Version July 2025
INTRODUCTION —
Juvenile idiopathic arthritis (JIA) is the most common rheumatic disease affecting children and includes multiple disease subtypes (ie, systemic JIA [sJIA], oligoarticular JIA, polyarticular JIA, psoriatic JIA [psJIA], and enthesitis related arthritis [ERA]). This topic discusses common challenges in JIA, including proper administration of vaccinations and certain complications such as temporomandibular joint (TMJ) arthritis and uveitis. Other aspects of the management and prognosis of specific JIA subtypes are reviewed separately:
●(See "Systemic juvenile idiopathic arthritis: Complications".)
●(See "Systemic juvenile idiopathic arthritis: Treatment and prognosis".)
●(See "Oligoarticular juvenile idiopathic arthritis".)
●(See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis".)
●(See "Psoriatic juvenile idiopathic arthritis: Management and prognosis".)
●(See "Spondyloarthritis in children".)
The classification, epidemiology, pathogenesis, and complications of JIA are also discussed in detail elsewhere. (See "Classification of juvenile idiopathic arthritis" and "Juvenile idiopathic arthritis: Epidemiology and immunopathogenesis".)
IMMUNIZATIONS
General approach — We assess patients' vaccination status and indications/contraindications for additional vaccines at least annually. The approach to immunization in patients with juvenile idiopathic arthritis (JIA) depends upon whether patients are taking immunosuppressive therapy:
●Patients not taking immunosuppression – In such patients, we maintain a routine childhood immunization schedule. (See 'Patients not taking immunosuppression' below.)
●Patients taking immunosuppression – In such patients, we modify the routine childhood immunization schedule by deferring live vaccines, administering additional doses of some vaccines (eg, the pneumococcal vaccine), and, for patients taking certain potent immunosuppressive medications, adjusting the timing of non-live vaccines [1]. (See 'Patients taking immunosuppression' below.)
The potential risks of routine vaccination in JIA are relatively minor and are thought to be outweighed by the benefits, as follows:
●Benefits of vaccination – The routine administration of childhood vaccinations has dramatically decreased the incidence of vaccine-preventable illnesses (see "Standard immunizations for children and adolescents: Overview", section on 'Benefits of vaccines'). Multiple studies of patients with rheumatic diseases suggest that vaccinations can still generate a protective benefit [2], even when patients require immunosuppressive therapy [3-5].
Prophylactic vaccinations are especially important for patients with JIA, who are often more vulnerable to infection. As an example, a study found that the rate of bacterial infection requiring hospitalization was higher in a cohort of 8400 patients with JIA compared with a group of patients with attention-deficit hyperactivity disorder (ADHD; adjusted hazard ratio [aHR] 2.0, 95% CI 1.5-2.5) [6]. However, patients with JIA have a relatively low completion rate of on-schedule vaccinations and administration of the measles, mumps, and rubella (MMR) vaccine in particular [7], underscoring the importance of continued efforts to improve vaccine uptake.
●Risks of vaccination – Data are mixed about whether there is an increased risk of flares of JIA after vaccinations, and the risk likely varies based on the specific vaccine [5,8]. In addition, very limited data have questioned whether rubella infection or vaccination may be related to the development of JIA (see "Juvenile idiopathic arthritis: Epidemiology and immunopathogenesis", section on 'Infection' and "Viral arthritis: Causes and approach to evaluation and management", section on 'Rubella and rubella vaccine virus'). However, subsequent studies have suggested that the MMR and MMR/varicella (MMR/V) booster vaccines are generally safe in patients with JIA [9,10], and we continue to administer rubella vaccination in all patients with JIA who are otherwise eligible to receive it (ie, patients without contraindications to live vaccinations, such as ongoing use of immunosuppressive therapy). (See 'Avoidance of live vaccines' below.)
Our approach is generally consistent with the guidelines for immunization in JIA issued by the American College of Rheumatology (ACR) in 2021 [1], as well as recommendations for children with autoimmune inflammatory rheumatic diseases from the European Alliance of Associations for Rheumatology (EULAR) and the Pediatric Rheumatology European Society (PRES) in 2021 [11]. Further studies are needed in this area, particularly regarding the importance of the subtype of JIA, primary immunization with live-virus vaccines in non-immunosuppressed patients with inactive disease, and live vaccine boosters in patients who have moderate to high disease activity and/or who are on biologic disease-modifying antirheumatic drugs (bDMARDs).
Additional information on the importance of routine childhood vaccinations, caregiver hesitancy or refusal of vaccines, and immunization in adults with rheumatic diseases is provided separately:
●(See "Standard immunizations for children and adolescents: Overview", section on 'Benefits of vaccines'.)
●(See "Standard childhood vaccines: Caregiver hesitancy or refusal".)
●(See "Immunizations in autoimmune inflammatory rheumatic disease in adults".)
Patients not taking immunosuppression — For children with JIA who are not taking immunosuppression, we maintain a routine immunization schedule regardless of disease activity, including administration of the MMR vaccine and other live vaccinations [12]. (See "Standard immunizations for children and adolescents: Overview", section on 'Routine schedule'.)
