including scores for the 25-question Geriatric Locomotive Function Scale (GLFS-25)7 and SARC-F,8 were available for 654 patients, all of whom fulfilled the 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) classification criteria for RA. LS is diagnosed when an individual scores 16 points or higher on the GLFS-25 questionnaire, a tool developed for the early detection of LS. The questionnaire consists of 25 items divided into the following categories: "Body pain," "Movement-related difficulty," "Usual care," "GLFS-5," "Social activities," and "Cognitive."7 Responses are rated on a scale from 0 (no impairment) to 4 (severe impairment), with a total possible score ranging from 0 (no symptoms) to 100 (most severe symptoms). The SARC-F is a self-administered questionnaire consisting of five questions, including Strength (S), Assistance walking (A), Rising from a chair (R), Climbing stairs (C), and Falls (F), rated on a scale from 0 (no difficulty) to 2 (a lot of difficulty) with a total score of 0 (best) to 10 (worst); the cutoff score for sarcopenia is ≥4 points.8 The SARC-F has a very high specificity of 90% for predicting sarcopenia, with a sensitivity of 21%.9 Statistical analyses were performed with EZR.10 A total of 654 patients (female, 71.4%) were included, with a mean age of 68.5 years, disease duration of 12.7 years, Clinical Disease Activity Index (CDAI) score of 6.2, GLFS-25 score of 20.5 points, and SARC-F score of 2.4 points. There were 185 patients with sarcopenia (28.3% of the total), of whom 176 (95.1%) had LS; among 284 patients with LS (43.4% of the total), 176 (62.0%) had sarcopenia. In RA patients with LS, those who had sarcopenia had significantly higher matrix metalloproteinase-3 (MMP-3), visual analog scale (VAS), CDAI, Health Assessment Questionnaire-Disability Index (HAQ-DI), and GLFS-25, and had significantly lower grip strength, compared with those who did not have sarcopenia (Table 1). Sarcopenia without LS (nine patients) tended to have higher GLFS-25 scores compared with robust patients, but had lower GLFS-25 scores compared with LS without sarcopenia (Table 1). The correlation coefficient between SARC-F and GLFS-25 scores was 0.829 (p < .001, Spearman's rank correlation coefficient), indicating a strong positive correlation. Specifically, the highest correlation for each SARC-F question with the GLFS-25 categories was as follows: Strength (S) had the highest correlation with GLFS-5 (0.650), Assistance walking (A) with Movement-related difficulty (0.683), Rising from a chair (R) with Movement-related difficulty (0.782), Climbing stairs (C) with GLFS-5 (0.768), and Falls (F) with Cognitive (0.347; Table S1). Conversely, the Body pain category of the GLFS-25 showed low correlations with each of the SARC-F questions, and Falls (F) had low correlations with all GLFS-25 categories. In receiver operating characteristic curves, the cutoff score of GLFS-25 corresponding to sarcopenia was 22 points (sensitivity 87.6%, specificity 87.4%, the area under the curve [AUC] 0.944, 95% confidence interval [95%CI] 0.928–0.960), and the cutoff score of SARC-F corresponding to LS was 3 points (sensitivity 78.2%, specificity 91.1%, AUC 0.920, 95%CI 0.899–0.941). The proportion of patients with LS and sarcopenia increased with age, increasing CDAI, and worsening stage of sarcopenia and LS, respectively (Figure 1). Multivariable logistic regression analysis revealed significant associations for sarcopenia with disease duration (odds ratio [OR] 0.96, 95% CI 0.93–0.99), BMI (OR 1.11, 95% CI 1.04–1.19), Steinbrocker stage (3/4) (OR 1.99, 95% CI 1.06–3.76), CDAI (OR 1.05, 95% CI 1.00–1.09), and HAQ-DI (OR 1.34, 95% CI 1.27–1.43; Table S2). For LS, significant associations were found with age (OR 1.04, 95% CI 1.02–1.07), BMI (OR 1.07, 95% CI 1.00–1.15), CDAI (OR 1.05, 95% CI 1.01–1.10), and HAQ-DI (OR 1.64, 95% CI 1.49–1.80). This is the first study to investigate the association between sarcopenia and LS in RA patients, aiming to elucidate the specific factors involved. Our multivariable logistic regression analysis identified common factors for both sarcopenia and LS, including high BMI, CDAI, and HAQ-DI. High BMI, although inconsistent with some previous reports,11 can be explained by the emerging concept of sarcopenic obesity,12 underscoring the importance of managing proper BMI through adequate exercise and nutrition. CDAI and HAQ-DI were significantly associated with both conditions, indicating that managing disease activity and maintaining physical function is crucial for preventing and managing both sarcopenia and LS in RA patients. Additionally, patients with LS and sarcopenia had worse disease activity and