You are here

  • Home
  • Archive
  • Volume 10, Issue 1
  • Validity of an ultrasonographic joint-specific scoring system in juvenile idiopathic arthritis: a cross-sectional study comparing ultrasound findings of synovitis with whole-body magnetic resonance imaging and clinical assessment

Article Text

Article menu

Download PDFPDF

Paediatric rheumatology
Original research
Validity of an ultrasonographic joint-specific scoring system in juvenile idiopathic arthritis: a cross-sectional study comparing ultrasound findings of synovitis with whole-body magnetic resonance imaging and clinical assessment
  1. Nina Krafft Sande1,2,
  2. Eva Kirkhus3,
  3. Vibke Lilleby1,
  4. Anders Høye Tomterstad3,
  5. Anna-Birgitte Aga1,
  6. Berit Flatø1,2 and
  7. Pernille Bøyesen1
  1. 1Department of Rheumatology, Oslo University Hospital, Oslo, Norway
  2. 2Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway
  3. 3Department of Radiology, Oslo University Hospital, Oslo, Norway
  1. Correspondence to Nina Krafft Sande; ninamkrafft{at}hotmail.com

Abstract

Objective To assess the validity of an ultrasonographic scoring system in juvenile idiopathic arthritis (JIA) by comparing ultrasound detected synovitis with whole-body MRI and clinical assessment of disease activity.

Methods In a cross-sectional study, 27 patients with active JIA underwent clinical 71-joints examination, non-contrast enhanced whole-body MRI and ultrasound evaluation of 28 joints (elbow, radiocarpal, midcarpal, metacarpophalangeal 2–3, proximal interphalangeal 2–3, hip, knee, tibiotalar, talonavicular, subtalar and metatarsophalangeal 2–3). One rheumatologist, blinded to clinical findings, performed ultrasound and scored synovitis (B-mode and power Doppler) findings using a semiquantitative joint-specific scoring system for synovitis in JIA. A radiologist scored effusion/synovial thickening on whole-body MRI using a scoring system for whole-body MRI in JIA. At patient level, associations between ultrasound synovitis sum scores, whole-body MRI effusion/synovial thickening sum scores, clinical arthritis sum scores, and the 71-joints Juvenile Arthritis Disease Activity Score (JADAS71) were calculated using Spearman’s correlation coefficients (rs). To explore associations at joint level, sensitivity and specificity were calculated for ultrasound using whole-body MRI or clinical joint examination as reference.

Results Ultrasound synovitis sum scores strongly correlated with whole-body MRI effusion/synovial thickening sum scores (rs=0.74,p<0.01) and the JADAS71 (rs=0.71,p<0.01), and moderately with clinical arthritis sum scores (rs=0.57,p<0.01). Sensitivity/specificity of ultrasound in detecting synovitis were 0.57/0.96 and 0.55/0.96 using whole-body MRI or clinical joint examination as reference, respectively.

Conclusion Our findings suggest that ultrasound is a valid instrument to detect synovitis, and that ultrasound synovitis sum scores can reflect disease activity and may be an outcome measure in JIA.

  • Arthritis, Juvenile
  • Inflammation
  • Magnetic Resonance Imaging
  • Synovitis
  • Ultrasonography

Data availability statement

Data are available upon reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/rmdopen-2023-003965

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Ultrasound is increasingly being used in the management of patients with juvenile idiopathic arthritis (JIA). However, the correlation between ultrasound findings and other imaging modalities and the validity of ultrasound detected synovitis in patients with JIA is not well known.

    WHAT THIS STUDY ADDS

    • Ultrasound synovitis sum scores strongly correlated with whole-body MRI effusion/synovial thickening sum scores and the 71-joints Juvenile Arthritis Disease Activity Score (JADAS71).

    • A novel ultrasonographic joint-specific scoring system with age-divided reference atlas may provide a valid assessment of synovitis in patients with JIA.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • Our findings suggest that ultrasound synovitis sum scores can reflect overall disease activity and may be an important outcome measure in clinical practice and research.

    Introduction

    Patients with juvenile idiopathic arthritis (JIA) demonstrate a wide range of disease manifestations and the clinical symptoms and signs can be variable. The number of joints with inflammation and the pattern of joint involvement are known to be of prognostic importance and guide the physician in treatment decisions.1 2 Detection of joints with inflammation is therefore important for the diagnosis, treatment and follow-up of these patients.3 Consequently, valid methods to detect and quantify joints with inflammation, and to assess disease activity in patients with JIA are needed.

    Traditionally, clinical joint examination with the evaluation of joint swelling, tenderness and limited range of motion in 71 joints is the standard disease activity assessment in patients with JIA, but can be challenging in children due to vague symptoms and complex anatomic regions.4–6 Contrast enhanced MRI is considered to be the gold standard in detecting synovitis when imaging a target joint,7 8 but a limitation is that only one joint or a limited number of joints can be evaluated in each session. Ultrasound is a readily available imaging modality that can assess many joints, is well accepted by children, and has demonstrated to be more sensitive in the evaluation of joint inflammation than clinical examination.9

    The process of standardising the use of ultrasound in children has evolved in recent years.10–15 Standardised ultrasound scanning approaches and scoring systems have shown to provide reliable findings of synovitis.13–15 However, it is less known whether ultrasound detected synovitis in children correlates with other imaging modalities and validated measures of disease activity.

