Introduction
The term “sarcopenia,” derived from the Greek words sarx (flesh) and penia (loss), was first coined by I.H. Rosenberg in 1989 to describe the age-related decline in skeletal muscle mass. Initially conceptualized as a geriatric syndrome, sarcopenia is recognized as a major contributor to frailty, disability, and mortality in the elderly population [1].
The definition and diagnostic criteria for sarcopenia have shifted from a mass-centric to a strength-focused model (Fig. 1). Initially, sarcopenia was defined solely as low muscle mass [2, 3], but evidence showed muscle strength to be a better predictor of adverse outcomes.
This prompted multi-component guidelines by the European Working Group on Sarcopenia in Older People (EWGSOP1), the Foundation for the National Institutes of Health (FNIH), and the Asian Working Group for Sarcopenia (AWGS) [4–6]. Recent statements from the updated EWGSOP2 [7], the Australian and New Zealand Society for Sarcopenia and Frailty Research (ANZSFFR) [8], now place low muscle strength at the core of sarcopenia diagnosis.
Fig. 1The evolution of diagnostic frameworks for sarcopenia. Abbreviations: EWGSOP, European Working Group on Sarcopenia in Older People; FNIH, Foundation for the National Institutes of Health; IWGS, International Working Group on Sarcopenia; ANZSFFR, Australian and New Zealand Society for Sarcopenia and Frailty Research
More recently, the paradigm has shifted from viewing sarcopenia as exclusively age-related to recognizing it as a significant complication of various chronic conditions, a concept known as secondary sarcopenia [9]. A key driver of this process across many chronic diseases is systemic inflammation [10]. This link is particularly evident in rheumatic diseases, where sarcopenia is now recognized as a prevalent but often underappreciated comorbidity in conditions like rheumatoid arthritis (RA), systemic lupus erythematosus, and juvenile idiopathic arthritis (JIA) itself [11, 12].
In adults with RA, for example, this has led to the concept of “rheumatoid sarcopenia,” where chronic inflammation, driven by pro-inflammatory cytokines like tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6), creates a catabolic state that promotes muscle protein breakdown and inhibits synthesis. This leads to a loss of both muscle mass and strength that often exceeds what is expected from aging alone [13].
This expanded understanding is particularly relevant for JIA, the most common chronic rheumatic disease in children. The pediatric years are a critical window for musculoskeletal development, and the presence of chronic inflammation can have profound and lasting consequences. This leads to a triad of interconnected problems: growth retardation, impaired bone health, and significant muscle deficits, reflecting both impaired muscle development and active wasting [14, 15].
Within this triad, this complex muscle pathology is now increasingly recognized not merely as a separate comorbidity but as an integral part of the disease’s functional impact. It arises directly from both local joint pathology—leading to pain, disuse, and arthrogenic muscle inhibition—and the systemic catabolic effects of inflammation, ultimately compromising the patient’s physical capabilities [16]. More recently, this failure to accrue and maintain normal muscle mass has also been recognized as a key indicator of malnutrition in this population, driven by a multifactorial process [17].
While systemic inflammation, pain-related immobility, reduced physical activity, and the catabolic effects of medications are well-recognized contributors, recent evidence from RA suggests that other mechanisms, such as oxidative stress and hormonal shifts, also play a significant role in what is termed illness-related muscle loss (secondary sarcopenia) [18]. This complex interplay of factors not only impacts immediate physical function but also disrupts the crucial “muscle-bone unit,” which is fundamental for both normal linear growth and optimal bone accretion and bone mass, ultimately increasing the risk of short stature and fractures later in life.
Despite a clear rationale for muscle deficits in JIA, muscle mass and strength in this population have never been systematically reviewed. A previous systematic review by Houttu et al. [19] focused on body composition more broadly and reported inconsistent findings, limited by disease heterogeneity, treatment-related confounding, and the frequent use of body composition as a secondary outcome. Furthermore, while substantial progress has been made toward consensus diagnostic criteria for sarcopenia in older adults—particularly with strength-focused models such as EWGSOP2, which effectively predict adverse outcomes, and with recent initiatives like the Global Leadership Initiative on Sarcopenia (GLIS), which has achieved high agreement on core concepts [7, 8, 20] —a critical gap remains.
These criteria have not been validated or adapted for pediatric or young adult populations with childhood-onset conditions. This review therefore specifically examines muscle alterations in JIA and considers the implications of sarcopenia for long-term management.
Methods
This scoping review was conducted and reported in adherence to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) checklist, where applicable. Various recommendations and guidelines for producing a narrative biomedical review were taken into account, along with additional suggestions and notable points [21]. Approval from an institutional review board was not deemed necessary for this study.
Search strategy
A comprehensive electronic search was conducted across PubMed/MEDLINE, EMBASE, and the Cochrane Library databases to identify relevant studies published from database inception to September 2025. The search strategy combined Medical Subject Headings (MeSH) and free-text terms for JIA with terms related to muscle mass (lean mass, body composition) and muscle strength (strength, dynamometry, torque). The PubMed search strategy was as follows: (“juvenile idiopathic arthritis” [MeSH Terms] OR “juvenile idiopathic arthritis” [All Fields] OR “juvenile rheumatoid arthritis” [All Fields]) AND (“sarcopenia” [MeSH Terms] OR “sarcopenia” [All Fields] OR “muscle mass” [All Fields] OR “lean mass” [All Fields] OR “body composition” [All Fields] OR “muscle strength” [All Fields] OR “dynamometry” [All Fields] OR “torque” [All Fields]).
Searches were limited to English-language publications. The titles and abstracts of all records were screened for relevance, and full-text articles were reviewed when eligibility was uncertain. Duplicate records were removed prior to screening.
A flowchart of the study selection process is presented in Fig. 2.
