Terminology, definitions, and classification

Previously, there were two clinical classifications of arthritis: the European, juvenile chronic arthritis (JCA) and the American, juvenile rheumatoid arthritis (JRA). They had different inclusion and exclusion criteria, which complicated international communication, the diagnosis and treatment of patients, and the interpretation of outcomes research. In order to define homogeneous subgroups of arthritis in children, to ensure uniform collection of information, and to clarify communication between clinicians worldwide, the ILAR introduced new nomenclature in 1995 (with revision in 2004) combining the European and American classifications into one system. The goal of this change in nomenclature was to advance understanding of the pathogenesis of juvenile arthritis and to improve treatment and outcomes in affected children. In this article, the designation JIA as defined and accepted by the ILAR is substituted for the previous designation JRA.

JIA is a broad term, describing a group of heterogeneous arthritides, which includes all forms of arthritis in children. Patients with JIA have a systemic disease that results in inflammation in the synovium and destruction of hard and soft tissues in joints. The temporomandibular joint (TMJ) is involved in more than 40% of patients diagnosed with JIA.

JIA is the most common pediatric rheumatologic condition and affects at least 300,000 children in the USA. The international prevalence ranges from 8 to 400 per 100,000, reflecting variations in different geographic locations. The etiology and pathogenesis of JIA are unknown; a combination of environmental and certain immunogenic factors may contribute to the disease. Without appropriate treatment, it may cause irreversible damage and destruction of hard and soft tissues in joints (including the TMJs) and viscera, leading to devastating physical limitations and disfigurement. There is no single pathognomonic abnormality that defines these disorders. JIA is categorized into subgroups based on history, clinical characteristics, and laboratory findings ( Table 1 ). Most children with JIA have a disease that is phenotypically different from that of adults with rheumatoid arthritis. However, 5–10% of children with JIA demonstrate serum positivity for rheumatoid factor (RF) and anti-cyclic citrullinated peptide (CCP) antibodies and are thus diagnosed with childhood onset rheumatoid arthritis (CORA), or CCP/RF-positive JIA. JIA is also differentiated from idiopathic condylar resorption (ICR), which affects only the TMJs and is not accompanied by joint inflammation and synovitis.

Diagnosis of juvenile idiopathic arthritis

The diagnosis of JIA is based on a combination of medical history, clinical presentation, and radiologic and laboratory abnormalities. Patients with juvenile arthritis must have at least two of the following clinical features: joint swelling, pain/tenderness with motion, limitation of joint motion, and calor (warmth or heat) overlying the joint for more than 6 weeks. Appropriate and timely diagnosis and management are essential to relieve symptoms and to prevent, control, or limit the multi-organ sequelae of this disease. There are seven categories of JIA: (1) systemic onset, (2) oligoarthritis, (3) RF-positive polyarthritis, (4) RF-negative polyarthritis, (5) psoriatic arthritis, (6) enthesitis (inflammation at the tendon/ligament insertion into bone)-related arthritis, and (7) undifferentiated arthritis ( Table 1 ).

In systemic JIA, children present with fever, rash, poor weight gain/failure to thrive, decreased growth velocity, anemia, lymphadenopathy, hepatosplenomegaly, and serositis. Skin and nail changes (e.g., dactylitis) are commonly found in children with psoriatic arthritis. Uveitis is a reported complication of oligoarthritis, enthesitis-related arthritis, and less commonly polyarthritis and psoriatic arthritis.

Clinical findings vary and depend on the location, subtype of arthritis, and duration of disease. Children demonstrate a variety of findings, including a limp, morning stiffness, and/or antalgic gait, with difficulty walking or running. They may exhibit extremity overgrowth or undergrowth and asymmetries such as leg length discrepancies in the axial skeleton. Depending on the duration of symptoms, patients will have muscular atrophy around the affected joints and contractures.