    Few studies validating ultrasound findings of synovitis in children are available, and the studies conducted are difficult to compare as they vary in ultrasound definitions of synovitis, number of joints assessed, scoring systems, and type of comparator used.5 6 16–23 Most previous studies evaluating the validity of ultrasound have assessed single joints with ultrasound using clinical examination or contrast enhanced MRI as comparators.5 16 17 20 23 Other studies have examined multiple joints with ultrasound using clinical examination as a comparator.18 21 22 To the best of our knowledge, no studies have examined multiple joints with ultrasound using MRI as a comparator. This is now possible with whole-body MRI, which can depicture the entire axial skeleton and peripheral joints in each session.8 The use of contrast in whole-body MRI is challenging as prolonged scan time can lead to incorrect interpretation of synovial contrast enhancement.24 However, inflammation can be evaluated without contrast by detecting and quantifying effusion and synovial thickening.

    The objective of this study was to assess the construct validity of ultrasound in patients with JIA, using a standardised scanning approach and a joint-specific semiquantitative scoring system for synovitis with age-divided reference atlas, by comparing ultrasound detected synovitis with non-contrast enhanced whole-body MRI findings of effusion/synovial thickening and clinical assessment of disease activity.

    Methods

    This cross-sectional study was conducted at the Department of Rheumatology, Oslo University Hospital (OUH) Rikshospitalet from September 2021 to December 2022. Patients fulfilling the International League of Associations for Rheumatology (ILAR) criteria for JIA,25 who were referred to an MRI of a joint or a whole-body MRI on clinical indication and attending the paediatric rheumatology clinic at OUH were consecutively included. Signed informed consent was obtained from parents and from patients when aged 16 years and older. The study was done in accordance with the Declaration of Helsinki and approved by the Norwegian Regional Committee for Medical and Health Research Ethics (REK 2018/805).

    Study assessment

    On study inclusion, the assigned MRI examination was converted to a non-contrast enhanced whole-body MRI since the intention was to evaluate many joints. Each patient was first clinically assessed by their treating rheumatologist, then ultrasound and whole-body MRI examinations were performed within an average of 1–2 days. No changes were made to the patient’s medications between the clinical, ultrasound and whole-body MRI assessments. The rheumatologists, the ultrasonographer and the radiologist were blinded to each other’s findings, but the age of the patient was known to all assessors. The ultrasonographer and the radiologist were also blinded to clinical information.

    Clinical and laboratory assessment

    Each patient was assessed by experienced rheumatologists who performed clinical 71-joints examination including the following joints: the temporomandibular, sternoclavicular, acromioclavicular, shoulder, elbow, wrist, metacarpophalangeal (MCP) 1–5, proximal interphalangeal (PIP) 1–5, distal interphalangeal (DIP) 2–5, hip, knee, ankle, subtalar, intertarsal, metatarsophalangeal (MTP) 1–5, toe 1–5, and the cervical spine. Clinical active arthritis (active joint) was defined as the presence of joint swelling or, if no swelling was present, the presence of joint pain/tenderness and limited range of motion.26 Presence or absence of active joints was recorded and an active joint sum score was calculated for each patient (range 0–71). Patient and disease variables including age, sex, JIA subgroup, disease duration, medications, duration of morning stiffness (min), patient/parent global assessment (PGA) of well-being (0–10 cm Visual Analogue Scale (VAS), 0=very well and 10=very poor) and the physician global assessment (PhGA) of disease activity (0–10 cm VAS, 0=no activity and 10=maximum activity) were recorded. Biochemical analyses included erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), antinuclear antibody, rheumatoid factor and anticyclic citrullinated peptide. Disease activity was measured using the 71-joints Juvenile Arthritis Disease Activity Score (JADAS71), with a total score of 0–101.27 The JADAS71 is calculated as the sum of the score of the four following components: the number of joints with active arthritis up to a maximum of 71 joints, PGA (0–10 cm VAS), PhGA (0–10 cm VAS) and the normalised ESR (0–10).27