Fig. 2PRISMA flow diagram of the study selection process
Inclusion and exclusion criteria
To be included in this review, studies had to meet the following criteria: (i) original peer-reviewed research; (ii) included patients with JIA; (iii) reported a quantitative assessment of either muscle/lean mass (by dual-energy X-ray absorptiometry (DXA), peripheral quantitative computed tomography (pQCT), or bioelectrical impedance analysis (BIA)) or muscle strength (using dynamometry); (iv) written in the English language. Studies measuring muscle parameters only by ultrasound or skinfold thickness, as well as editorials and viewpoint papers, were excluded.
Data extraction
Two independent reviewers (M.K. and O.Y.) performed the literature search, screening, and data extraction. For each eligible study, data were collected on study design, participant characteristics, assessment methods, and key quantitative outcomes. Any discrepancies between reviewers were resolved through discussion, with a third reviewer (MD) adjudicating any unresolved disagreements.
Given the heterogeneity of study designs, measurement techniques, and reported outcomes, a narrative synthesis was conducted rather than a meta-analysis.
Results
The initial literature search identified 195 articles. After a systematic screening process involving the removal of duplicates and irrelevant studies, a final review of 37 articles was conducted, encompassing a total of 2,040 patients with JIA (Fig. 2).
To reflect the evolution of research in this field and align with the core objectives of this review, the included studies were categorized into three groups for analysis: 21 studies assessing muscle mass only (Table 1), 10 studies assessing muscle strength only (Table 2), and 6 comprehensive studies assessing multiple components of sarcopenia (Table 3). Where multiple publications reported outcomes from the same cohort, they were grouped and counted as one study [22, 23]. It is noteworthy that while a majority of the included studies (26 of 37 studies) were specifically designed to investigate muscle parameters as a primary outcome (in body composition context), a substantial subset (11 of 37 studies) provided evidence on muscle mass or strength as a secondary endpoint.
This variability in study design may contribute to the heterogeneity of the reported findings.
Table 1Characteristics and key findings of included studies assessing muscle mass onlyN.Author(s) & YearStudy DesignParticipantsAssessment Method(s)Key Quantitative FindingsMain Conclusion1.Felin et al. (2007) [24]Cross-sectional48 JIA patients (17 pauci-, 23 poly-, 8 systemic) vs. 266 controlspQCT, DXAmCSA Z-scores (SD from control): By JIA subtype: Polyarticular: -0.87; Pauciarticular: -0.74; Systemic: -0.33.
By treatment exposure: GCs + MTX: -1.08; MTX only: -0.57; No treatment: -0.35. Predictive Role of Muscle: Total lean mass was the strongest positive predictor of bone geometry and strength (p < 0.001), explaining 33–96% of the variance in bone parameters.Prolonged GC exposure is a significant risk factor for muscle atrophy, and this loss of muscle mass is the primary determinant of compromised bone geometry and strength in JIA.2.Henderson et al. (1997) [25]Cross-sectional48 prepubertal, GC-naïve JIA patients vs.
25 controlsDXAJIA patients with low BMD (29% of cohort) had significantly lower total lean mass compared to JIA patients with normal BMD (7,400 gm vs. 9,900 gm, p = 0.001).Low lean mass is linked to higher disease activity and low BMD even in GC-naïve patients.3.Henderson et al. (2000) [26]Cross-sectional36 GC-naïve postpubertal female JIA patients vs.
51 controlsDXAOverall Comparison: JIA patients had a trend towards lower lean mass vs. controls (39.8 kg vs. 42.2 kg, p = 0.07). Subgroup analysis: JIA patients with low bone mass had significantly lower lean mass than JIA patients with normal bone mass (33.7 kg vs.
42.5 kg, p < 0.001). Predictive Role: Total lean mass was the sole significant predictor of total-body bone mass, accounting for 76.3% of its variance (p = 0.01).Lean mass is the most powerful predictor of bone status in postpubertal, non-GC-treated female JIA patients.4.Bechtold et al. (2005) [27]Controlled study17 JIA on GH vs.
17 untreated controlspQCTmCSA (mm²): GH therapy group: 2539 ± 89; Control group: 1692 ± 596 (p < 0.01). Body Composition: relative fat mass was significantly lower in the GH group vs. controls (20.8% vs. 30.1%, p < 0.05).GH treatment is associated with significantly greater muscle mass (1.5 times higher) and lower fat mass compared to no treatment, indicating a strong anabolic effect on body composition.5.Bechtold et al.
(2010) [28]Longitudinal12 prepubertal JIA on GH (no control group)pQCTOver a mean of 5.3 years, mCSA Z-scores significantly increased. Height-corrected mCSA Z-score improved from − 0.98 at baseline to + 0.16 at final height (p = 0.046).Long-term GH treatment leads to a significant increase and normalization of muscle mass in JIA patients.6.Simon et al. (2003) [29]3-year longitudinal intervention13 JIA patients on GC and GHDXAOver 3 years, median lean mass increased from 19.8 kg to 26.0 kg (+ 31%, p < 0.05), while fat mass remained stable (7.5 kg → 7.8 kg, median Fat %: 29% → 19% (p = 0.01).GH treatment can significantly improve body composition by increasing lean mass in JIA patients on long-term steroid therapy.7.Lien et al.
(2005) [30]2-year prospective longitudinal108 children with early JIA vs. controlsDXAOver 2 years, the JIA group had lower gains in lean mass (a difference of 1.2 kg, 95% CI 0.5–1.8; p < 0.001) and higher gains in fat (a difference of 2.8%, 95% CI 1.2–4.4; p = 0.001) compared to controls.JIA impairs normal muscle accrual and promotes fat gain during growth, leading to an adverse shift in body composition over time.8.Burnham et al. (2006) [31]12-month longitudinal42 JIA patients vs. 192 controlspQCTBaseline: mCSA lower by -6.4% (p = 0.01) & bone strength lower by -14.4% (p = 0.001).