Patients taking immunosuppression
Avoidance of live vaccines — Live vaccines (table 1) should not be given to children who are receiving systemic immunosuppression. Live vaccines commonly included in routine childhood vaccination schedules include MMR, varicella zoster virus (VZV), MMR/varicella (MMR/V), oral rotavirus, and live attenuated (intranasal) influenza vaccines. Of note, immunosuppressed patients may still receive the recombinant influenza vaccine, which is not a live vaccine. Other live immunizations that are sometimes used for patients who live in or are traveling to certain regions include the oral typhoid and yellow fever vaccinations.
Limited data suggest that the administration of some live vaccinations may be safe for patients taking less potent immunosuppressive medications; however, further studies are needed before recommendations are likely to change. Studies on specific vaccinations include the following:
●MMR and MMR/V – Data from a small, open-label trial and observational studies suggest that a booster MMR vaccine is safe and immunogenic in patients with JIA on methotrexate. Specifically, several observational studies of MMR vaccination in patients with JIA did not identify any systemic MMR infections or increases in JIA disease activity or medication use, including among patients taking methotrexate with or without various bDMARDs [10,13,14]. These findings were confirmed in a randomized, open-label trial of 137 patients (131 analyzed) with JIA who were randomly assigned to receive a booster MMR vaccine or no vaccination [9]; 15 patients were taking bDMARDs (discontinued prior to vaccination) and 60 patients were on methotrexate. After 12 months, JIA disease activity was similar and seroprotection rates were higher in patients who received the booster compared with those who did not. Of note, most patients had low disease activity at baseline and oligoarticular JIA was the most common subtype (65 percent of cases). (See "Juvenile idiopathic arthritis: Epidemiology and immunopathogenesis", section on 'Environmental factors' and "Viral arthritis: Causes and approach to evaluation and management", section on 'Rubella and rubella vaccine virus' and "Secondary immunodeficiency induced by biologic therapies".)
●VZV – Limited observational data suggest that primary vaccination with the VZV vaccine may be safe and immunogenic in children with JIA. One case-control study of 49 children with rheumatic disease (39 with JIA) taking methotrexate with or without other immunosuppressive medications (glucocorticoids or bDMARDs) found similar antibody responses in patients and controls, and no disease flares [15]. Adverse reactions were infrequent and minor, with fever in one patient, and mild temperature elevation (<38°C) and mild vesicular rash in three patients.
Adjustment of non-live vaccines — While it is generally safe to administer non-live vaccinations to patients with JIA who are taking immunosuppressive medications, these therapies may keep patients from mounting a protective response to the vaccine. The degree of vaccine response depends on both the type of immunosuppression and the type of vaccination. This variability has prompted the development of guidelines for adults with rheumatic diseases on how to optimally time non-live vaccinations with immunosuppression [16]. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults", section on 'Vaccine efficacy and immunogenicity'.)
While specific recommendations for children taking immunosuppression have not been released, we consider the optimal timing of vaccinations when patients are taking the following medications:
●Methotrexate – In adults, patients taking methotrexate generally have a higher rate of seroprotection to vaccination when they hold methotrexate. However, it is unclear whether this translates into clinical differences in the number or severity of infections, and patients can develop a disease flare while holding therapy. For children with JIA taking methotrexate, we counsel the patient and caretaker(s) about the potential risks and benefits of holding methotrexate for one to two weeks after vaccination to maximize vaccine immunogenicity.
●Rituximab – The humoral response to vaccination is impaired in patients taking rituximab. Accordingly, multiple studies have demonstrated that adults taking rituximab are less likely to develop an antibody response to vaccinations. However, the cell-mediated immune response can still provide some degree of protection. While data in children are limited, we generally administer the influenza, COVID, and pneumococcal vaccines. The remainder of vaccinations are generally not given until at least six months after the last rituximab dose. Data describing the vaccine response in adults receiving rituximab are discussed in detail elsewhere. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults", section on 'Vaccine efficacy and immunogenicity'.)
●High doses of glucocorticoids – The response to certain vaccinations may be reduced in patients taking higher doses of glucocorticoids; specifically, the Advisory Committee on Immunization Practices (ACIP) has defined the threshold for reduced vaccination response as ≥14 days of ≥2 mg/kg or ≥20 mg/day of prednisone or its equivalent (in patients >10 kg) [17]. Such doses of glucocorticoids are rarely required for patients with JIA. However, in such cases, we generally defer routine vaccinations other than the influenza vaccine until patients taper to a lower dose of glucocorticoids. (See "Glucocorticoid effects on the immune system", section on 'Impact on vaccination'.)
If patients are not taking the above therapies, then we generally maintain the routine immunization schedule of non-live vaccinations [12]. (See "Standard immunizations for children and adolescents: Overview", section on 'Routine schedule'.)