    Ultrasound assessment

    One rheumatologist (NKS), with broad experience in musculoskeletal ultrasound (9 years), performed the ultrasound examinations according to a recently published standardised scanning protocol for patients with JIA.14 The following 14 joints (19 joint regions) were bilaterally assessed: anterior elbow, posterior elbow, radiocarpal, midcarpal, MCP2-3 (dorsal), PIP2-3 (dorsal and volar), hip, knee (suprapatellar recess and lateral parapatellar recess), tibiotalar, talonavicular, anterior subtalar, posterior subtalar and MTP2-3 (dorsal). Synovitis detection by ultrasound included B-mode (BM) and power Doppler (PD) findings and was defined according to the preliminary definitions for synovitis in children.12 Synovitis could be detected based on BM findings alone, but not on Doppler findings alone. BM findings included synovial effusion (abnormal, intra-articular, anechoic or hypoechoic material that is displaceable) and/or synovial hypertrophy (abnormal, intra-articular, anechoic or hypoechoic material that is non-displaceable), and PD signals had to be detected within synovial hypertrophy to be considered as a sign of synovitis.12 The images were scored at the time of acquisition for BM and PD findings according to a semiquantitative joint-specific scoring system for BM synovitis (grades 0–3) and PD activity (grades 0–3) with age-divided reference atlas for patients with JIA for each joint region.14 A printed version of the scoring system with reference atlas was available during the ultrasound assessment and the examination took an average of 27 min (range 13–39 min). Ultrasound synovitis sum scores were calculated for each patient for BM scores and PD scores (range 0–84) and for a combined score (BM+PD scores (range 0–168)). To avoid increased weighting of joints that were assessed from more than one view by ultrasound (elbow anterior/posterior, PIP2-3 dorsal/volar, knee suprapatellar/lateral parapatellar recess and subtalar anterior/posterior), the highest BM and PD value for each joint region was selected. The rheumatologist used the same ultrasound machine, a GE Logiq S8 machine with multifrequency linear probe (6–15 MHz) and hockey stick (8–18 MHz), and standardised settings for BM and PD (pulse repetition frequency 0.6 kHz, frequency 7.7 MHz and low wall filter) in all examinations. To ensure a clear description of the study, the European Alliance of Associations for Rheumatology (EULAR) recommendations checklist for the reporting of ultrasound studies in rheumatic and musculoskeletal diseases was used.28

    Whole-body MRI acquisition and assessment

    Non-contrast enhanced whole-body MRI was performed using Avanto fit 1.5T, Aera 1.5T and Vida Fit 3T (Siemens Healthineers, Erlangen, Germany), with local receiver coils covering the entire body. General anaesthesia was given to the youngest patients (under 5 years of age) to avoid motion artefacts during the examination. The protocol included T2 Turbo spin echo Dixon sequences: coronal plane from the skull base to the thighs; sagittal plane in both knees and ankles; oblique coronal plane in the sacroiliac joints and sagittal plane in the spine. The image acquisition parameters were preset: repetition time (TR)>2000 ms/echo time (TE) 92–111 ms, field of view (FOV) 15–35 cm, slice thickness 3–4 mm and in-plane resolution 0.39–0.55 mm2. Acquisition time was approximately 30 min. Images were deidentified and analysed using Sectra Picture Archiving and Communications Systems. Joint inflammation on whole-body MRI included effusion/synovial thickening defined as hyperintense signal intensity within the joint space distending the joint capsule on fluid sensitive sequence, as suggested by the MRI in JIA (JAMRI) Outcome Measures in Rheumatology (OMERACT) working group.29 The whole-body MRI images were scored by a radiologist (EK) with extensive experience in musculoskeletal MRI (more than 20 years). Effusion/synovial thickening was scored as absent, mild or moderate (grades 0–2) in large joints (elbows, hips and knees) and as absent or present (grades 0–1) in small joints according to a newly developed semiquantitative whole-body MRI scoring system for JIA.29 Effusion/synovial thickening scores from the same 28 joints that were assessed with ultrasound were included in the analyses, and a whole-body MRI effusion/synovial thickening sum score was calculated for each patient (range 0–34). Whole-body MRI was performed in all patients except for one where the consent to the examination was withdrawn by the parents.

    Patient and public involvement

    Patients did not actively participate in the development of this study, but the participants and the Norwegian Rheumatism Association have shown a great interest in the study and supported the project. Study results will be disseminated to patients and the public through the patient organisation’s website and newsletter.

    Statistical analysis

    Descriptive statistics were calculated as number (%), mean (range) or median (IQR) as appropriate. At patient level, Spearman’s correlation coefficient (rs) was used to calculate associations between ultrasound synovitis sum scores and whole-body MRI effusion/synovial thickening sum scores, and clinical active joint sum scores. In addition, associations between ultrasound synovitis sum scores and the JADAS71, CRP, ESR, PhGA and PGA were calculated. The strength of rs was defined as: very weak 0.0–0.19; weak 0.20–0.39; moderate 0.40–0.59; strong 0.60–0.79 and very strong 0.80–1.0. P-values<0.05 were regarded as statistically significant. Sum scores for ultrasound, whole-body MRI and clinical joint examination were calculated without any missing values. The JADAS71 was not calculated for patients missing PhGA or PGA (n=2 in total).

    At joint level, the 28 joints (elbow, radiocarpal, midcarpal, MCP2-3, PIP2-3, hip, knee, tibiotalar, talonavicular, subtalar and MTP2-3 joints) assessed in each patient, with complete ultrasound, whole-body MRI and clinical joint examination data, were included in the analyses, whereas joints with missing data were excluded. Findings in the wrist joint was recorded as one joint in the clinical assessment, but in the analysis at joint level this finding was allocated to both the radiocarpal and the midcarpal joints. Semiquantitative scores on ultrasound and whole-body MRI were dichotomised for further analyses. The threshold for abnormality was chosen to be BM grade≥2 for ultrasound findings. There is no consensus on the cut-off value for abnormal findings on ultrasound in children, but a BM grade≥2 as cut-off for synovitis has been used in another ultrasound study in JIA.30 We also performed sensitivity analysis with BM grade≥1 as threshold for abnormality. For whole-body MRI findings, the threshold for abnormality was chosen to be grade≥1. Sensitivity, specificity, positive predictive value and negative predictive value were calculated for findings of synovitis on ultrasound using whole-body MRI or clinical joint examination as reference. Analyses were performed using SPSS V29.