Over 12 months, muscle accrual was 2.0% greater in JIA (p = 0.02), but bone strength did not improve.First longitudinal evidence of a dysfunctional muscle-bone unit in JIA.9.Burnham et al. (2008) [32]Cross-sectional101 JIA patients vs. 830 controlspQCTmCSA Z-scores by subtype: Lower in Polyarticular JIA (-0.56) & SpA (-0.50).
Normal in Oligoarticular JIA (0.06). Higher in systemic JIA (+ 0.58). Muscle-Bone Unit: Bone strength Z-score was significantly lower than mCSA Z-score (p < 0.001) after adjustments.Muscle deficits are subtype-dependent and linked to disability.
The muscle-bone unit is dysfunctional in JIA, with bone strength being inappropriately low for the existing muscle mass.10.Stagi et al. (2014) [33]Cross-sectional & Longitudinal138 JIA patients vs. 43 jSLE patients vs.
79 controlsDXA, pQCTmCSA (mm²): JIA: 2126.9; JSLE: 2440.9; Controls: 2300.6. JIA significantly lower than both JSLE (p < 0.0005) and controls (p < 0.001). Fat CSA (mm²): JIA: 1161.8; JSLE: 1222.3; Controls: 822.0.
Both JIA & JSLE significantly higher than controls (p < 0.0001).Significant muscle loss appears to be a specific feature of JIA, more pronounced than in JSLE, while increased adiposity is common to both conditions.11.Markula-Patjas K. (2015) [34]Cross-sectional50 JIA patients (with GC exposure) vs. 88 controlsDXANo significant difference in lean mass (JIA: 31.7 kg vs.
Controls: 30.6 kg; p = 0.852). Significantly lower BMC-to-lean mass ratio in JIA (p = 0.002).Normal body composition can be observed in JIA cohorts with low disease activity, despite a history of GC use.12.Caetano et al. (2012) [35]Cross-sectional controlled42 female JIA patients vs.
35 controlsDXAHigher adiposity in JIA girls (p < 0.001). No significant difference in lean mass index (p = 0.212).JIA in girls is associated with increased fat, but not necessarily lower indexed lean mass.13.Roth et al. (2007) [36]4-year longitudinal25 JIA patients (13 oligoarticular, 12 polyarticular)pQCTOligoarticular group: Showed significant improvement in mCSA Z-score (from − 1.58 to -0.55, p < 0.05).
Polyarticular group: persistent deficits with no significant improvement in mCSA Z-score (from − 2.26 to -2.03, NS).The potential for musculoskeletal recovery is subtype-dependent, with notable improvement in muscle mass in the oligoarticular cohort, while deficits persist in the polyarticular group.14.Stagi et al. (2014) [37]Cross-sectional & Longitudinal245 JIA patients vs. 80 controlsDXA, pQCT, QUSAt baseline, JIA patients had significantly lower mCSA (2112.7 vs.
2398.5 mm², p < 0.005) and higher fat CSA (1143.1 vs. 918.3 mm², p < 0.0001) vs. controls. Deficits were most prominent in polyarticular and systemic JIA.
Longitudinally, these deficits persisted.JIA patients have significant and persistent deficits in muscle mass and increased fat mass, which do not fully normalize over time despite therapy, indicating a mixed defect of bone accrual and lower muscle forces.15.Haugen et al. (2002) [38]Cross-sectional220 adult JIA patients vs. 95 controlsDXATotal Lean Mass (kg): Women in Remission (n = 84): JIA: 45.1 ± 6.1; Controls: 42.1 ± 5.0 (p < 0.05, JIA higher).
Women with Active Disease (n = 54): JIA: 42.7 ± 5.5; Controls: 42.1 ± 5.0 (p > 0.05, no difference); Men (all patients, n = 82): No significant difference vs. controls.Young adult women with JIA in remission may have a compensatory higher lean body mass compared to healthy peers. Men with a history of JIA show no significant long-term alterations in body composition.16.Diaz-Cordovés Rego et al. (2021) [39]Cross-sectional case-control80 JIA patients vs.
80 controlsDXANo significant overall difference in lean mass (High activity: 28.5 kg vs. Low activity: 24.8 kg vs. Controls: 25.0 kg; p = 0.396).
Patients with high disease activity had significantly higher fat mass (High activity: 18.1 kg vs. Low activity: 9.7 kg vs. Controls: 12.2 kg; p = 0.003)High inflammatory activity, not JIA itself, is the primary driver of increased adiposity.17.Billiau et al.
(2010) [40]Prospective comparative16 MTX-resistant JIA (on ETN + MTX) vs. 8 new-onset JIA (on MTX)DXAA significant increase in the lean: fat mass ratio was observed in the ETN + MTX group over 18 months (e.g., from median 2.7 to 3.2 at 12 months, p = 0.04). No significant change was observed in the MTX-only group.Biologic therapy is more effective than MTX alone in improving body composition, suggesting a deeper level of inflammatory control is needed to reverse adverse metabolic changes.18.Vasil et al.
(2024) [41]Cross-sectional case-control21 adolescents with JIA vs. 21 controlsDXANo significant group difference in lean mass. Lean mass was a positive predictor of aBMD (β = 0.05, p = 0.03) and BMC (β = 0.06, p < 0.001).Lean mass remains a key predictor of bone health even when total mass is not reduced.19.El Badri et al.
(2014) [22] & Hari et al. (2015) [23]Cross-sectional (same cohort)33 children with JIADXA, Dietary RecallEl Badri: Positive correlation between lean mass and total body BMD (r = 0.41, p = 0.04).Hari: Positive correlation between carbohydrate intake and lean mass (r = 0.4, p = 0.03).Lean mass is a key determinant of bone health in this cohort and is associated with carbohydrate intake.20.Räsänen et al. (2024) [42]Cross-sectional case-control79 JIA patients vs.
79 controlsDXA, BIANo significant difference in lean mass (JIA: 28.5 kg vs. Controls: 29.3 kg; p = 0.081) or fat mass (JIA: 12.2 kg vs. Controls: 10.7 kg; p = 0.077).