If patients seek medical attention soon after the onset of symptoms, plain radiographs often do not show joint or skeletal abnormalities. Subtle radiographic findings that suggest arthritis in the early stages include mild peri-articular osteopenia with soft tissue swelling. In aggressive or rapidly progressive disease, or when there is a significant delay in the diagnosis and initiation of treatment, effusions, erosions, bony overgrowth, and joint space narrowing may be seen. Magnetic resonance imaging (MRI) with intravenous gadolinium is the gold standard imaging method to detect active synovial inflammation (enhancement of the intra-articular soft tissue). The extent of synovitis helps differentiate active inflammation and joint damage from long-standing chronic arthritis.

The detection of synovitis is a key aspect of JIA diagnosis and management and may prevent long-term disability. Although, synovitis can be diagnosed clinically by joint swelling, pain, tenderness, and decreased range of motion, gadolinium-enhanced MRI more accurately reflects synovitis than clinical examination. Active synovitis may be identified in presumed clinically inactive patients with this method.

Laboratory studies frequently provide additional information. For example, patients with significant inflammation may have anemia and/or elevated inflammatory markers such as erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). However, normal laboratory studies do not rule out active arthritis. For example, it is a common misconception that negative antinuclear antibody (ANA) and RF studies exclude the diagnosis of JIA. Of all the children with JIA, only 5–10% have the RF-positive polyarticular JIA subtype. On the other hand, RF-negative polyarticular JIA represents 15–25% of all JIA cases.

Arthritis in children is not simply adult rheumatoid arthritis in a child. There are significant differences in the diseases. Most children with JIA who receive appropriate treatment do not develop the typical ‘rheumatoid’ appearance, i.e., hand/limb/spine deformities. However, the longer the disease is active, the more the child is at risk of developing these deformities. Polyarthritis in combination with an elevated RF is also less common in children. Unlike adults, children may not complain of pain at first, and joint swelling may not be noticeable or can be delayed for a long time after the original fever episodes. The lack of complaints often delays appropriate medical care. However, children may show indirect signs of pain, such as limping, moving slowly, avoiding chewing/eating certain foods, and/or avoiding physical activities like sports. They may have stiffness, most notable upon awakening or with cold weather. This usually improves with movement or time. Families may report that children look worse on rainy or cold days and improve with warm weather or showers. Untreated active arthritis usually results in greater long-term deformity, disability, and morbidity in actively growing or developing children than in mature adult patients.

The metabolism of medications may be more rapid in pediatric patients when compared to adults. For example, it is common for children to need higher doses of non-steroidal anti-inflammatory drugs (NSAIDs) and methotrexate than those tolerated by adults.

JIA can be a lifelong disorder. About one third of these pediatric patients will continue to be affected by the disease in their adulthood. Children who have intermittent episodes of disease tend to do better in the long term than those with continuous disease. Even after years of remission, flares of disease can occur in adulthood. In future years, the prevalence of disease in adults may decrease because of continuous improvements in diagnosis and drug therapy.

Diagnosis of juvenile idiopathic arthritis

The diagnosis of JIA is based on a combination of medical history, clinical presentation, and radiologic and laboratory abnormalities. Patients with juvenile arthritis must have at least two of the following clinical features: joint swelling, pain/tenderness with motion, limitation of joint motion, and calor (warmth or heat) overlying the joint for more than 6 weeks. Appropriate and timely diagnosis and management are essential to relieve symptoms and to prevent, control, or limit the multi-organ sequelae of this disease. There are seven categories of JIA: (1) systemic onset, (2) oligoarthritis, (3) RF-positive polyarthritis, (4) RF-negative polyarthritis, (5) psoriatic arthritis, (6) enthesitis (inflammation at the tendon/ligament insertion into bone)-related arthritis, and (7) undifferentiated arthritis ( Table 1 ).

In systemic JIA, children present with fever, rash, poor weight gain/failure to thrive, decreased growth velocity, anemia, lymphadenopathy, hepatosplenomegaly, and serositis. Skin and nail changes (e.g., dactylitis) are commonly found in children with psoriatic arthritis. Uveitis is a reported complication of oligoarthritis, enthesitis-related arthritis, and less commonly polyarthritis and psoriatic arthritis.