    Results

    Twenty-seven patients (89% girls), with a median age 12 years (IQR 3–14) were included. The median number of active joints was 4 (IQR 2–6) and the median JADAS71 was 13.4 (IQR 9.4–28.6). Demographic and clinical characteristics are described in table 1.

    View this table:
    Table 1

    Demographic and clinical characteristics of the patients with juvenile idiopathic arthritis (n=27)

    Patient level

    The median ultrasound synovitis sum score was 7.0 (IQR 4.0–12.0), 1.0 (IQR 0.0–5.0) and 9.0 (IQR 4.0–17.0), for BM synovitis sum scores, PD synovitis sum scores and combined synovitis sum scores, respectively. The median whole-body MRI effusion/synovial thickening sum score was 2.5 (IQR 0.8–7.0). Ultrasound combined synovitis sum scores strongly correlated with whole-body MRI effusion/synovial thickening sum scores (rs=0.74, p<0.01), and moderately with clinical active joint sum scores (rs=0.57, p<0.01). The correlation between ultrasound combined synovitis sum scores and the JADAS71 was also strong (rs=0.71, p<0.01). Correlations between ultrasound synovitis sum scores and measures of disease activity are presented in table 2.

    View this table:
    Table 2

    Spearman’s correlations (rs) between ultrasound synovitis sum scores ((B-mode (BM), power Doppler (PD) and combined (BM+PD) synovitis sum score) and whole-body MRI effusion/synovial thickening sum scores, active joint sum scores and other measures of disease activity at patient level

    Joint level

    In total, 692 joints were assessed with ultrasound, whole-body MRI and clinical examination, identifying 533 joints with normal findings.

    For whole-body MRI, 32 joints were missing as they were outside the field of view (elbows) or of poor image quality (PIP and MTP joints). For ultrasound, one MCP2, one MCP3, one PIP2 and one PIP3 joint were missing due to a venous catheter in the hand of a small patient (2 years old) that made it impossible to assess these joints.

    Ultrasound findings of synovitis at joint level demonstrated a sensitivity of 0.57 and a specificity of 0.96, a positive predictive value of 0.71 and a negative predictive value of 0.93 using whole-body MRI findings of effusion/synovial thickening as reference. When using active joints found on clinical examination as reference, ultrasound findings of synovitis at joint level had a sensitivity and specificity of 0.55 and 0.96, respectively, while the positive predictive value was 0.71 and the negative predictive value was 0.92. When performing sensitivity analysis with BM grade≥1 as threshold for abnormality, ultrasound demonstrated a higher sensitivity (0.64/0.66), but lower specificity (0.91/0.91) using whole-body MRI or clinical examination as reference, respectively.

    Synovitis was most frequently found in the knee (n=16) and the tibiotalar (n=12) joints with ultrasound, while effusion/synovial thickening was most frequently seen in the knee (n=19) and the subtalar joints (n=17) with whole-body MRI (figure 1). Active joints were most frequently found in the knee (n=18) and the subtalar joints (n=14). The overall frequency of normal and inflammatory findings in the 14 joints assessed with ultrasound, whole-body MRI and clinical examination, are presented in figure 2.

    Figure 1
    Figure 1

    Illustration of ultrasound detected synovitis (images A and C) (arrows) and non-contrast enhanced whole-body MRI findings of effusion/synovial thickening (images B and D) (arrows) in three joints in a 15-year-old with juvenile idiopathic arthritis. Longitudinal (suprapatellar recess) and transverse (lateral parapatellar recess) ultrasound scan of the knee joint showing B-mode (BM) grade 3 (A). Sagittal MRI scan showing effusion in the knee joint (B). Longitudinal dorsal ultrasound scan of the tibiotalar joint showing BM grade 2, and lateral ultrasound scan of the posterior subtalar joint showing BM grade 3 (C). Sagittal MRI scan showing effusion in the tibiotalar and subtalar joints (D).

    Figure 2
    Figure 2

    (A) Frequency of normal findings in 14 joints assessed with ultrasound, non-contrast enhanced whole-body MRI and clinical joint examination. (B) Frequency of joint inflammation in 14 joints assessed with ultrasound, non-contrast enhanced whole-body MRI and clinical joint examination. MCP, metacarpophalangeal; MTP, metatarsophalangeal; PIP, proximal interphalangeal.

    On ultrasound and whole-body MRI, 58/692 joints (8.4%) had findings of synovitis or effusion/synovial thickening, while 566/692 joints (81.8%) had normal findings on both modalities. Ultrasound detected synovitis in 24/590 joints (4.1%) that were normal on whole-body MRI, while effusion/synovial thickening was found in 44/610 joints (7.2%) on whole-body MRI that were considered normal on ultrasound.