Increased adiposity was significant in patients with inactive disease.Increased adiposity is a feature of JIA, persisting even in inactive disease.21.Więch et al. (2018) [43]Cross-sectional case-control46 JIA patients vs. 46 controlsBIAJIA vs.
Controls: Muscle Mass %: 46.0% vs. 49.5% (p = 0.005). Body Cell Mass %: 50.6% vs.
52.7% (p = 0.010). Phase Angle: 5.45 vs. 5.85 (p = 0.010).
Subtype Analysis: Polyarthritis group had significantly lower muscle mass % vs. controls. Oligoarthritis group showed no significant difference vs. controls.Children with JIA are at high risk of malnutrition, as evidenced by lower muscle and body cell mass. This risk is predominantly concentrated in the polyarticular subtypeAbbreviations: aBMD, areal bone mineral density; BIA, bioelectrical impedance analysis; BMC, bone mineral content; BMD, bone mineral density; CI, confidence interval; mCSA, muscle cross-sectional area; DXA, dual-energy X-ray absorptiometry; ETN, etanercept; GC, glucocorticoid; GH, growth hormone; JIA, juvenile idiopathic arthritis; jSLE, juvenile systemic lupus erythematosus; LBM, lean body mass; LMI, lean mass index; MTX, methotrexate; NS, not significant; pauci-, pauciarticular; poly-, polyarticular; pQCT, peripheral quantitative computed tomography; QUS, quantitative ultrasonography; SD, standard deviation; SpA, spondyloarthritis
Table 2Characteristics and key findings of included studies assessing muscle strength onlyNo.Author(s) & Year [Ref]Study DesignParticipantsAssessment Method(s)Key Quantitative FindingsMain Conclusion1.Saarinen et al. (2008) [44]Cross-sectional case-control25 JIA patients vs. 25 controlsDynamometryJIA vs.
Controls: Knee Extension 80: Right: 215.9 vs. 259.6 (p < 0.05).Left: 205.7 vs. 245.1 (p < 0.05).Knee Extension 30°: Right: 128.0 vs.
136.7. Left: 120.0 vs. 134.4.
Knee Flexion: Right: 117.0 vs. 122.4.Left: 113.0 vs. 114.1.
Ankle Plantarflexion: Right: 298.7 vs. 294.8. Left: 240.4 vs.
280.5. Ankle Dorsiflexion: Right: 131.0 vs. 136.3.
Left: 125.9 vs. 134.3.Muscle weakness in JIA is present but localized, with a statistically significant deficit only in knee extension strength. Strength in other muscle groups was comparable to controls in this cohort with inactive disease.2.Broström et al.
(2004) [45]Cross-sectional case-control10 female JIA patients vs. 10 controlsIsokinetic DynamometrySignificantly lower isometric strength in JIA: Plantarflexion: 42.7 ± 22.7 Nm vs. 82.1 ± 25.8 Nm (p = 0.004).
Dorsiflexion: -10.9 ± 4.8 Nm vs. -17.7 ± 4.9 Nm (p = 0.010)Girls with JIA have a significant (40–50%) deficit in ankle plantar- and dorsiflexor strength compared to healthy peers.3.Baydogan et al. (2015) [46]Randomized Controlled Trial30 JIA patients (15 per group)Handheld DynamometerKnee Extension Strength (change from baseline): Balance Group: +10.1 kg. Strengthening Group: +5.9 kg (p < 0.01 between groups).
Hip Flexion Strength (change from baseline): Balance Group: +8.1 kg. Strengthening Group: +4.9 kg (p < 0.01 between groups)Both strengthening and balance-proprioceptive exercises improve muscle strength, but the latter may be more effective for improving overall lower extremity function.4.Rashed et al. (2018) [47]Cross-sectional case-control23 JIA patients vs.
46 controlsHandheld DynamometerSignificantly lower HGS in JIA: Dominant Hand: 16.1 ± 7.1 kg vs. 22.8 ± 8.1 kg (p < 0.001). Non-dominant Hand: 14.7 ± 6.7 kg vs.
20.9 ± 7.9 kg (p < 0.001). Correlations: HGS vs. JADAS-27: r = -0.467 (p = 0.025)Hand grip strength is significantly reduced in JIA and serves as an independent predictor of disease activity, disability, and quality of life.5.Sandstedt et al.
(2013) [48]Randomized Controlled Trial54 JIA patients (randomized to exercise or control)Handheld DynamometerBaseline: >50% of patients had low grip strength. Hip extensor and abductor strength was below normative values. Post-intervention (Exercise Group): Hip Extensor Strength (Right): 47.5 to 54.3 Nm (p < 0.03).
Knee Extensor Strength (Right): 57.3 to 65.8 Nm (p < 0.03). Knee Extensor Strength (Left): 55.8 to 61.7 Nm (p < 0.03).Grip Strength (Right Hand, Nm): Baseline: 157.9 (Exercise Group) vs. 197.1 (Control Group).
Lower Limb Strength (Nm): Significant increase in hip and knee extensor strength post-intervention in exercise group (e.g., Right Knee Ext: 57.3 to 65.8, p < 0.03).A home-based weight-bearing exercise program is well-tolerated and effective in significantly improving lower extremity muscle strength (hip and knee extensors) in children with JIA.6.Houghton et al. (2018) [49]Pre-post intervention trial24 JIA patients (13 completers)Dynamometry, DXA, HR-pQCTNo significant improvement post-intervention: Grip Strength (Left, kg): 19.5 → 19.8.Grip Strength (Right, kg): 20.9 → 22.0.Knee Extensor Peak Torque (Nm): 182.0 → 219.7A home-based exercise program was safe but did not lead to significant gains in either muscle strength or bone outcomes, primarily due to low adherence (47%).7.Elnaggar et al. (2024) [50]Randomized Controlled Trial58 JIA patients (29 per group)Isokinetic DynamometryGreater improvement in Knee Extensor Strength in AcVR group: Right Leg (120°/s): 38.1 → 43.5 Nm (AcVR) vs.