In addition, children with JIA who are taking immunosuppression are more vulnerable to infection and may therefore benefit from additional doses of certain non-live vaccinations, potentially including the following:
●Pneumococcal vaccination – Children taking certain forms of immunosuppression, including those taking bDMARDs, may benefit from additional pneumococcal vaccinations beyond those standardly given as part of routine childhood immunizations. (See "Pneumococcal vaccination in children", section on 'Immunization of high-risk children and adolescents'.)
●Coronavirus disease 2019 (COVID-19) vaccination – Immunocompromised patients may require additional doses of the COVID-19 vaccine. (See "COVID-19: Vaccines", section on 'Immunocompromised individuals'.)
●Herpes zoster vaccination – Immunocompromised patients ≥19 years old may be eligible for vaccination against herpes zoster with the non-live recombinant glycoprotein E vaccine (designated recombinant zoster vaccine [RZV]), even if they have previously received a varicella vaccination. (See "Vaccination for the prevention of shingles (herpes zoster) in adults", section on 'Immunocompromised adults'.)
Vaccination of household contacts — We generally encourage routine vaccination of all household contacts of patients with JIA [11]. However, when patients with JIA are taking immunosuppression, the following types of vaccinations in close contacts may require special precautions:
●Rotavirus vaccination – If an infant who has close contact with an immunocompromised patient receives a live rotavirus vaccination, all of the infant's caregivers should carefully wash their hands after changing diapers [11].
●Varicella vaccination – If a close contact receives a live varicella vaccine and subsequently develops a rash, then they should avoid direct physical contact with the immunocompromised person until their symptoms resolve [11].
●Smallpox vaccination – The replication-competent smallpox vaccine (ACAM2000) should not be given to close contacts of immunocompromised patients. Another non-replicating version of the smallpox vaccine is available. (See "Vaccines to prevent smallpox, mpox (monkeypox), and other orthopoxviruses", section on 'Contraindications and precautions'.)
COMPLICATIONS —
The most common complications associated with juvenile idiopathic arthritis (JIA) are temporomandibular joint (TMJ) arthritis, uveitis, leg-length discrepancy, and short stature. The incidence of AA amyloidosis in patients with JIA is extremely low.
The topics on the treatment of specific JIA subtypes provide additional information on the approach to monitoring disease activity and prognosis. Complications associated with systemic JIA (sJIA), such as macrophage activation syndrome (MAS) and sJIA-associated lung disease (sJIA-LD), are also discussed in detail elsewhere. (See "Systemic juvenile idiopathic arthritis: Complications".)
Temporomandibular joint arthritis — TMJ arthritis frequently affects patients with JIA and may initially be relatively asymptomatic [18]. It can lead to permanent changes in facial appearance, including jaw asymmetry and micrognathia; it is therefore important to regularly monitor for TMJ involvement and, if present, escalate therapy until achieving remission [19].
Epidemiology — Normal TMJ anatomy is shown in the picture (figure 1). TMJ arthritis is concerning in children who are actively growing since the mandibular growth plate is close to the joint fibrocartilage. Damage to this growth plate from uncontrolled inflammation may lead to jaw undergrowth or micrognathia [20], which can occur even in the absence of symptoms.
Based on observational studies of TMJ imaging in patients with JIA, TMJ arthritis is relatively common [21-23]. In one retrospective study of 7473 patients with JIA, 939 patients (12.6 percent) had a diagnosis of TMJ arthritis made clinically or by magnetic resonance imaging (MRI) [22]. However, this is likely an underrepresentation of the true prevalence because a minority of children were assessed via physical examination. In another cross-sectional study of over 3300 children, TMJ involvement was noted in 11.6 percent of patients based on presence of TMJ pain or limited maximal incisal opening [21]. In that study, the factors most strongly associated with the development of symptomatic TMJ arthritis included cervical spine involvement, prolonged disease duration (>4.4 years), and higher levels of disability.
Clinical presentation and evaluation — The clinical presentation of TMJ arthritis in patients with JIA is described below:
●Symptoms – While patients often are asymptomatic at presentation, they may describe pain in their jaw or inner ear, especially with activity (eg, talking, chewing), as well as morning stiffness and difficulty fully opening their mouth [18,24]. In one study of 32 patients with a new diagnosis of JIA who underwent MRI of the TMJ, focal pain and dysfunction were 100 percent specific but only 26 percent sensitive for TMJ arthritis [25].
●Physical findings – The evolution of TMJ involvement on physical examination is often slow and virtually imperceptible [25]. Focal swelling of the TMJ is rarely present and cannot be appreciated on physical examination [18,24]. Clinical findings that are suggestive of TMJ involvement include crepitus, pain with TMJ motion, limited maximal incisal opening, and limited translation (ie, reduced forward movement of the mandibular condyle with mouth opening) [26]. Of these signs, pain and limited translation are predictors of radiographic damage [26]. With longstanding disease, patients can also develop clinically apparent jaw asymmetry (ie, if there is unilateral disease, the chin may point to the affected side) [27], micrognathia (ie, mandibular hypoplasia), and/or retrognathia (ie, malocclusion from the mandible being set farther back relative to the maxilla) [18].
●Laboratory testing – There are not any known laboratory findings that are clearly associated with a higher risk of developing TMJ arthritis or ongoing TMJ inflammation.