    In total, 534/571 joints (93.5%) that were scored as BM grade 0 on ultrasound had normal findings on whole-body MRI. Of the joints scored as BM grade 1 on ultrasound, 32/39 joints (82.1%) were scored as normal on whole-body MRI. Of all joints scored as BM grade 2 or BM grade 3 on ultrasound, 21/38 joints (55.3%) and 37/44 joints (84.1%) were scored with findings of effusion/synovial thickening on whole-body MRI, respectively. In large joints (the elbow, hip and knee) scored as BM grade 3 on ultrasound, 8/12 joints (66.7%) were scored as grade 2 on whole-body MRI and 4/12 joints (33.3%) were scored as grade 1 on whole-body MRI. No joint scored as BM grade 3 on ultrasound had normal findings on whole-body MRI for these joints.

    Joint inflammation was found in 58/692 joints (8.4%) on both ultrasound and clinical examination, whereas 562/692 joints (81.2%) had normal findings. Ultrasound detected synovitis in 24/586 joints (4.1%) that were normal on clinical examination. In joints deemed normal on ultrasound, 48/610 (7.9%) were considered as active on clinical examination.

    Discussion

    To the best of our knowledge, this is the first study comparing ultrasound assessment of multiple joints with non-contrast enhanced whole-body MRI and clinical assessment of disease activity in patients with JIA. Our study shows that ultrasound synovitis sum scores are strongly correlated with whole-body MRI effusion/synovial thickening sum scores and clinical measures of disease activity in children and adolescents with JIA.

    A strong association between ultrasound synovitis sum scores and whole-body MRI joint inflammation sum scores has also been reported in adults with rheumatoid arthritis.31 To the best of our knowledge, no study has compared ultrasound findings of synovitis in multiple joints with MRI in patients with JIA. Laurell et al compared ultrasound with contrast enhanced MRI of the wrist, knee and ankle joints to evaluate disease activity in JIA, but only one joint was assessed with MRI in each patient. Their results were descriptive but indicated that MRI and ultrasound might be valuable in the evaluation of disease activity in JIA.32

    Reliable and validated tools to assess disease activity are needed for optimal management of patients with JIA. The ultrasonographic scoring system for synovitis with age-divided reference atlas used in this study has previously shown moderate to excellent reliability. The next step in the validation process was therefore to test the scoring system in relation to other imaging modalities and measures of disease activity. We found a strong correlation between ultrasound synovitis sum scores and whole-body MRI effusion/synovial thickening sum scores, CRP, PhGA and the JADAS71. Magni-Manzoni et al found a poor correlation between ultrasound synovitis findings and the JADAS,18 others have found a moderate correlation.15 The weak correlation between ultrasound synovitis findings and PGA in our study is in line with previously published findings.18 While ultrasound evaluates the level of inflammation, PGA might also be affected by other confounding factors.

    Ultrasound demonstrated high specificity, but lower sensitivity in detecting synovitis using whole-body MRI or clinical joint examination as reference. It has previously been shown that ultrasound is more sensitive than clinical examination in the evaluation of joint inflammation.9 Vega-Fernandez et al found that ultrasound had a sensitivity of 0.83 for diagnosing synovitis in the knee joint when using contrast-enhanced MRI as reference.23 The sensitivity demonstrated in this study may have many contributing explanations. The participants were referred to an MRI or whole-body MRI on clinical indication prior to inclusion. MRI is most often performed when there is diagnostic doubt, and this usually applies to joints or joint regions that are difficult to assess clinically and with ultrasound. The affected joints could therefore have been complex and challenging to evaluate. The treating rheumatologist that performed the clinical joint examination was not blinded to the medical history, reason for admission or clinical symptoms of the patients. The ultrasonographer was blinded to all clinical information except for the age of the patient, which was known. This may have influenced the findings. Additional explanations may be involvement of several complex joints, the age of the patient and cooperation during the examinations.

    Contrast enhanced MRI is considered the gold standard for assessing synovitis,7 8 and since whole-body MRI can assess many joints in each examination we found it suitable as a comparator to ultrasound and our joint-specific scoring system where many joints are being evaluated. The most frequently detected joint with inflammation on ultrasound, whole-body MRI and clinical examination was the knee joint, which is also the most commonly affected joint in JIA.3 Whole-body MRI detected inflammation in the knee joint more often than ultrasound. This is in line with others who found that contrast enhanced MRI was superior to ultrasound in the evaluation of the knee joint.16 Whole-body MRI also detected effusion/synovial thickening more often in the subtalar and the talonavicular joints than ultrasound detected synovitis, but ultrasound found more frequently synovitis in the tibiotalar joint. This may be due to the complex anatomy in the ankle joint.5 6 33 34 In addition, the positioning of the knee and the ankle joint during the ultrasound and whole-body MRI examinations differed, which may have affected the distribution and organisation of the effusion, as also suggested by others.32