37.8 → 40.0 Nm (Control); p = 0.032. Left Leg (120°/s): 35.9 → 39.6 Nm (AcVR) vs. 34.1 → 36.2 Nm (Control); p = 0.039A 6-week accommodating variable-resistance (AcVR) training is more effective than usual exercise for improving knee extensor strength and muscle architecture in adolescents with JIA.8.Van Oort et al.
(2013) [51]Pre-post intervention pilot study7 JIA patientsDynamometry, UltrasoundStrength (Nm): No significant change in leg extension (111.6 → 105.0) or elbow flexion (28.3 → 30.1).Muscle Thickness (cm): Significant increase in vastus lateralis thickness (3.0 → 3.2 cm, p = 0.017).A 6-week home-based resistance program can induce muscle hypertrophy (increased thickness) even without a significant increase in maximal strength.9.Eid et al. (2016) [52]Randomized Controlled Trial36 JIA patients (18 per group)Isokinetic DynamometryGreater improvement in Quadriceps Strength in Biofeedback group after 12 weeks: Right Leg: 39.1 → 50.2 Nm (Biofeedback) vs. 38.2 → 42.3 Nm (Control); p < 0.0001.
Left Leg: 35.6 → 47.3 Nm (Biofeedback) vs. 34.3 → 40.4 Nm (Control); p < 0.0001EMG biofeedback training is a useful adjunct to conventional physical therapy, leading to significantly greater improvements in quadriceps strength.10.Arman et al. (2019) [53]Randomized Clinical Trial62 JIA patients (31 per group)Handheld DynamometerGreater improvement in Upper Limb Strength in Video Games group: Shoulder Flexion (Right): +3.59 lb (Video) vs. +1.15 lb (Traditional); p = 0.001.
Elbow Flexion (Right): +3.67 lb (Video) vs. +1.60 lb (Traditional); p = 0.001Both traditional and video games-based training improve upper limb strength, but the video games-based approach is significantly more effective.Abbreviations: AcVR, accommodating variable-resistance; DXA, dual-energy X-ray absorptiometry; EMG, electromyographic; HGS, handgrip strength; HR-pQCT, high-resolution peripheral quantitative computed tomography; JADAS-27, Juvenile Arthritis Disease Activity Score-27; JIA, juvenile idiopathic arthritis
Table 3Characteristics and key findings of included studies assessing multiple components of sarcopenia (Muscle Mass, Strength, and/or Function)No.Author(s) & Year [Ref]Study DesignParticipantsAssessment Method(s)Key Quantitative FindingsMain Conclusion1.Roth et al. (2004) [54]Cross-sectional57 JIA patients (27 oligo-, 20 poly-, 10 systemic) vs. reference dataMass: pQCT (mCSA); Strength: Hand dynamometry (Grip strength)mCSA Z-scores, age-matched median: Significantly reduced in all subgroups. Polyarticular: -2.57; Oligoarticular: -1.72; Systemic: -1.39; Strength (Grip Strength): Not reported as absolute values or Z-scores.
Showed a very strong positive correlation with mCSA (r = 0.91, p < 0.001).A primary problem in JIA is the lack of muscle mass and force, which leads to secondary abnormalities in bone geometry (e.g., thinner cortices), rather than a primary defect in bone metabolism.2.Risum et al. (2019) [55]Cross-sectional case-control60 JIA patients vs. 60 controlsMass: DXAStrength:DynamometryNo significant difference in Total Lean Mass (JIA: 33.4 ± 8.5 kg vs.
Controls: 35.7 ± 10.0 kg).Strength: Significantly lower strength in JIA: Grip Strength (dominant): 23.2 ± 7.4 kg vs. 27.9 ± 8.4 kg (p < 0.001). Knee Extensor Strength (dominant, normalized): 124 ± 38 vs.
146 ± 34 (Nm/kg x 100) (p < 0.001)Muscle strength is significantly impaired in JIA even when total muscle mass is comparable to healthy controls, indicating a deficit in muscle quality/function.3.Sule & Fontaine (2019) [56]Randomized Controlled Trial33 patients with polyarticular JIAMass: DXAStrength:DynamometryBaseline: only 35% of patients met CDC guideline for lean mass (> 60% of total body mass). No significant change in body composition post-intervention.Low adherence to exercise programs is a major barrier to improving body composition in JIA.4.Brabnikova Maresova et al. (2014) [57]Cross-sectional31 adult JIA patients vs.
31 controlsMass: DXA. Strength/Function: Chair rise test, Walking speedMass: JIA patients had significantly lower lean mass (39.4 vs. 44.1 kg, p = 0.04) and higher fat mass (21.1 vs.
16.3 kg, p = 0.05). Function: JIA patients demonstrated reduced performance in the chair rise test and slower walking speed (specific values not reported).Adult patients with active JIA have significant deficits in both muscle mass and physical function compared to healthy controls.5.Ozdemir et al. (2024) [58]Cross-sectional case-control32 JIA patients vs.
18 controlsMass: BIA.Strength: Handheld DynamometerNo significant difference vs. controls: Lean Body Weight: 41.5 ± 11.3 kg vs. 39.5 ± 15.2 kg (p = 0.599). Knee Extensor Strength: 22.1–23.6 kg (JIA) vs.
25.7–27.3 kg (Controls) (p > 0.1). Correlations in JIA: Physical activity vs. Lean Mass: r = 0.423 (p = 0.016)In JIA patients with low disease activity, muscle mass and strength may be comparable to healthy peers.
