From Hands to Imaging: The Emerging Role of Musculoskeletal Ultrasonography in Juvenile Idiopathic Arthritis Disease Activity Assessment

Dear Editor,

Disease activity assessment tools are widely and successfully used in clinical practice to evaluate disease activity and guide treatment accordingly. Juvenile idiopathic arthritis is no exception, and disease activity scores specific to JIA have been developed and validated in recent years [1, 2, 3]. Although minor differences exist between these scoring systems in terms of their components, all include clinical joint examination [1, 2, 3]. While the burden of JIA extends beyond articular involvement, arthritis remains the central feature of the disease. However, the assessment of arthritis may be more complex than initially appreciated.

Physical examination (PE) remains the cornerstone of joint assessment but has important limitations compared with advanced imaging modalities, such as contrast‐enhanced magnetic resonance imaging (MRI). Hemke et al. have shown that PE of the knee joint, which is the most commonly involved and reliably accessed joint, failed to identify MRI‐confirmed synovitis in 35.9% of patients [4]. Likewise, whole‐body MRI (WB‐MRI) has further highlighted the limitations of PE by revealing up to 49% subclinical joint inflammation in JIA patients [5]. However, despite its superior sensitivity, MRI remains impractical for routine disease assessment because of its high cost, limited availability, prolonged acquisition time, and the frequent need for sedation in young children. Accordingly, there remains a need for an imaging modality that combines the sensitivity of advanced imaging with the feasibility required for routine clinical assessment, a role that musculoskeletal ultrasonography (MSUS) is increasingly well positioned to fulfill.

The use of MSUS in the diagnosis and monitoring of JIA has grown considerably [6, 7]. The standardization of normal anatomical landmarks and pathological thresholds for pediatric ultrasonography by Outcome Measurement in Rheumatology (OMERACT) and Childhood Arthritis and Rheumatology Research Alliance (CARRA) has enabled more advanced research comparing MSUS with physical examination and MRI [8, 9, 10]. Although efforts toward standardization and validation are still ongoing, MSUS has shown superior sensitivity compared to PE; Magni‐Manzoni et al. demonstrated that MSUS captured 5.5% subclinical synovitis in JIA patients [11]. Furthermore, this finding has been confirmed by various studies, with subclinical synovitis rates varying by joint and being particularly higher at joints other than the knee, consistent with adult rheumatoid arthritis (RA) studies [12, 13, 14]. Although not as sensitive as MRI, Sande et al. compared MSUS with WB‐MRI and found a strong correlation between synovitis scores (Spearman's correlation coefficients = 0.74), compatible with the study by Vega‐Fernandez et al., in which a strong correlation was demonstrated between MSUS and contrast enhanced MRI scores [15, 16].

Consequently, investigators further explored the clinical implications of MSUS‐detected subclinical synovitis. Although initial reports by Magni‐Manzoni et al. and Zhao et al. were unsupportive, De Lucia et al. have shown that MSUS‐detected abnormalities in clinically inactive patients increased the risk of flare‐up by approximately 4 times, compatible with Fingerhutová et al.'s and adult RA studies [14, 17, 18, 19, 20]. Additionally, Miotto E Silva et al. have shown that Power Doppler positivity predicted joint damage along with risk of flare in clinically inactive patients [21]. Although MSUS‐detected subclinical synovitis may predict relapse, not all patients who exhibited subclinical synovitis relapsed, and it can be detected in up to 11.1% of healthy controls [22]. Notably, there is considerable heterogeneity across these studies regarding patient selection, treatment, follow‐up duration, and MSUS operators and technique. Therefore, further evidence and validation are needed before evidence‐based recommendations can be established for patients with MSUS‐detected subclinical synovitis.

Another advantage of MSUS over PE is its ability to grade arthritis. OMERACT has developed scoring systems for B‐mode and Color and Power Doppler imaging allowing clinicians to evaluate and grade effusion and inflammatory activity through visualization of synovial hypertrophy and vascularity [10]. The ability to grade inflammatory activity enhances disease activity assessment to capture changes in inflammatory activity more sensitively. In addition, MSUS is particularly valuable in detecting extra‐articular involvement, such as enthesitis, an important inflammatory manifestation in enthesitis‐related arthritis (ERA) and psoriatic arthritis [23]. Weiss et al. have demonstrated that physical examination and examination with a standardized dolorimeter have poor accuracy and reliability compared to MSUS in ERA patients [24]. However, with respect to structural damage, such as bone erosions, MSUS is only as sensitive as X‐ray, which is lower than MRI [25].