●Imaging findings – The most common imaging findings on MRI observed in TMJ arthritis are synovial membrane enhancement, condylar erosions, and abnormally shaped condyles, including condylar flattening [28]. It is important to correlate imaging findings with the patient's symptoms and examination, since contrast enhancement on MRI may also be seen in healthy children [29,30]. Other potential changes in TMJ arthritis include articular disc alterations (eg, deformities, effusion), subchondral sclerosis, pannus formation, and increased fluid and/or enhancement in the intraarticular space [28]. Synovial thickening may be observed as an early finding [31].
We monitor patients with JIA for the development of TMJ arthritis based on history and physical examination [32]. We routinely ask about difficulty opening their mouth and orofacial pain, especially when opening the mouth or chewing. On examination, we watch for the development of jaw asymmetry, abnormal position of the mandible in the sagittal position, TMJ crepitus or other abnormal condylar movements during jaw opening, tenderness of the TMJ or masticatory muscles, and limited maximal incisal opening (ie, the ability to open <3 finger-breadths wide).
We obtain imaging in patients who have concerning examination findings or new complaints of jaw pain or dysfunction. The optimal imaging modality is MRI with contrast [33]. Disc characteristics are helpful in distinguishing inflammatory arthritis from other causes of TMJ pain and dysfunction; as an example, TMJ arthritis in JIA is positively associated with disc flattening and negatively associated with anterior displacement [34]. While radiography and computed tomography (CT) can detect structural changes, they are not able to distinguish acute synovitis from chronic damage [24]. Compared with MRI, ultrasonography appears to be less sensitive for the diagnosis of TMJ arthritis [35].
Diagnosis and differential diagnosis — Symptoms of jaw pain, stiffness, and/or dysfunction are relatively common in patients with JIA [36]; while they may be related to either active TMJ arthritis or inactive, chronic TMJ arthritis, they are not specific enough in isolation to confirm a diagnosis. Similarly, relying only on findings on physical examination can lead to both over- and underdiagnosis of TMJ arthritis [37,38]. If TMJ arthritis is suspected and its presence will impact treatment decisions, then imaging to confirm the diagnosis is paramount. However, if the presence of TMJ arthritis would not alter the treatment approach (eg, in a patient who is escalating disease-modifying antirheumatic drug [DMARD] therapy for other areas of active arthritis), then it may be reasonable to make a clinical diagnosis of TMJ arthritis without imaging confirmation, especially if imaging is difficult to obtain (eg, in a younger child who would require sedation). In such cases, we monitor patients closely for worsening symptoms or evidence of structural damage that may warrant reconsideration of imaging.
Important alternative diagnoses to consider that can cause orofacial pain and dysfunction in children include the following:
●TMJ disorders – TMJ disorders can be classified as a type of secondary headache disorder and are a common cause of jaw pain and dysfunction in older teenagers and adults. They are associated with fibromyalgia, which can accompany various forms of chronic inflammatory arthritis. TMJ disorders are rare in children under age 10. They should not be associated with jaw asymmetry or micrognathia. (See "Temporomandibular disorders in adults".)
●Anterior disc displacement – Anterior displacement of the TMJ can also cause dysfunction of the mandibular condyle and is associated with evidence of inflammation on MRI. In one study of clinical TMJ degeneration in 18 patients with JIA and 16 without JIA, both patient groups had bone marrow edema (66 and 53 percent, respectively) and synovial thickening (34 and 25 percent, respectively) [34]. Disc flattening and anterior disc displacement on MRI may help distinguish these disorders, with disc flattening being more common in patients with JIA versus those without JIA (39 versus 9 percent, respectively) and anterior displacement being less common (0 percent versus 28 percent, respectively).
●Other causes of condylar resorption – Condylar resorption can be related to osteoarthritis, trauma, and, rarely, systemic rheumatic diseases other than JIA (eg, scleroderma, systemic lupus erythematosus [SLE]) [39]. Idiopathic condylar resorption is especially common in female adolescents who are experiencing a growth spurt [40,41]. Condylar resorption can often be differentiated from TMJ arthritis in JIA based on history and the presence of other clinical features (eg, inflammatory arthritis in other joints); however MRI of the TMJ may be helpful in cases of diagnostic uncertainty.
●Chronic noninfectious osteomyelitis (CNO) – CNO (also known as chronic recurrent multifocal osteomyelitis [CRMO]) can affect the mandible and may be associated with inflammatory arthritis in other joints. It can be distinguished from TMJ arthritis based on imaging. (See "Chronic nonbacterial osteomyelitis (CNO)/chronic recurrent multifocal osteomyelitis (CRMO) in children".)
