The joints in juvenile idiopathic arthritis
- 2.8k Downloads
Juvenile idiopathic arthritis is the most common rheumatic entity in childhood. Imaging has become an important supplement to the clinical assessment of children with JIA. Radiographs still play an important role in the workup, and long-term follow-up in children with JIA, but are not sensitive to findings in the early disease stage. Both ultrasound and MRI are more sensitive to inflammatory changes than clinical assessment alone. However, the differentiation between normal findings and pathology can be challenging, particularly in early disease. The objective of this review is to discuss the role of imaging in JIA, describe the typical findings on different modalities and highlight the challenges we face regarding the reliability and accuracy of the different methods for imaging the joints in children with JIA.
• Imaging is an important supplement to the clinical examination in JIA.
• Ultrasound is more sensitive for detecting synovitis than clinical examination alone.
• MRI can depict all relevant structures in joint inflammation.
• The differentiation between normal variants and pathology is difficult in children.
KeywordsJuvenile idiopathic arthritis Magnetic resonance imaging Ultrasound Paediatrics Musculoskeletal
Summary of the main features of the six different subtypes of JIA
Subset of JIA
Age at onset
Early childhood, peak 2–4 years
Four or fewer joints involved the first 6 months
F> > M
Poly RF negative
Early peak 2–4; late peak 6–12 years
Four or more joints involved within the first 6 months, absence of IgM RF. Heterogeneous disease with three subsets. Prognosis varies with the disease subset
Poly RF positive
Four or more joints involved within the first 6 months, IgM RF positive. Resembles adult RA. Involvement of small joints. Progressive and diffuse joint involvement
F> > M
Characterised by enthesitis and arthritis. Often HLA-B27 positive. Commonly hip involvement at presentation. Often a mild and remitting course but may progress with sacroiliac and spinal joint involvement, resembling ankylosing spondylitis
M> > F
Early peak 2–4 years; late peak 9–11 years
Arthritis and psoriatic rash or psoriasis in close family. Controversial definition, resembles oligoarthritis but more often with dactylitis and involvement of both small and large joints
F > M
Arthritis and quotidian fever plus one or more of the following symptoms: characteristic rash, hepatomegaly, splenomegaly, lymphadenopathy, serositis. Variable course; 5–8 % develop macrophage activation syndrome
F = M
Typical manifestations of the articular and periarticular inflammation of JIA in early and late disease. The order of presentation is not absolute; some of the features may not be present at all or the features may overlap
Periarticular soft tissue change
Periarticular bony change
Destructive change of bone and cartillage
Erosive change of bone
Joint space narrowing, bone fusion, malalignment
The objective of this review is to discuss the role of imaging in JIA, describe the typical findings on different modalities and highlight the challenges we face regarding the reliability and accuracy of the different methods for imaging the joints in children with JIA.
The role of imaging in JIA
Currently, the diagnosis of JIA is based on clinical and laboratory findings and does not include imaging. However, clinical assessment of children with joint complaints is challenging and laboratory findings are often equivocal. This has led to an increased use of imaging for both diagnosis and follow-up, particularly of joints that are frequently affected but may be asymptomatic, such as the temporomandibular joint (TMJ) .
Imaging is used to determine the presence and extension of joint inflammation and can more accurately distinguish between arthritis and tenosynovitis than clinical examination alone. In some cases, it can also help define the subtype of JIA. Imaging has a role in the assessment of differential diagnosis, such as leukaemia, avascular necrosis, trauma or bone tumours, and also shows potential in the evaluation of treatment response.
There are no unifying international recommendations for imaging in JIA and both the choice of imaging modality and when and whom to refer for radiological investigations vary between centres. One of the major challenges when interpreting imaging studies is to distinguish between normal findings and pathology in the early stages of the disease, particularly on MRI. Moreover, the lack of standardised scoring systems for all modalities makes the objective evaluation of the degree of inflammation and destruction challenging. Plain radiographs of the affected joint are often not helpful to establish the diagnosis but are advised as a baseline study for longitudinal follow-up and to help narrow the list of differentials or suggest alternative diagnoses, for example an underlying bone tumour or a fracture. Ultrasound and MRI form the basis of more advanced imaging assessment in paediatric arthritis, while cone-beam computed tomography play a role in imaging of the TMJs and conventional CT in imaging of the sacro-iliac and facet joint of the cervical spine .