Clinical findings vary and depend on the location, subtype of arthritis, and duration of disease. Children demonstrate a variety of findings, including a limp, morning stiffness, and/or antalgic gait, with difficulty walking or running. They may exhibit extremity overgrowth or undergrowth and asymmetries such as leg length discrepancies in the axial skeleton. Depending on the duration of symptoms, patients will have muscular atrophy around the affected joints and contractures.

If patients seek medical attention soon after the onset of symptoms, plain radiographs often do not show joint or skeletal abnormalities. Subtle radiographic findings that suggest arthritis in the early stages include mild peri-articular osteopenia with soft tissue swelling. In aggressive or rapidly progressive disease, or when there is a significant delay in the diagnosis and initiation of treatment, effusions, erosions, bony overgrowth, and joint space narrowing may be seen. Magnetic resonance imaging (MRI) with intravenous gadolinium is the gold standard imaging method to detect active synovial inflammation (enhancement of the intra-articular soft tissue). The extent of synovitis helps differentiate active inflammation and joint damage from long-standing chronic arthritis.

The detection of synovitis is a key aspect of JIA diagnosis and management and may prevent long-term disability. Although, synovitis can be diagnosed clinically by joint swelling, pain, tenderness, and decreased range of motion, gadolinium-enhanced MRI more accurately reflects synovitis than clinical examination. Active synovitis may be identified in presumed clinically inactive patients with this method.

Laboratory studies frequently provide additional information. For example, patients with significant inflammation may have anemia and/or elevated inflammatory markers such as erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). However, normal laboratory studies do not rule out active arthritis. For example, it is a common misconception that negative antinuclear antibody (ANA) and RF studies exclude the diagnosis of JIA. Of all the children with JIA, only 5–10% have the RF-positive polyarticular JIA subtype. On the other hand, RF-negative polyarticular JIA represents 15–25% of all JIA cases.

Arthritis in children is not simply adult rheumatoid arthritis in a child. There are significant differences in the diseases. Most children with JIA who receive appropriate treatment do not develop the typical ‘rheumatoid’ appearance, i.e., hand/limb/spine deformities. However, the longer the disease is active, the more the child is at risk of developing these deformities. Polyarthritis in combination with an elevated RF is also less common in children. Unlike adults, children may not complain of pain at first, and joint swelling may not be noticeable or can be delayed for a long time after the original fever episodes. The lack of complaints often delays appropriate medical care. However, children may show indirect signs of pain, such as limping, moving slowly, avoiding chewing/eating certain foods, and/or avoiding physical activities like sports. They may have stiffness, most notable upon awakening or with cold weather. This usually improves with movement or time. Families may report that children look worse on rainy or cold days and improve with warm weather or showers. Untreated active arthritis usually results in greater long-term deformity, disability, and morbidity in actively growing or developing children than in mature adult patients.

The metabolism of medications may be more rapid in pediatric patients when compared to adults. For example, it is common for children to need higher doses of non-steroidal anti-inflammatory drugs (NSAIDs) and methotrexate than those tolerated by adults.

JIA can be a lifelong disorder. About one third of these pediatric patients will continue to be affected by the disease in their adulthood. Children who have intermittent episodes of disease tend to do better in the long term than those with continuous disease. Even after years of remission, flares of disease can occur in adulthood. In future years, the prevalence of disease in adults may decrease because of continuous improvements in diagnosis and drug therapy.

Diagnosis of TMJ arthritis

Minimal complaints of pain, a lack of swelling or limitation of motion, and the poor sensitivity of conventional radiographs inhibit the diagnosis of TMJ JIA and active synovitis. As a result, the reported prevalence of TMJ disease varies based on inclusion criteria, disease duration, and imaging methods used. The incidence and prevalence of TMJ arthritis, therefore, may be underestimated.