Physical activity is positively correlated with muscle health.6.Kulyk & Dzhus (2024) [59]Cross-sectional70 young adult JIA patientsMass: DXA.Strength: Dynamometry.Function: Gait SpeedPrevalence: Low handgrip strength: 64%.Sarcopenia (low mass + low strength): 59%. Severe sarcopenia (+ low function): 34%. Risk Factors for Sarcopenia: DAS28-ESR: OR 2.08 (95% CI 1.15–3.76; p = 0.01).
JADI-A (articular damage): OR 2.29 (95% CI 1.23–4.25; p = 0.009)Sarcopenia is highly prevalent in young adults with JIA and is strongly associated with disease activity, accumulated damage, and poor functional capacity, establishing it as a critical, measurable comorbidity.Abbreviations: BIA, bioelectrical impedance analysis; CDC, Centers for Disease Control and Prevention; CI, confidence interval; mCSA, muscle cross-sectional area; DAS28-ESR, Disease Activity Score in 28 joints using erythrocyte sedimentation rate; DXA, dual-energy X-ray absorptiometry; JADI-A, Juvenile Arthritis Damage Index-articular; JIA, juvenile idiopathic arthritis; mCSA, muscle cross-sectional area; oligo-, oligoarticular; OR, odds ratio; poly-, polyarticular; pQCT, peripheral quantitative computed tomography
Findings on muscle mass
The 21 studies in this category reveal a heterogeneous but clear pattern. Among the 14 studies that directly compared JIA patients to healthy controls, a majority (9 studies, 64%) reported statistically significant lower lean mass or muscle cross-sectional area (mCSA) in JIA patients [24–26, 30–33, 37, 43]. In contrast, five studies (36%) found no significant difference, a finding primarily observed in cohorts with low disease activity [34, 35, 39, 41, 42].
The remaining studies included four intervention trials, all of which demonstrated positive effects of therapies like growth hormone or biologics on muscle mass [27–29, 40], and two unique cohorts (reported across three publications) with designs that highlighted subtype-specific trajectories and correlational findings [22, 23, 36].
Deficits in muscle mass were shown to be highly dependent on disease subtype. Studies utilizing pQCT and BIA demonstrated that significant deficits in mCSA and muscle mass percentage were most consistently observed in polyarticular JIA, whereas findings in systemic JIA were more variable across studies; muscle mass was often preserved in the oligoarticular subtype [32, 37, 43]. This subtype-dependent trajectory was confirmed longitudinally, with a 4-year pQCT follow-up showing notable improvement in mCSA in the oligoarticular cohort, while deficits persisted in the polyarticular group [36].
Overall, early JIA appears to disrupt normal musculoskeletal development from its onset, with prospective data showing that patients exhibit a blunted accrual of total lean mass (by DXA) compared to healthy peers [30].
In parallel with muscle deficits, increased adiposity was identified as a key feature of JIA. A pivotal study by Stagi et al. [33] highlighted that while significant muscle loss was a specific feature of JIA compared to juvenile systemic lupus erythematosus, increased fat mass was common in both conditions. This increase in adiposity was observed even in female patients who did not have significant deficits in lean mass index [35] and, importantly, was found to persist even in patients with inactive disease, suggesting that adverse metabolic changes may outlast clinical remission [42].
However, it is noteworthy that while many studies reported higher body fat percentage, the use of clinical obesity metrics, such as body mass index (BMI) or waist circumference, was rarely reported, limiting a comprehensive assessment of sarcopenic obesity phenotypes.
Disease activity emerged as a critical modulator of these changes. In cohorts with predominantly low disease activity, significant differences in total lean mass (by DXA) were often not observed [34, 39], suggesting that effective control of inflammation is key to preserving muscle mass. The influence of therapies further complicates this picture.
While prolonged glucocorticoid (GC) use was identified as a significant risk factor for muscle atrophy (mCSA by pQCT) [24], studies on GC-naïve cohorts still demonstrated a link between lower total lean mass (by DXA) and higher disease activity [25]. More advanced therapies showed greater promise; Billiau et al. [40] demonstrated that add-on etanercept, unlike methotrexate (MTX) monotherapy, led to a significant improvement in the lean-to-fat mass ratio (by DXA). Specific interventions like growth hormone (GH) therapy have also shown powerful anabolic effects, with both controlled and longitudinal studies documenting a significant increase and even normalization of muscle mass over time [27–29].
Long-term outcomes in adulthood appear to be sex-specific. A large study of young adults by Haugen et al. [38] using DXA revealed that women in clinical remission had a significantly higher total lean mass compared to healthy controls, suggesting a potential for compensatory recovery. In contrast, no significant long-term alterations in total lean mass were observed in male patients.
A strong link between muscle and bone health was identified across multiple studies. Lean mass, measured by DXA, was consistently found to be the single most powerful positive predictor of bone mineral content and density [22, 26, 41]. One study attributed over 76% of the variance in bone mass to lean mass alone [26].
However, this “functional muscle-bone unit” appears to be uniquely compromised in JIA. Burnham et al. [31], using pQCT, provided the first direct evidence of this dysfunction, showing that an observed increase in mCSA in JIA patients over a year did not translate into the expected gains in bone strength. Other factors may also play a role, as suggested by a study that found a positive correlation between carbohydrate intake and total lean mass [23].
Findings on muscle strength
The 10 studies in this category provide consistent evidence of muscle weakness. All three cross-sectional studies that directly compared JIA patients with healthy controls reported significantly lower muscle strength [44, 45, 47]. The remaining seven studies were interventional trials exploring the efficacy of rehabilitation.
Of these, a majority (five studies, 71%) demonstrated that targeted exercise programs, particularly those incorporating neuromuscular control or technology, could lead to statistically significant improvements in muscle strength [46, 48, 50, 52, 53]. In contrast, two studies (29%) found no significant improvement, an outcome attributed to low patient adherence [49, 51].
Early cross-sectional assessments revealed that weakness is not always global but can be localized to specific muscle groups. For instance, Saarinen et al. [44] found that in a cohort with inactive disease, a significant strength deficit was present only in the knee extensors. Similarly, Broström et al. [45] documented a profound, isolated strength deficit in the ankle plantarflexors and dorsiflexors of prepubertal girls with polyarticular JIA.