physical function compared with those with LS without sarcopenia. Factors specific to sarcopenia included Steinbrocker stage, reflecting the extent of joint destruction, and shorter disease duration, suggesting that greater joint damage and rapid muscle loss may occur early in RA onset. Age was identified as a specific factor for LS, reflecting the increased vulnerability to mobility issues with advancing age. Given that the cutoff score of SARC-F for LS (3 points) is lower than that for sarcopenia (4 points), and that the cutoff score of GLFS-25 for sarcopenia (22 points) indicates proximity to LS Stage 3 (24 points) in the later stages, it suggests that LS Stage 2 (i.e., LS, 16 points) encompasses a broader range of functional impairments, including those related to sarcopenia. Additionally, the "Movement-related difficulty" category of GLFS-25, which is associated with the A and R components of SARC-F, was more affected in the later stages of LS, indicating proximity to LS Stage 3.7, 13 This suggests that LS and sarcopenia, while overlapping, have distinct progression patterns, with LS affecting a wider array of physical functions. However, interestingly, the Body pain category of the GLFS-25 showed low correlations with each of the SARC-F questions, and Falls (F) had low correlations with all GLFS-25 categories. These findings suggest that while SARC-F and GLFS-25 capture overlapping aspects of physical function, they also highlight distinct areas of impairment, emphasizing the need for comprehensive assessment tools to accurately evaluate the multifaceted nature of functional decline in RA patients. There are several limitations in this study. First, assessments of locomotive functions, such as the stand-up test, the two-step test, and the Timed Up and Go (TUG) test, were not performed. Second, the present study did not use the criteria for sarcopenia by AWGS, which are widely recognized internationally.14 Although the SARC-F questionnaire is straightforward and can be administered in clinical settings, its low sensitivity may lead to an underestimation of sarcopenia prevalence and affect observed associations. Despite this, our study found a sarcopenia prevalence of 28.3% (mean age 68.5 years) using SARC-F, which is similar to the 28.0% reported by the AWGS definition (mean age 66.1 years).11 Another study reported a prevalence of 14.1% (mean age 63.9 years) using SARC-F.15 Therefore, it is likely that few RA patients have undiagnosed sarcopenia by SARC-F and do not have LS. Future studies should consider using more comprehensive diagnostic criteria, such as those proposed by AWGS, to improve diagnostic accuracy. Third, the cohort was taken from three specific hospitals in Japan, which may limit the generalizability of the findings. This geographic and demographic concentration might affect the applicability of the results to broader RA patient populations. Fourth, the mean duration of disease in our cohort was 12.7 years, which was relatively long and may impact the study results by influencing factors related to LS and sarcopenia. Patients with a longer disease duration may have more severe or advanced stages of RA, potentially leading to a higher prevalence of LS and sarcopenia. Finally, the cross-sectional study design did not allow us to establish a causal link between sarcopenia and LS. In the future, prospective longitudinal studies to monitor RA patients over time will be needed to obtain a deeper understanding of the link between sarcopenia and LS. Our study highlights the complex interplay between sarcopenia and LS in RA patients and underscores the importance of a comprehensive approach to managing these conditions. Future research should explore the impact of different RA treatments on sarcopenia and LS, providing insights into targeted interventions and new directions for managing RA to improve patient outcomes. Y.S., M.S., and Y.O. contributed to Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Visualization, Writing – Original draft preparation, and Writing, Reviewing and Editing. R.S., H.K., Y.O., J.H., T.S., K.T., and S.A. contributed to Investigation, Data curation, and Writing-Reviewing and Editing. S.I. contributed to Supervision, Conceptualization, Methodology, and Writing-Reviewing and Editing. All authors read and approved the final manuscript. We thank Dr. Koji Funahashi, Dr. Hiroshi Koshima, Dr. Nobuyuki Okui, Dr. Hisato Ishikawa, Ms. Sachiko Kato, Ms. Emi Yokota, Ms. Ritsuko Otake, and Ms. Takako Sashikata for their assistance in information collection. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Figure S1. Table S1. Table S2. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.