The problem is compounded because the clinician must also differentiate between TMJ arthritis and myofascial pain dysfunction. Both can present with similar signs and symptoms. Too often, any type of jaw pain is assumed to be a result of TMJ arthritis, and myofascial pain dysfunction may not be considered. In addition, TMJ symptoms and signs may be masked by anti-rheumatic therapy.

The reported prevalence of myofascial pain dysfunction in healthy children ranges from 4% to 15%. However, orofacial pain occurs more frequently in children with JIA than in control subjects. Müller and colleagues did not find a statistically significant correlation between jaw pain and active TMJ arthritis. Twilt and colleagues reported that pain with jaw excursion was a predictor of TMJ involvement, while pain at rest was not a statistically significant predictor of disease. Weiss and colleagues determined that the positive predictive value of TMJ pain for synovitis was 100%, but the sensitivity was 26%. They concluded that TMJ pain was a poor screening tool for the detection of TMJ synovitis.

Our group recently reported an overlap in diagnoses of active arthritis and myofascial pain dysfunction in JIA patients with jaw pain. In addition, jaw symptoms were equally as likely to result from myofascial pain dysfunction as arthritis. A subjective finding of pain was not always predictive of TMJ synovitis. Therefore, differentiation between the two is critical, as the management strategies for TMJ arthritis/synovitis and myofascial pain dysfunction are different.

Variability in the nature and severity of physical signs inhibits the diagnosis of TMJ synovitis. Clinicians must be aware of subtle physical findings indicating early TMJ synovitis. We reported that among children with JIA, limited maximal incisal opening (MIO) and deviation of the jaw with opening had high sensitivity and specificity for synovitis (positive predictive value of 1.00). Therefore, it is important for clinicians to obtain MIO and lateral excursive measurements at the earliest time point in the disease process to establish a baseline for follow-up. These measurements should be made with a rigid equilateral triangle measuring device ( Fig. 1 ). Longitudinal open and closed mouth photographs demonstrating jaw opening and facial photographs demonstrating jaw growth over time and the presence of jaw asymmetry should be taken ( Fig. 2 ). The typical presentation of micrognathia (‘bird facies’) ( Fig. 3 ) is less common now, as disease awareness has increased and aggressive treatments are introduced early in the disease process.

Technique for the measurement of the maximal incisal opening with a rigid equilateral triangle measuring device. (A) The patient opens their mouth maximally and the triangle is inserted in between the incisors. Since the device is an equilateral triangle, it is rotated until the numbers on the upper and lower sides are equal, and that is the MIO. (B) Rigid equilateral triangle measuring device by KLS Martin LP, Jacksonville, FL, USA.

A 12-year-old girl with JIA and unilateral TMJ involvement. (A) Frontal view at rest demonstrating the typical physical findings: facial asymmetry with deviation of the chin to the right, flattening of the right mandibular contour while the left one looks fuller, and shortening of the right posterior facial height. (B) Frontal view smiling illustrating the occlusal cant and deviation of the chin to the right. (C) Submental view demonstrating an accentuated right antegonial notch, chin deviation to the right, and a contour deficiency of the right mandible and soft tissue. (D) Open-mouth panoramic radiograph showing the right accentuated antegonial notch, right condylar head flattening, short right condylar neck, and shorter right ramus–condyle unit compared to the left side.

A 12-year-old boy with JIA and bilateral TMJ involvement. (A) Frontal view demonstrating the symmetric mandibular contour, centered chin point, no occlusal cant, and short posterior facial height. (B) Submental view illustrating the bilateral accentuated antegonial notch. (C) Profile view demonstrating the typical ‘bird facies’: micrognathia, short chin-to-throat distance, convex profile, and high mandibular plane. (D) Lateral cephalogram demonstrating significant mandibular retrusion with an increased overjet and open bite. (E) Open-mouth panoramic radiograph showing bilateral condylar head flattening, short condylar necks, and short ramus–condyle units.