The clinical relevance of these deficits was highlighted by Rashed et al. [47], who showed that handgrip strength was not only significantly lower in JIA patients but also served as a robust proxy for overall disease burden, correlating strongly with disease activity, disability, and quality of life.
Given these deficits, numerous interventional studies have explored the potential for recovery through exercise, revealing both the efficacy of rehabilitation and its practical challenges. Home-based and supervised exercise programs were shown to be effective in significantly improving lower extremity muscle strength, particularly in the hip and knee extensors [48]. However, the challenge of long-term implementation was underscored by Houghton et al. [49], who attributed a lack of sustained improvement in muscle strength to very low patient adherence.
Interestingly, a pilot study by Van Oort et al. [51] suggested that even short-term resistance training, while not producing significant gains in maximal strength, can induce structural changes such as muscle hypertrophy, indicating an early physiological response to exercise.
To overcome the limitations of traditional exercise and low adherence, more advanced and engaging therapeutic strategies have been investigated. Studies have shown that interventions focusing on enhancing neuromuscular control, such as balance-proprioceptive exercises [46] or electromyographic (EMG) biofeedback-guided training [52], can yield superior strength gains compared to standard strengthening alone. Furthermore, modern, technology-based approaches have proven highly effective.
Elnaggar et al. [50] demonstrated that instrumented variable-resistance training was more effective than usual exercise for improving knee extensor torque and muscle architecture. Similarly, a trial by Arman et al. [53] found that a rehabilitation program based on video games led to statistically superior gains in upper limb muscle strength compared to a traditional task-oriented program, suggesting that technology can be a more effective and motivating tool for muscle rehabilitation in JIA.
Comprehensive studies on both muscle mass and strength
The six studies in this category assessed both muscle mass and strength. Among the five cross-sectional studies, a clear majority (four studies, 80%) identified significant deficits in either muscle mass, muscle strength, or both [54, 55, 57, 59]. The single study that found no significant deficits was conducted in a cohort with low disease activity [58].
The one interventional trial in this group also reported no significant change in body composition, an outcome attributed to low patient adherence [56].
Several of these studies identified a critical dissociation between muscle mass and strength. Risum et al. [55] found that JIA patients had significantly lower muscle strength in both handgrip and knee extension/flexion despite having comparable total lean mass to healthy controls. This supported earlier findings by Roth et al. [54], who, using pQCT and dynamometry, demonstrated a strong positive correlation between reduced mCSA and lower grip strength, and linked these combined deficits to secondary abnormalities in bone geometry.
The relationship between muscle mass and strength appeared to be modulated by disease activity. In a cohort with low disease activity, Ozdemir et al. [58] found no significant differences in either lean body mass or knee extension muscle strength compared to controls but did report a positive correlation between physical activity and both muscle parameters. In contrast, a study of adults with active JIA by Brabnikova et al. [57] identified significant deficits in both total lean mass and functional performance (chair rise test and walking speed).
The difficulty of improving these deficits through intervention was also reported. A randomized controlled trial by Sule & Fontaine [56] found that a 12-week resistance exercise program did not lead to statistically significant changes in body composition, an outcome the authors attributed to very low patient adherence.
The application of the full, modern EWGSOP2 diagnostic algorithm allowed for the direct quantification of sarcopenia prevalence. In a study of 70 young adults, Kulyk & Dzhus [59] combined assessments of muscle mass (by DXA), strength (by dynamometry), and physical performance (by gait speed). The study reported a high prevalence of both sarcopenia and severe sarcopenia and identified associations between the sarcopenic phenotype and high disease activity, accumulated damage, and poor functional capacity.
Discussion
This scoping review of 37 studies shows a shift from isolated observations of musculoskeletal deficits in JIA to recognizing sarcopenia as a key clinical syndrome. Evidence of “mass–strength dissociation” indicates that marked weakness can occur despite preserved muscle mass, reflecting deficits in both muscle quantity and quality [55]. Using modern diagnostic criteria, in one study of young adults, sarcopenia was reported in 59% of patients and was strongly linked to higher disease activity and accumulated damage [59].
A primary finding of this review is that JIA promotes a shift in body composition towards reduced total lean mass and increased fat mass [33, 57]. This profile, which resembles what is sometimes termed «sarcopenic obesity», is likely driven by the catabolic state induced by chronic inflammation. It is important to note, however, that a formal diagnosis of sarcopenic obesity requires specific criteria, such as the combination of low muscle function and high adiposity (e.g., BMI ≥ 30 kg/m²), as outlined in the recent The European Society for Clinical Nutrition and Metabolism (ESPEN) and the European Association for the Study of Obesity (EASO) consensus statement [60].
Pro-inflammatory cytokines such as TNF-α and IL-6, which are central to JIA pathogenesis, are known to stimulate muscle protein breakdown and induce anorexia, contributing to both muscle wasting and malnutrition. This phenomenon is not unique to JIA and mirrors findings in other chronic inflammatory conditions, such as RA and inflammatory bowel disease, where systemic inflammation is a key driver of secondary sarcopenia. The consistent observation of increased fat mass, even when lean mass index remains unchanged in some studies [34, 35] suggests that JIA impacts overall metabolic health, not just muscle quantity.
Perhaps the most critical clinical implication highlighted by this review is the profound impact of muscle deficits on bone health. Our findings consistently show that total lean mass is the single most important predictor of bone mineral content (BMC) and density in JIA [26, 41, 57], with one study attributing over 76% of the variance in BMC to lean mass [26]. This strongly supports the “muscle-bone unit” theory, where bone adapts its structure and strength in response to mechanical loads generated by muscle contractions.