Ultrasound has many advantages in children. It is readily available, non-ionising, dynamic and well tolerated without recourse to anaesthesia or sedation; hence, it is a frequently used radiological examination in a child with JIA. Small patient size is both an advantage and disadvantage—using modern high-resolution transducers, exquisite detail is possible but requires a meticulous approach and adaptation of scanning equipment. Small “hockey-stick” transducers can be used for small joints but also for larger joints, e.g. ankle scanning in small children. Judicious use of warm jelly, often as a stand-off for small joints, and a child-friendly environment increase the chances of success.
Ultrasound may also be used in interventional procedures, for both diagnostic purposes, i.e. aspiration of a joint, and treatment, i.e. tendon sheath or joint injection . Despite the stated advantages of ultrasound, there are several limitations. Musculoskeletal ultrasound (MSK US) is not a widespread skill, particularly amongst paediatric radiologists, and there is a learning curve when undertaking these examinations. Ultrasound is an operator-dependent modality, taking time to become proficient. Close collaboration between the clinical team and radiologist, with multidisciplinary discussion where there are discordant findings between the clinical examination and the sonographic assessment, facilitates appropriate management. It is recognised that US will depict more synovitis in JIA than clinically detected , but this may represent old inactive changes rather than active disease. Doppler examination may be used to differentiate active from inactive disease, but there is no unifying definition of hyperaemia, and the Doppler appearance will vary with different probes, settings, vendors and the imaging technique (e.g. probe pressure). Therefore the interpretation of the findings, particularly in early disease/subtle findings, is challenging. Comparison with a contra-lateral unaffected joint or the adjacent non-inflamed synovium may help in the evaluation. Furthermore, ultrasound cannot assess changes in the bone marrow. Finally the stages of development and US imaging of the normal growing peri-articular skeletal tissues throughout childhood are in general not well documented (aside from Spannow et al. described above); hence, normal standards for US anatomy are not yet properly established.
Despite the challenges of MSK US in paediatrics there is a growing acceptance of its use and availability. US lends itself to one-stop clinics to establish a diagnosis, formulate a treatment plan and possibly US-guided aspiration or injection.
T1 SE sequence,
T2 fat-suppressed sequence or a STIR and
T1 fat-suppressed sequence pre- and post-contrast.
Some may prefer to add a sequence for visualisation of cartilage, such as a proton-weighted or 3D DESS sequence. The DIXON fat-suppression technique is increasingly being used because it gives robust, homogeneous fat suppression and can be combined with high-resolution T2-weighted images, giving a better signal-to-noise ratio compared to the STIR sequence, without increasing the scan time . Contrast administration is necessary for the assessment of synovitis; hence contrast enhanced MRI is part of the standard protocol in JIA. The preferred image planes will vary depending on the joint but the MRI protocol should allow multiplanar reformatting so potential findings, particularly evidence of joint destruction, could be verified in at least two planes.
Cone-beam computed tomography (CBCT) is currently accepted as the imaging modality of choice for assessment of permanent bony changes to the TMJs. It can visualise joint space narrowing and erosive change, as well as growth disturbances with a relatively low radiation dose .
Computed tomography (CT) may be helpful in the assessment of the facet joints, in particular in the cervical spine. Its use is, however, limited by the relatively high radiation dose.
Validation of imaging markers for active disease and permanent joint damage
There have been major advances in the treatment of JIA during the last decade. The development of new therapeutic agents and new, individually tailored treatment strategies has led to significant improvement in functional outcome in children with JIA [24, 25]. The paramount goal of current treatment in JIA is to achieve inactive disease and remission with or without medication . In order to evaluate the therapeutic response, sensitive and reproducible tools for assessment of change in the severity of inflammation and for bony destruction have become crucial. Of the diagnostic tools currently available, imaging studies are thought best suited for these purposes . The radiological investigations in JIA should ideally be able to determine the presence and degree of (1) active inflammation, (2) precursors of permanent articular damage, (3) established articular damage and (4) complications to treatment.