Multiple radiographic techniques for imaging the TMJs in JIA patients have been evaluated. MRI with gadolinium has been shown to be the best method to detect early inflammation and is the gold standard for the diagnosis of TMJ synovitis ( Figs 4 and 5 ).

MRI of the patient in Fig. 2 . (A) Post contrast T1 coronal cut of the right condyle demonstrating typical findings of active disease: flattening of the condylar head and synovial enhancement (arrow). (B) Post contrast T1 coronal cut of the left condyle representing a normal condyle. Note the absence of synovial enhancement and the larger condylar head compared to the right side.

MRI of the patient in Fig. 3 . Post contrast T1 coronal cut of the bilateral condyles demonstrating typical findings of synovitis.

In an attempt to minimize the potential high costs associated with MRI, other radiographic modalities have been studied. Skeletal changes resulting from prior inflammation cannot be distinguished from current active disease by computed tomography (CT) scans, but heterotopic ossification of the TMJs can be identified and can be associated with severe arthritis. However, based on recent evidence of an increase in childhood malignancies associated with high diagnostic radiation dosages, CT scanning is not encouraged. In certain cases where MRI with contrast is not possible (i.e., kidney disease), another alternative to MRI may be TMJ ultrasound. Weiss and colleagues found that ultrasound detected TMJ involvement in only 28% of patients. Jank and colleagues used ultrasound to visualize disk dislocation, condylar erosion, and effusion. However, they were not able to visualize active synovial hypertrophy and bone marrow edema, which are often indicative of early disease activity. In a recent study using ultrasound, Assaf and colleagues were able to detect condylar erosions and abnormal synovial thickness in a majority of their patients. However, they detected bone irregularities and/or joint effusion in less than 25% of patients. Our recently reported that abnormal condyle morphology and accentuated antegonial notching, as demonstrated on panoramic radiographs, are significantly correlated with TMJ synovitis. Therefore, in situations where MRI scanning is not possible or contraindicated (e.g., kidney disease), these panoramic radiographic findings can be used to support the diagnosis of synovitis.

A condition that may mimic JIA and confuse the diagnosis for OMFS is ICR. This condition is uncommon but well known and is defined as a progressive alteration of condylar morphology and loss of condylar mass of unknown etiology. These patients usually present with a chief complaint of progressive loss of chin prominence. They demonstrate progressive clockwise rotation of the mandible, retrognathia, and an open bite. The patients are overwhelmingly female, between ages 15 and 44 years, and the disorder is usually bilateral and symmetrical. Their work-up for systemic rheumatoid disease is negative. They do not demonstrate synovitis on MRI scans. The disease is self-limited and usually burns itself out over time. The entire condyle may be resorbed down to the sigmoid notch, but the disk and synovium are not affected.

Although the exact etiology and pathogenesis of ICR remain unclear, three theories have been proposed. Arnett and colleagues correlated ICR with increased, abnormal joint loading and subsequent pressure resorption. Chuong and Piper postulated that the mechanism of condylar resorption is similar to that of avascular necrosis of the femoral head. Gunson et al., in a recent paper, described an association between 17β-estradiol deficiency and ICR in a series of young females. This subject is discussed in more detail in the Discussion and controversies section below.

Since ICR is not a systemic illness, the malocclusion and jaw deformity can be treated successfully with an appropriate orthognathic surgical procedure when it has burned itself out. During the active phase of the disease, condylectomy is curative, permitting ramus–condyle unit and jaw reconstruction as appropriate.

Management

In this article, only the management of TMJ involvement in JIA is discussed. The management strategies for myofascial pain dysfunction and ICR in children have been described elsewhere in the literature and are not the subject of this review. The treatment of JIA is individualized and depends on the severity of disease, current activity, number of joints involved, and current physical limitations. Attempts to develop guidelines to treat JIA are underway but are currently based on expert opinion and consensus rather than multicenter randomized clinical trials.

The basic categories of disease are multiple joint involvements (polyarticular JIA) with or without systemic involvement, or one affected joint. When multiple joints are affected, systemic therapy is carried out in a hierarchical fashion.