Treatment and prognosis — TMJ involvement is treated similarly to other forms of peripheral arthritis in JIA and includes the following elements:
●Multidisciplinary care involving rheumatology, orthodontics, oral and maxillofacial surgery, and physical therapy
●Systemic immunosuppression (ie, DMARD therapy)
●As-needed analgesics (eg, nonsteroidal antiinflammatory drugs [NSAIDs])
●For those with persistent symptoms, consideration of locally delivered therapy (eg, intraarticular lavage with or without glucocorticoids)
●For those who have or at high risk of developing dentofacial deformities, consideration of functional orthodontic appliances, which can hopefully avoid the need for future corrective surgery
●Education for the patient and family about potential exacerbating factors (eg, chewing gum)
Our approach to the initial treatment of TMJ arthritis is outlined in the algorithm (algorithm 1). For patients starting therapy for active TMJ arthritis, we add or switch DMARD therapy to include methotrexate and/or a biologic DMARD (bDMARD). The exact choice and combination of DMARD is guided by baseline therapies for JIA and other active clinical manifestations (eg, peripheral versus axial arthritis, enthesitis, uveitis). As examples, a patient with newly diagnosed JIA involving the TMJ and peripheral joints may be started on methotrexate monotherapy, while a patient who develops TMJ arthritis while on methotrexate for ≥3 months may add or switch to a bDMARD. Some providers reserve methotrexate for mild cases (ie, minimal to no symptoms or evidence of jaw abnormalities on physical examination) and otherwise escalate directly to tumor necrosis factor (TNF) inhibitors, given the importance of rapidly controlling inflammation and preserving function in this joint.
If patients have persistent symptoms after an adequate trial of systemic immunosuppression, we reconsider the diagnosis and, if not done recently (eg, within three to six months), consider repeating an MRI. Depending on the degree of response and other active JIA manifestations, we add or switch immunosuppressive agents; there are no data to favor the use of one bDMARD over another. In addition, we refer patients with persistent disease for TMJ arthrocentesis and lavage, which is typically performed by interventional radiologists or oral and maxillofacial surgeons.
Skeletally mature patients with refractory disease despite the above measures may benefit from the addition of intraarticular glucocorticoid injections; however, we carefully review the potential risks of this procedure, including heterotopic bone formation [42]. If significant micrognathia develops, surgical correction may be possible once the facial bones are fully developed.
Our approach differs somewhat from the treatment guidelines for TMJ arthritis in JIA issued by the American College of Rheumatology (ACR) in 2021 [19], as well as recommendations from an interdisciplinary panel of experts published in 2023 [33]. Specifically, we typically do not use NSAIDs as monotherapy for TMJ arthritis since TMJ involvement in JIA is a poor prognostic factor and typically prompts initiation of DMARD therapy (see "Oligoarticular juvenile idiopathic arthritis", section on 'Overview of our approach'). In addition, we avoid intraarticular glucocorticoid injections in skeletally immature patients and attempt TMJ arthrocentesis and lavage prior to an intraarticular glucocorticoid injection, given the potential increased risks associated with glucocorticoids and unclear benefit beyond lavage alone.
Data directly comparing various therapeutic approaches to treat TMJ arthritis in JIA are extremely limited. One study randomly assigned 38 affected TMJs in 21 patients with JIA to undergo arthrocentesis alone or arthrocentesis with intraarticular triamcinolone hexacetonide injection [43]. Compared with baseline scores, pain and function improved at three- and eight-month follow-up assessments; the degree of improvement was similar between the two treatment groups. By contrast, a study of 41 children with JIA observed outcomes in 21 patients who had lavage with glucocorticoid injection, 8 with lavage alone, and 12 without any local intervention [44]; based on follow-up clinical assessments and MRI after six months, improvement from baseline was similar across the three treatment groups, with the exception of greater improvement in dysfunction index scores in the glucocorticoid and lavage group compared with the lavage group.
Additional supportive evidence for specific types of therapies is summarized below:
●DMARDs – While data on DMARDs to treat TMJ arthritis in JIA are very limited, observational data suggests that they are effective. As an example, in a longitudinal study of 38 patients with TMJ involvement due to JIA who did not receive IA TMJ glucocorticoid injections and who were followed for a median of 3.6 years, DMARD therapy was associated with improvement of MRI findings and less severe osseous deformities [45]. Similarly, a prospective study of 18 patients with TMJ arthritis related to JIA who took a combination of a conventional synthetic DMARD (csDMARD) and a TNF inhibitor had improved symptoms and MRI findings over two years of observation [46]. This latter study supports anecdotal observations that the frequency of clinically apparent jaw abnormalities in patients with JIA has also decreased over time with advances in DMARD therapy [24].
●Intraarticular glucocorticoid injections – Support for the use of intraarticular glucocorticoid injections in treating TMJ arthritis in JIA comes from systematic reviews [47,48] and subsequent studies [49]. As an example, a systematic review in 2020 identified 11 studies involving intraarticular glucocorticoid injections for TMJ arthritis; it found improvement in both pain resolution and increase in mouth-opening capacity for many patients, but also a high degree of heterogeneity and lack of a control group in many studies. In another study, 15 adolescent patients with JIA and active TMJ arthritis on MRI received a total of 22 intraarticular glucocorticoid injections, which were mostly given as a single injection per joint (22 cases) [49]. After two months, there were minimal clinical improvements in pain and maximal incisal opening. Over two years of follow-up, joint damage detected on MRI remained stable in 10 patients, improved in 3, and worsened in 2, which was in the setting of 10 patients (67 percent) also adjusting DMARD therapy over this timeframe.