Synovial contrast enhancement, bone marrow oedema (BMO) and increased joint fluid are all thought to be signs of active inflammation. The presence of bone erosions is a sign of structural joint damage and the presence of BMO may be predictive of later bone erosions [27, 28]. These definitions of active inflammation and destruction are based on research on patients with rheumatoid arthritis and have been adapted for use in children. Recent studies have shown that definitions extrapolated from research in the adult population cannot automatically be used in children and may lead to both over- and under-staging of disease [21, 29, 30, 31]. Therefore separate validation of the different modalities when used in children must be performed in order to provide systems that can reliably quantify the degree of inflammation and destruction in JIA. Large efforts have been made over the last decade to provide objective imaging tools for the quantification of disease activity in children with JIA. A summary of the current status is provided below:
Radiographic scores are only feasible for structural change and not for inflammation. Radiographic scoring systems specific for joint destruction at the wrist in JIA have been devised and show good reproducibility and clinical validity . One initial validation of a scoring system for chronic changes at the hip has been published  but validated radiographic scoring systems for other joints are lacking.
Apart from studies of the age-dependent cartilage thickness , normal appearances for the various joints in a growing child throughout childhood are not well documented on ultrasound. In addition, to date there are no unifying definitions of synovial thickening, pathological amounts of joint fluid, hyperaemia or destructive change on ultrasound; hence, no validated scoring systems for disease activity and progression are available and the interpretation of the US findings in JIA remains subjective [12, 34].
Great efforts have been made by different national and international study groups, e.g. the Health-e-Child Radiology Group and the Outcome Measures in Rheumatology Clinical Trials (OMERACT) Special Interest Group ‘MRI in JIA’ to create and validate a universal paediatric MRI scoring system . This work has shown that in multi-centre studies the assessment and semi-quantification of BMO and tenosynovitis have acceptable repeatability for clinical use, whereas synovial enhancement and overall inflammation can be accurately scored by the same observer while inter-observer agreement was less favourable, but acceptable for the assessment of the radio-ulnar and radio-carpal joints [36, 37]. Moreover, recent studies showing that many of the findings that were previously defined as pathology also appear in healthy children have altered the whole perception of the accuracy of the MRI features included in previous scoring systems and the main challenge we now face is how to differentiate normal findings from disease. Bone erosion is the only feature of destruction included in the suggested scoring systems for JIA on MRI. However, it has previously been suggested, when using radiographs in the assessment, that bone erosions may not even be the most significant sign of bone destruction at the wrist, particularly in younger children, and that carpal crowding or squaring of the carpal bones causing a small and deformed wrist is more characteristic for JIA [10, 38]. This was later confirmed in a study comparing healthy children and those with JIA showing that children with JIA had significantly smaller wrists, but there was no difference in the number of erosion-like changes . Including erosions only in the assessment of bone destruction in children may therefore lead to both over- and under-staging of disease. There is ongoing multinational collaboration among several research groups to create accurate and reliable scoring systems for JIA, for both large and small joints, but to date an internationally validated and accepted scoring system for JIA is not yet available for any of the joints on MRI.
Imaging is a useful diagnostic supplement to the clinical assessment of children with JIA. Following the introduction of new, potent biologic drugs, there is an urgent need for accurate markers of disease, both active arthritis and permanent joint damage. Imaging, in particular MRI, has great potential, but more studies addressing the precision and accuracy as well as the clinical validity of the findings should be done.
- 35.Damasio MB, Malattia C, Tanturri de Horatio L et al (2012) MRI of the wrist in juvenile idiopathic arthritis: proposal of a paediatric synovitis score by a consensus of an international working group. Results of a multicentre reliability study. Pediatr Radiol 42:1047–1055CrossRefPubMedGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.