    Ultrasound demonstrated a sensitivity of 0.57 using whole-body MRI as reference, but only 55.3% of the joints scored as BM grade 2 on ultrasound had findings of effusion/synovial thickening on whole-body MRI. This may be explained by a higher threshold to score abnormality on whole-body MRI using a binary scoring (vs semiquantitative on ultrasound) and the lack of contrast enhanced images. Differences in the distribution of effusion due to the positioning of the joints during the examinations could also be a contributing factor. In addition, the complex anatomy of some joints and the use of different anatomical planes (for instance, sagittal vs coronal plane), may have affected the scoring in joints like the ankle, wrist and MTP-joints. However, most of the joints with severe findings on ultrasound (BM grade 3) were also scored with effusion/synovial thickening findings on whole-body MRI (84.1%), and a high number of joints with BM grade 0 and BM grade 1 on ultrasound had normal findings on whole-body MRI. For large joints (the elbow, hip and knee), all joints scored as a BM grade 3 on ultrasound had corresponding effusion/synovial thickening on whole-body MRI. This suggests that the scoring system corresponds well with the severity of synovitis. Interestingly, 32/39 joints (82.1%) with a BM grade 1 on ultrasound (defined as normal in this study) were scored as normal on whole-body MRI. However, if PD signals are detected within synovial hypertrophy in a joint scored as BM grade 1 on ultrasound, it is considered abnormal.12 35 This was not an issue in this study, but it can be discussed whether a BM grade 1 on ultrasound represents a normal finding or mild synovitis since there is a lack of defined cut-off levels. In addition, most of the joints were scored binary on whole-body MRI potentially leading to the loss of minimal findings, as also indicated by others.36

    Limitations of this study include that only one rheumatologist performed the ultrasound examinations. However, this was done for feasibility reasons and the rheumatologist has shown moderate to excellent reliability in a previous study.14 Another limitation is that only one radiologist scored the whole-body MRI images. The radiologist has extensive experience in musculoskeletal MRI in children. Not all ultrasound and whole-body MRI examinations were performed on the same day, which may have caused discrepancies in the findings. However, there were no adjustments in the patients’ medications in the time between the examinations. The whole-body MRI missed many elbow joints because the joint often was outside the field of view. In addition, small joints (PIP) were difficult to visualise on whole-body MRI compared with standard MRI. Furthermore, we did not use contrast during the whole-body MRI examinations, which might have made it more difficult to evaluate joint pathology. However, since we wanted to assess many joints in one session, this was not possible. The small number of patients in this study may also provide uncertain results. Finally, since this was not a longitudinal study, we cannot say if ultrasound is responsive to change, this should be addressed in a future study.

    The strengths of this study are that many joints were assessed with the same ultrasound machine, using a standardised scanning approach. In addition, that ultrasound definitions for synovitis in children were used and findings scored semiquantitatively with a joint-specific scoring system for synovitis with reference atlas for patients with JIA. Further, that the whole-body MRI images were scored according to a newly developed whole-body MRI scoring system for patients with JIA and that a standardised scanning method was used.

    In summary, our findings provide a validation of ultrasound in JIA showing that ultrasound synovitis sum scores strongly correlate with non-contrast enhanced whole-body MRI effusion/synovial thickening sum scores and clinical measures of disease activity. This indicates that the scanning protocol and scoring system provides a valid assessment of synovitis and suggests that ultrasound synovitis sum scores can reflect overall disease activity and may be a useful outcome measure in clinical practice and research. Future longitudinal studies are needed to evaluate the ultrasound synovitis sum score as a measure of disease activity and to explore the optimal set of joints to assess with ultrasound in children with JIA to increase feasibility.

    Data availability statement

    Data are available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants. The study was approved by the Norwegian Regional Committee for Medical and Health Ethics (REK 2018/805) and done in accordance with the Declaration of Helsinki. Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    We would like to thank all the patients and their families who contributed to this study. We also thank Lien My Diep, Researcher (statistician), MSc, for statistical advice. The authors are grateful for the support from the Norwegian Rheumatism Association and the Simon Fougner Hartmann’s Family Foundation.