The Juvenile Arthritis Disease Activity Score (JADAS) and Wallace criteria primarily focus on active joint count, whereas the American College of Rheumatology (ACR) Pediatric criteria additionally incorporate the number of joints with limited range of motion (ROM), highlighting chronic structural damage [1, 2, 3]. Yet chronic articular sequelae in JIA may present with a spectrum of findings, including chronic anatomical changes and persistent tenderness along with restricted ROM, making it difficult to distinguish acute inflammatory findings from chronic sequelae [26]. This becomes particularly evident in patients with longstanding disease, residual joint sequelae, and recurrent flares, which is a common scenario in JIA. As a result, discrepancies arise between clinicians when disease activity scoring relies solely on physical examination, with Guzman et al. reporting only moderate interobserver agreement (κ = 0.47) even for joint swelling [27]. This reflects a fundamental shortcoming of clinical assessment: limited interobserver agreement and the difficulty of distinguishing chronic changes from active inflammation, an area where musculoskeletal ultrasonography (MSUS) can be remarkably advantageous, especially with Power Doppler imaging [6]. Supporting this concept, Fingerhutová et al. demonstrated that MSUS can resolve uncertainty in clinically equivocal joints by identifying active synovitis and that joints with MSUS‐confirmed synovitis carry a significantly higher risk of subsequent relapse than MSUS‐inactive joints [20].

These advantages prompted investigators to evaluate MSUS's correlation with standardized composite disease activity measures [16]. Across two studies, Vega‐Fernandez et al. demonstrated a moderate correlation of MSUS‐detected synovitis with cJADAS10 (rs = 0.45) and patient global assessment (rs = 0.44, p = 0.014), whereas the correlation with active joint count was weaker (rs = 0.37, p = 0.048) [13, 28]. These findings were further supported by Snipaitiene et al. even after disease remission has been achieved [29]. Consistent with this, Licciardi et al. reported that the burden of subclinical synovitis per patient correlated with JADAS‐27 (p = 0.03) [30]. Taken together, these findings suggest that MSUS reflects a different, and possibly more objective, domain of disease activity than traditional PE‐based measures. This may be particularly relevant in patients with clinically inactive disease or clinically equivocal arthritis. However, it should be noted that ultrasound‐guided treatment strategies did not yield clear superiority in the ARCTIC and TaSER trials in adult rheumatoid arthritis [31, 32]. While direct extrapolation of these studies to JIA is not feasible, further evidence is required before adopting purely MSUS‐guided disease activity scoring and treatment strategies in JIA.

The search for optimal methods to diagnose and monitor joint inflammation has been ongoing since the earliest use of X‐ray. Magnetic resonance imaging can address many unmet needs, and promising advances are being made in near‐infrared optical imaging, photoacoustic techniques, and thermal imaging; however, MRI has not displaced traditional physical examination due to limited accessibility and cost. Likewise, MSUS will not end this quest, as challenges regarding operator dependency, technique heterogeneity, and validation persist. Nevertheless, MSUS offers higher sensitivity than PE, the ability to grade arthritis, and direct visualization of active inflammation and extra‐articular structures, such as entheseal structures, along with substantial technical advantages, including wide availability, ease of use, absence of ionizing radiation, and a non‐invasive, sedation‐free nature (Table 1). Therefore, moving toward MSUS‐aided clinical assessment represents a logical and timely step in refining disease activity scoring in JIA and informing future research directions.

TABLE 1.

Comparison of imaging modalities used for the diagnosis and monitoring of juvenile idiopathic arthritis.

Feature	Conventional radiography (X‐ray)	MSUS	MRI
Synovitis	−	++/+++ a	+++
Joint effusion	−	++/+++ a	+++
Active vascular inflammation	−	++	+++
Bone marrow edema	−	−	+++
Cartilage	+	++	+++
Bone erosions	++	++	+++
Dynamic assessment	−	+++	−
Examination time	Short	Short	Long
Radiation	Yes	No	No
Cost	Low	Low—Moderate	High
Accessibility	+++	++	+
Bedside availability	++	+++	−
Repeatability	++	+++	++
Need for sedation	−	−	Sometimes
Guidance for joint injection	−	+++	−

Note: The ratings represent the overall utility of each modality in JIA, taking into account diagnostic performance, anatomical coverage, technical limitations, and feasibility, rather than accuracy alone. Scoring: −, not applicable or not useful; +, limited; ++, moderate; +++, excellent.

^a Performance may be limited in deep or poorly accessible joints.

Author Contributions

Conceptualization: B.K., Ç.Y., and S.A.B.; Writing – original draft: B.K. and Ç.Y.; Writing – review and editing: S.A.B.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.