However, in JIA, this unit appears to be uniquely compromised. A key finding from study by Burnham et al. [31] was that even when muscle mass increased in JIA patients, it failed to produce the expected gains in bone strength, indicating a clear dissociation in the muscle-bone relationship. This suggests that chronic inflammation during the critical period of growth may uncouple the normal physiological link between muscle and bone.
The combination of reduced muscle mass, inflammation-driven inactivity, and pain leads to decreased mechanical loading on the skeleton, resulting in impaired bone geometry and lower bone density, as evidenced by multiple studies in our review [24, 25, 54]. This contrasts with findings in adults with established RA, where the functional muscle-bone unit appears to be preserved. This finding of a maintained muscle-bone relationship in RA has been demonstrated at both weight-bearing (tibia) and non-weight-bearing (metacarpal) sites, suggesting that in RA, bone deficits are largely an appropriate adaptation to the primary loss of skeletal muscle mass [61, 62].
This distinction highlights a potentially unique pathophysiological aspect of JIA, where the timing of the inflammatory insult during skeletal development may be critical, a conclusion that is strongly supported by recent critical overviews on bone health in pediatric rheumatology [63].
The review revealed considerable heterogeneity in musculoskeletal outcomes, suggesting that the impact of JIA is not uniform. Disease activity and subtype appear to be major determinants, with several studies identifying polyarticular and systemic JIA as subtypes with more pronounced muscle deficits compared to oligoarticular JIA [32, 33, 43]. This finding is consistent across different assessment methods, including pQCT and BIA, with the latter also showing significant muscle mass reduction only in the polyarthritis group [30, 43].
This aligns with the higher inflammatory burden in these subtypes. Furthermore, our findings suggest that achieving clinical remission can lead to a recovery of total lean mass, with one study showing that adult women in remission had even higher total lean mass than controls [38]. However, the trajectory of recovery may be complex.
The study by Räsänen et al. [42], for instance, found that while mean lean mass was not significantly lower in their JIA cohort with generally low disease activity, these patients still exhibited other signs of altered body composition, such as increased adiposity. This aligns with findings from Ozdemir et al. [58], who also found no significant differences in either lean mass or muscle strength in a cohort with low disease activity, further emphasizing that the most severe deficits are strongly linked to uncontrolled inflammation. This suggests that even in well-controlled disease, the underlying processes affecting musculoskeletal health may not be fully normalized, and a lack of significant difference in total lean mass does not necessarily equate to optimal muscle health.
The influence of treatment on muscle health is complex and multifactorial. While GC use is linked to muscle atrophy [24, 57], effective control of inflammation remains paramount, as adverse body composition changes are most prominent in patients with high disease activity [39]. The choice of antirheumatic therapy is also critical; advanced treatments like etanercept have shown a superior ability to restore a healthy lean-to-fat mass ratio compared to MTX alone, suggesting a deeper level of inflammatory control is needed [40].
Beyond this, specific interventions like GH-therapy have shown significant promise in increasing muscle mass [27–29]. However, translating these physiological insights into effective rehabilitation presents significant challenges. While targeted exercise can, in principle, reverse muscle weakness [48], its real-world efficacy is often severely limited by low patient adherence [49, 56].
Furthermore, the localized nature of weakness, with specific deficits documented in key muscle groups like knee extensors and ankle muscles even in patients with inactive disease [44, 45], suggests that rehabilitation programs must be highly specific and tailored, rather than general. These challenges may be unique to a pediatric-onset disease [64]. The “mass-strength dissociation” and the localized nature of weakness, for instance, may be partly explained by impaired neuromuscular adaptations.
In healthy individuals, initial strength gains from resistance training are primarily driven by neural factors, not muscle hypertrophy. A child with early-onset JIA, who may have developed poor physical literacy due to pain and activity limitation during critical developmental windows, might struggle to perform exercises effectively, thereby limiting these crucial neuromuscular adaptations. The concept of physical literacy—encompassing the motivation, confidence, and physical competence to engage in physical activities for life—is increasingly recognized as fundamental for long-term muscle health [65, 66].
This crucial finding suggests the primary barrier to improving muscle health in JIA may be not only behavioral and motivational but also developmental. In response, recent research has demonstrated that interventions incorporating modern technology and principles of neuromuscular control—such as video game-based rehabilitation [53] or EMG biofeedback [52]—are more effective and engaging than traditional approaches, representing a promising direction for future clinical practice. Finally, nutrition also plays a role.
One study found a positive correlation between carbohydrate consumption and total lean mass [23], highlighting that dietary patterns, in addition to inflammation and medication, are an important and potentially modifiable factor influencing muscle health in JIA.
Several limitations should be acknowledged. Our review highlights a significant gap in comprehensive sarcopenia assessment, as only a small subset of the included studies (6 of 37, 16%) evaluated multiple components of sarcopenia. Furthermore, the included studies themselves were often limited by small sample sizes, and power calculations were rarely reported, potentially affecting the reliability of findings showing no significant differences.
A major barrier to the formal recognition of sarcopenia in JIA is the lack of standardized assessment methods and validated, age-appropriate diagnostic criteria. Heterogeneity in imaging techniques and the absence of normative data for patients with childhood-onset disease hinder consistent diagnosis and cross-study comparisons [59, 67]. This is also true for the assessment of obesity; the variability in obesity metrics and the infrequent use of standardized cut-offs across studies hinder any precise estimation of the prevalence of sarcopenic obesity.
These gaps define priorities for future research: longitudinal cohorts tracking musculoskeletal development from childhood into adulthood; development and validation of JIA-specific consensus criteria for sarcopenia; and intervention studies that address not only the efficacy of exercise [68] but also the consistently reported low adherence and its behavioral and psychological determinants [49, 56]. Despite these limitations, our findings have clear clinical implications: clinicians should recognize the high risk of musculoskeletal deficits beyond the joints in individuals with a history of JIA and consider routine assessment of both muscle mass and strength, particularly in those with high disease activity or prolonged GC exposure.