Potential complications of intraarticular glucocorticoids include heterotopic bone formation that can further impair jaw function [42,50]. As an example, in a retrospective study of 238 children with TMJ arthritis related to JIA who received ≥1 IA glucocorticoid injection(s), 14 percent developed heterotopic bone formation [42]; this complication was more common in patients with an older age at diagnosis of JIA and a shorter time from diagnosis to first injection.
Patients with JIA who develop TMJ involvement appear to have a relatively worse prognosis, with poorer health-related quality of life and higher levels of disease activity and disability [21]. However, condylar alterations and facial asymmetry can improve over time in some children with treatment [45,51].
Chronic anterior uveitis — Chronic anterior uveitis (CAU) is one of the most potentially serious complications of JIA. CAU in patients with JIA typically lacks symptoms associated with active inflammation (light sensitivity, ocular injection, eye pain), both at the onset of CAU and often throughout the disease course. However, patients with refractory disease can develop severe vision impairment. Patients with JIA must therefore have regular monitoring with an ophthalmologist, including a complete slit lamp examination.
Epidemiology and risk factors — CAU or iridocyclitis refers to inflammation of the anterior uveal tract and the adjacent ciliary body (figure 2). This is the most common extraarticular complication in JIA, affecting approximately 10 to 20 percent of all patients [52-54].
The risk of developing CAU related to JIA varies by the subtype of JIA, with higher rates among those who have oligoarticular JIA, rheumatoid factor (RF) negative polyarticular JIA, psoriatic JIA (psJIA), and undifferentiated JIA [52-54]. The risk is also higher among female patients, those who are diagnosed at a young age (<6 years), and those with a positive antinuclear antibody (ANA) test [52-55].
The time from diagnosis of JIA to the diagnosis of related CAU is variable. In a single-center cohort study of over 1000 children with JIA, the mean time between diagnosis of both conditions was 1.8 years, with a wide range of -4.2 months (signifying onset of uveitis prior to JIA) to 10 years [56].
Clinical presentation and routine monitoring — CAU is often initially silent and can affect one or both eyes. Uncommonly, patients may present with eye pain, redness, photophobia, vision loss, and/or headache. Patients with longstanding disease may present with visual impairment due to complications such as posterior synechiae (ie, attachment of the iris to the lens) (picture 1), cataracts, glaucoma, band keratopathy, and macular edema. It is important to note that young children, who have a higher risk of developing uveitis, may have difficulty expressing vision impairment. (See "Uveitis: Etiology, clinical manifestations, and diagnosis", section on 'Symptoms and findings'.)
Most patients with CAU have an unremarkable external eye examination, without appreciable scleral injection or photophobia. A standard ophthalmology examination may reveal cells in the anterior chamber on slit lamp examination, as well as other complications (eg, impaired visual acuity, increased intraocular pressure, macular edema, posterior synechiae, corneal calcium deposition consistent with band keratopathy).
We refer all patients with JIA to an ophthalmologist for regular monitoring for CAU, which includes performing a complete slit lamp examination [57]. Optometric examination or office fundoscopy are not adequate substitutes. The optimal approach to monitoring for CAU in patients with JIA varies depending on the patient's risk factors, as is summarized in the algorithm (algorithm 2) and described below:
●Assessing the risk of uveitis – We determine the risk of uveitis based on the subtype of JIA and, for some JIA subtypes, ANA positivity, age at JIA onset, and disease duration:
•Patients with RF-positive polyarticular JIA, enthesitis related arthritis (ERA), and sJIA – Patients are at low risk.
•Patients with RF-negative polyarticular JIA, oligoarticular JIA, psJIA, and undifferentiated JIA – These patients are at high risk if they meet the following criteria:
-ANA positive and
-Age of JIA onset ≤6 years old and
-Disease duration ≤4 years
Otherwise, patients in this group are at moderate risk.
●Initiating monitoring based on risk – We initially monitor high-risk patients every three months (with the initial evaluation optimally within four weeks of JIA diagnosis), moderate-risk patients every six months, and low-risk patients every 12 months.
●Decreasing frequency of subsequent monitoring over time – Patients at moderate to high risk who do not develop CAU during their initial monitoring period may eventually fall into a lower-risk group that requires less frequent evaluation. Specifically, we decrease the frequency of monitoring as follows:
•Initially high-risk patients – We decrease the frequency of monitoring to every six months in high-risk patients who have not developed uveitis after four years. After seven years from disease onset, the frequency may be further reduced to every 12 months.
•Initially moderate-risk patients – We decrease the frequency of monitoring to every 12 months in moderate-risk patients who have not developed uveitis over a specific time period. For patients who were ≤6 years old at JIA onset, we wait seven years; otherwise, we wait four years.
Few data exist to guide whether or when it might be safe to discontinue monitoring for adult patients with JIA. While practices vary, we typically ask patients to continue visiting ophthalmology for an annual examination indefinitely. We also counsel older patients about symptoms that should prompt screening, such as vision loss, visual field deficits, or floaters.