    References

    1. ↵
      1. Flatø B,
      2. Lien G,
      3. Smerdel A, et al
      . Prognostic factors in juvenile rheumatoid arthritis: a case-control study revealing early predictors and outcome after 14.9 years. J Rheumatol 2003;30:386–93.
      OpenUrlAbstract/FREE Full Text
    2. ↵
      1. Eng SWM,
      2. Aeschlimann FA,
      3. van Veenendaal M, et al
      . Patterns of joint involvement in juvenile idiopathic arthritis and prediction of disease course: a prospective study with multilayer non-negative matrix factorization. PLoS Med 2019;16:e1002750. doi:10.1371/journal.pmed.1002750
    3. ↵
      1. Ravelli A,
      2. Martini A
      . Juvenile idiopathic arthritis. Lancet 2007;369:767–78. doi:10.1016/S0140-6736(07)60363-8
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Rooney ME,
      2. McAllister C,
      3. Burns JFT
      . Ankle disease in juvenile idiopathic arthritis: ultrasound findings in clinically swollen ankles. J Rheumatol 2009;36:1725–9. doi:10.3899/jrheum.080508
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Pascoli L,
      2. Wright S,
      3. McAllister C, et al
      . Prospective evaluation of clinical and ultrasound findings in ankle disease in juvenile idiopathic arthritis: importance of ankle ultrasound. J Rheumatol 2010;37:2409–14. doi:10.3899/jrheum.091262
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Lanni S,
      2. Marafon DP,
      3. Civino A, et al
      . Comparison between clinical and ultrasound assessment of the ankle region in children with juvenile idiopathic arthritis. Arthritis Care Res (Hoboken) 2021;73:1180–6. doi:10.1002/acr.24241
      OpenUrl
    7. ↵
      1. Damasio MB,
      2. Malattia C,
      3. Martini A, et al
      . Synovial and inflammatory diseases in childhood: role of new imaging modalities in the assessment of patients with juvenile idiopathic arthritis. Pediatr Radiol 2010;40:985–98. doi:10.1007/s00247-010-1612-z
      OpenUrlCrossRefPubMed
    8. ↵
      1. Malattia C,
      2. Tolend M,
      3. Mazzoni M, et al
      . Current status of MR imaging of juvenile idiopathic arthritis. Best Pract Res Clin Rheumatol 2020;34:101629. doi:10.1016/j.berh.2020.101629
      OpenUrl
    9. ↵
      1. Colebatch-Bourn AN,
      2. Edwards CJ,
      3. Collado P, et al
      . EULAR-Pres points to consider for the use of imaging in the diagnosis and management of juvenile idiopathic arthritis in clinical practice. Ann Rheum Dis 2015;74:1946–57. doi:10.1136/annrheumdis-2015-207892
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Roth J,
      2. Jousse-Joulin S,
      3. Magni-Manzoni S, et al
      . Definitions for the sonographic features of joints in healthy children. Arthritis Care Res (Hoboken) 2015;67:136–42. doi:10.1002/acr.22410
      OpenUrl
    11. ↵
      1. Collado P,
      2. Vojinovic J,
      3. Nieto JC, et al
      . Toward standardized musculoskeletal ultrasound in pediatric rheumatology: normal age-related ultrasound findings. Arthritis Care Res (Hoboken) 2016;68:348–56. doi:10.1002/acr.22670
      OpenUrl
    12. ↵
      1. Roth J,
      2. Ravagnani V,
      3. Backhaus M, et al
      . Preliminary definitions for the sonographic features of synovitis in children. Arthritis Care Res (Hoboken) 2017;69:1217–23. doi:10.1002/acr.23130
      OpenUrl
    13. ↵
      1. Ting TV,
      2. Vega-Fernandez P,
      3. Oberle EJ, et al
      . Novel ultrasound image acquisition protocol and scoring system for the pediatric knee. Arthritis Care Res (Hoboken) 2019;71:977–85. doi:10.1002/acr.23746
      OpenUrl
    14. ↵
      1. Sande NK,
      2. Bøyesen P,
      3. Aga A-B, et al
      . Development and reliability of a novel ultrasonographic joint-specific scoring system for synovitis with reference atlas for patients with juvenile idiopathic arthritis. RMD Open 2021;7:e001581. doi:10.1136/rmdopen-2021-001581
    15. ↵
      1. Vega-Fernandez P,
      2. Ting TV,
      3. Oberle EJ, et al
      . Musculoskeletal ultrasound in childhood arthritis limited examination: a comprehensive, reliable, time-efficient assessment of synovitis. Arthritis Care Res (Hoboken) 2023;75:401–9. doi:10.1002/acr.24759
      OpenUrl
    16. ↵
      1. El-Miedany YM,
      2. Housny IH,
      3. Mansour HM, et al
      . Ultrasound versus MRI in the evaluation of juvenile idiopathic arthritis of the knee. Joint Bone Spine 2001;68:222–30. doi:10.1016/s1297-319x(01)00269-x
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Karmazyn B,
      2. Bowyer SL,
      3. Schmidt KM, et al
      . US findings of metacarpophalangeal joints in children with idiopathic juvenile arthritis. Pediatr Radiol 2007;37:475–82. doi:10.1007/s00247-007-0438-9
      OpenUrlCrossRefPubMed
    18. ↵
      1. Magni-Manzoni S,
      2. Epis O,
      3. Ravelli A, et al
      . Comparison of clinical versus ultrasound-determined synovitis in juvenile idiopathic arthritis. Arthritis Rheum 2009;61:1497–504. doi:10.1002/art.24823
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Collado P,
      2. Jousse-Joulin S,
      3. Alcalde M, et al
      . Is ultrasound a validated imaging tool for the diagnosis and management of synovitis in juvenile idiopathic arthritis? A systematic literature review. Arthritis Care Res (Hoboken) 2012;64:1011–9. doi:10.1002/acr.21644
      OpenUrl
    20. ↵
      1. Eich GF,
      2. Hallé F,
      3. Hodler J, et al
      . Juvenile chronic arthritis: imaging of the knees and hips before and after Intraarticular steroid injection. Pediatr Radiol 1994;24:558–63. doi:10.1007/BF02012732
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Haslam KE,
      2. McCann LJ,
      3. Wyatt S, et al