The rationale for frequent monitoring for CAU is to minimize the risk of ocular complications due to delayed diagnosis and treatment, especially as these patients are asymptomatic. Our approach is consistent with the guidelines issued in 2019 by the ACR/Arthritis Foundation (AF) [57]. These guidelines simplified the approach initially outlined by the American Academy of Pediatrics in 2006 [55], which was subsequently modified in 2007 to fit modern JIA classification criteria [53].
Treatment — We typically treat patients with anterior uveitis related to JIA with a course of topical glucocorticoids and, if needed, escalate to subcutaneous methotrexate and often TNF inhibitor therapy.
Initial therapy based on disease severity — Initial therapy of CAU in patients with JIA depends on the severity of their ocular disease, as well as baseline therapies for other manifestations of JIA, as summarized in the algorithm (algorithm 3) and described in more detail below:
●Topical glucocorticoids for all patients – For all patients with JIA who develop related CAU, topical glucocorticoids are first-line therapy and are typically managed by ophthalmologists, who have the equipment and expertise to monitor treatment response and complications. These may be added to any existing therapies for JIA.
●Concurrent systemic immunosuppression for severe disease – In addition to topical glucocorticoids, we add or adjust DMARD therapy for patients with the following features suggestive of severe disease:
•Profound visual impairment or features that suggest a risk of significant vision loss (eg, macular edema)
•Factors associated with a poor prognosis, including band keratopathy, cataracts, posterior synechiae, and/or glaucoma
If patients develop these features despite an adequate trial of DMARD therapy for other JIA manifestations, they may need to switch the DMARD (eg, to another TNF inhibitor) or add another DMARD (eg, adding a TNF inhibitor to methotrexate). If patients are taking oral methotrexate, we change to subcutaneous administration to maximize absorption. For severe inflammation, it may be preferable to start infliximab over other types of TNF inhibitors since it is easier to increase the medication dose and/or frequency of infusions.
Otherwise, if patients develop features suggestive of severe CAU in the absence of systemic immunosuppression, then we start DMARD therapy. While the exact therapy depends on other extraocular manifestations of JIA, we start with subcutaneous methotrexate for most patients. (See 'Persistent or refractory disease' below.)
Finally, in cases of severe CAU, we also consider adding a short-term course of systemic glucocorticoids as bridging therapy while the efficacy of DMARD therapy is still building [57,58]. If we use systemic glucocorticoids, we taper and discontinue them as quickly as possible given their potential to cause significant adverse effects. (See "Major adverse effects of systemic glucocorticoids".)
●Adjunctive dilating drops – Patients may also benefit from the addition of dilating (ie, cycloplegic) drops to prevent the development of synechiae [59], as is done for adults with various forms of anterior uveitis. (See "Uveitis: Treatment", section on 'Dilating drops for anterior uveitis'.)
There are very few data to directly compare different approaches to the initial treatment of active CAU. Our approach is generally consistent with guidelines issued by the ACR/AF in 2019 [57] and the Single Hub and Access Point for Pediatric Rheumatology in Europe (SHARE) in 2018 [60]. Studies supporting the efficacy of specific medications are discussed below. (See 'Medication dosing and efficacy' below.)
Persistent or refractory disease — Our approach to the subsequent treatment of CAU in JIA is outlined in the algorithm (algorithm 4). For patients with JIA who have persistent CAU despite a three-month trial of topical glucocorticoids, or who cannot discontinue topical therapy, we start or change DMARD therapy. Specifically, for patients who were on topical glucocorticoids alone, we start subcutaneous methotrexate with folic acid. If they have subsequently developed severe disease (eg, severe, active chronic anterior uveitis, sight-threatening complications due to disease or glucocorticoid treatment), we also add a monoclonal TNF inhibitor. (See 'Medication dosing and efficacy' below.)
We define refractory disease as persistent inflammation despite three months of methotrexate therapy. In such cases, we add a monoclonal TNF inhibitor [57,61-64].
Patients who do not benefit from the combination of methotrexate and a TNF inhibitor may require additional intensification of therapy. If patients have a partial response to a TNF inhibitor, various approaches have been described:
●Increasing the dose – We ensure that the medication is appropriately dosed in growing children and, if need, consider increasing the dose. One study described the successful use of infliximab with doses of up to 20 mg/kg in children with CAU [65]. Alternatively, patients may respond to a "loading dose" of a TNF inhibitor. As an example, a study in adults used a one-time dose of 80 mg of adalimumab before decreasing to 40 mg every two weeks [66].
●Increasing the frequency – We consider increasing the frequency of TNF inhibitors. As an example, one study of 42 children with CAU found that 11 (41 percent) required escalation in the frequency of adalimumab from every 14 to every 7 days; after three to six months of treatment, seven of these patients (64 percent) improved and there were not any major safety concerns [67].