      . The detection of subclinical synovitis by ultrasound in oligoarticular juvenile idiopathic arthritis: a pilot study. Rheumatology (Oxford) 2010;49:123–7. doi:10.1093/rheumatology/kep339
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Filippou G,
      2. Cantarini L,
      3. Bertoldi I, et al
      . Ultrasonography vs. clinical examination in children with suspected arthritis. Does it make sense to use poliarticular ultrasonographic screening? Clin Exp Rheumatol 2011;29:345–50.
      OpenUrlPubMed
    23. ↵
      1. Vega-Fernandez P,
      2. Rogers K,
      3. Sproles A, et al
      . Diagnostic accuracy study of the pediatric-specific ultrasound scoring system for the knee joint in children with juvenile idiopathic arthritis. Arthritis Care Res (Hoboken) 2024;76:251–8. doi:10.1002/acr.25218
      OpenUrl
    24. ↵
      1. Panwar J,
      2. Patel H,
      3. Tolend M, et al
      . Toward developing a semiquantitative whole body-MRI scoring for juvenile idiopathic arthritis: critical appraisal of the state of the art, challenges, and opportunities. Acad Radiol 2021;28:271–86. doi:10.1016/j.acra.2020.01.022
      OpenUrl
    25. ↵
      1. Petty RE,
      2. Southwood TR,
      3. Manners P, et al
      . International League of associations for rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, 2001. J Rheumatol 2004;31:390–2.
      OpenUrlFREE Full Text
    26. ↵
      1. Ravelli A,
      2. Viola S,
      3. Ruperto N, et al
      . Correlation between conventional disease activity measures in juvenile chronic arthritis. Ann Rheum Dis 1997;56:197–200. doi:10.1136/ard.56.3.197
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Consolaro A,
      2. Ruperto N,
      3. Bazso A, et al
      . Development and validation of a composite disease activity score for juvenile idiopathic arthritis. Arthritis Rheum 2009;61:658–66. doi:10.1002/art.24516
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Costantino F,
      2. Carmona L,
      3. Boers M, et al
      . EULAR recommendations for the reporting of ultrasound studies in rheumatic and musculoskeletal diseases (RMDs). Ann Rheum Dis 2021;80:840–7. doi:10.1136/annrheumdis-2020-219816
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Panwar J,
      2. Tolend M,
      3. Redd B, et al
      . Consensus-driven conceptual development of a standardized whole body-MRI scoring system for assessment of disease activity in juvenile idiopathic arthritis: MRI in JIA OMERACT working group. Semin Arthritis Rheum 2021;51:1350–9. doi:10.1016/j.semarthrit.2021.07.017
      OpenUrl
    30. ↵
      1. Vega-Fernandez P,
      2. Oberle EJ,
      3. Henrickson M, et al
      . Musculoskeletal ultrasound and the assessment of disease activity in juvenile idiopathic arthritis. Arthritis Care Res (Hoboken) 2023;75:1815–20. doi:10.1002/acr.25073
      OpenUrl
    31. ↵
      1. Ng SN,
      2. Axelsen MB,
      3. Østergaard M, et al
      . Whole-body magnetic resonance imaging assessment of joint inflammation in rheumatoid arthritis-agreement with ultrasonography and clinical evaluation. Front Med (Lausanne) 2020;7:285. doi:10.3389/fmed.2020.00285
    32. ↵
      1. Laurell L,
      2. Court-Payen M,
      3. Nielsen S, et al
      . Comparison of ultrasonography with doppler and MRI for assessment of disease activity in juvenile idiopathic arthritis: a pilot study. Pediatr Rheumatol Online J 2012;10:23. doi:10.1186/1546-0096-10-23
    33. ↵
      1. Lanni S,
      2. Bovis F,
      3. Ravelli A, et al
      . Delineating the application of ultrasound in detecting synovial abnormalities of the subtalar joint in juvenile idiopathic arthritis. Arthritis Care Res (Hoboken) 2016;68:1346–53. doi:10.1002/acr.22846
      OpenUrl
    34. ↵
      1. Remedios D,
      2. Martin K,
      3. Kaplan G, et al
      . Juvenile chronic arthritis: diagnosis and management of Tibio-Talar and sub-Talar disease. Br J Rheumatol 1997;36:1214–7. doi:10.1093/rheumatology/36.11.1214
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Sande NK,
      2. Lilleby V,
      3. Aga A-B, et al
      . Associations between power doppler ultrasound findings and B-mode synovitis and clinical arthritis in juvenile idiopathic arthritis using a standardised scanning approach and scoring system. RMD Open 2023;9:e002937. doi:10.1136/rmdopen-2022-002937
    36. ↵
      1. D’Agostino M-A,
      2. Terslev L,
      3. Aegerter P, et al
      . Scoring ultrasound synovitis in rheumatoid arthritis: a EULAR-OMERACT ultrasound taskforce-part 1: definition and development of a standardised, consensus-based scoring system. RMD Open 2017;3:e000428. doi:10.1136/rmdopen-2016-000428

    Footnotes

    • Contributors All authors were involved in drafting the article or revising it critically for important intellectual content, approved the final manuscript to be published and agreed to be accountable for all aspects of the work. NKS designed the study, made substantial contribution to acquisition, analyses and interpretation of data. PB designed the study and made substantial contribution to acquisition and interpretation of data. EK, VL and BF participated in the study design and made substantial contributions to acquisition and interpretation of data. AHT participated in the study design and made substantial contribution to acquisition of data. ABA made substantial contribution to acquisition and interpretation of data. PB is the guarantor of the study.

    • Funding This study was funded by the DAM foundation.

    • Competing interests ABA reports personal fees from AbbVie, Eli Lilly, Novartis and Pfizer.

    • Provenance and peer review Not commissioned; externally peer reviewed.