If patients do not respond to or cannot tolerate TNF inhibitors, there are case reports of successful use of cyclosporine [68], rituximab [69], abatacept [70,71], and tocilizumab [72], especially for severe, treatment-resistant disease. Mycophenolate mofetil may also be beneficial in refractory disease [73] and can sometimes be combined with bDMARDs. The 2019 ACR/AF guidelines specify tocilizumab as the next category of bDMARD to try after TNF inhibitor failure [57]. (See "Uveitis: Treatment", section on 'Other therapies'.)
While systemic glucocorticoids can be added to glucocorticoid-sparing immunosuppression [58,74], they are associated with a range of adverse effects and should therefore only be used for short periods of time. (See "Major adverse effects of systemic glucocorticoids".)
Data to directly compare treatment approaches for persistent uveitis are very limited; however, the combination of methotrexate and a monoclonal TNF inhibitor appears to be more effective than methotrexate alone. In one trial, 90 children with active uveitis and JIA (mostly oligoarticular) on a stable dose of methotrexate were continued on their current therapy and also randomly assigned to adalimumab (20 or 40 mg based upon body weight, administered subcutaneously every two weeks) or placebo for 18 months or until treatment failure [75]. Treatment failure was observed in 16 of 60 patients (27 percent) on adalimumab compared with 18 of 30 (60 percent) in the placebo group, with a significant delay in time to treatment failure in the adalimumab group (hazard ratio [HR] 0.25, 95% CI 0.12-0.49). In addition, compared with the placebo group, the adalimumab group was able to taper or discontinue topical glucocorticoids more often and had a longer mean duration of sustained inactive disease. However, adverse effects, including some serious ones, were more common in the adalimumab group. Limitations of the trial include that it was stopped early (90 rather than 114 patients enrolled), which can lead to an overestimation of treatment effect, and its limited generalizability given that patients had more severe disease.
Medication dosing and efficacy — Data to support the use of specific medications are discussed below:
●Topical glucocorticoids – Options for topical glucocorticoids to treat active CAU include prednisolone acetate 1%, dexamethasone 0.1, difluprednate 0.05%, and loteprednol 0.5%. The frequency of administration varies greatly (ie, every hour to every 24 hours) depending on the severity of inflammation.
While some guidelines favor certain glucocorticoid formulations, data directly comparing such formulations are very limited; in the authors' experience, any formulation may be effective [57]. Topical glucocorticoids were first shown to be beneficial for treating uveitis in a randomized trial published in 1979 [76] and subsequently, several other studies have demonstrated the efficacy of various formulations [77,78]. Adverse effects of ophthalmic glucocorticoids that occur with chronic use include increased intraocular pressure and cataracts.
●Methotrexate – Experts have recommended using subcutaneous methotrexate rather than oral methotrexate to treat active CAU in patients with JIA, in order to maximize absorption [57]. Dosing is the same as that used for other manifestations of JIA. (See "Oligoarticular juvenile idiopathic arthritis", section on 'Methotrexate for patients with poor prognostic factors'.)
Methotrexate is typically the first DMARD used in the treatment of JIA-associated uveitis [57,60,74,79-81]. In a retrospective study of 38 children with JIA-associated uveitis, 25 received methotrexate (subcutaneous or oral) at a mean dose of 15 mg/m2 [80]; within this group, 20 achieved remission after a median of four months, while four required additional therapy due to an incomplete response and one discontinued therapy due to intolerance. More information on the use of methotrexate for other forms of noninfectious uveitis is provided elsewhere. (See "Uveitis: Treatment", section on 'Antimetabolites'.)
●TNF inhibitors – Monoclonal TNF inhibitors that may be used to treat JIA-associated uveitis include adalimumab [75,82,83], infliximab [84,85], golimumab [60,86], and certolizumab [87]. Etanercept is not effective for treating uveitis [88], and data to support the relative efficacy of biosimilars of other TNF inhibitors are very limited. We start therapy at the dose and frequency that is typically used for patients with other manifestations of JIA; however, these may need to be intensified over time for those with persistent disease. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Tumor necrosis factor inhibitors' and 'Persistent or refractory disease' above.)
In a retrospective, multicenter study, 56 children (half with JIA) with active noninfectious uveitis despite therapy with glucocorticoids (topical therapy with or without oral therapy) and other csDMARDs were treated with an anti-TNF agent (infliximab n = 42, etanercept n = 11, adalimumab n = 3) [62]. Forty-seven and 59 percent achieved quiescence of uveitis at three and six months, respectively. Within 12 months, 75 percent achieved inactive uveitis and most (36 of 42 patients) were also able to discontinue all glucocorticoids. Notably, etanercept was subsequently found to be ineffective for the treatment of uveitis [88]. More information on the use of monoclonal TNF inhibitors to treat noninfectious uveitis is provided elsewhere. (See "Uveitis: Treatment", section on 'Tumor necrosis factor-alpha inhibitors'.)
Subsequent monitoring and duration of therapy — Once uveitis is diagnosed, the frequency of subsequent examinations is based upon the presence of complications and the response to and changes in therapy [57]. In general, patients follow with ophthalmology at least every three months when they are in remission on or off of therapy, and more frequently when escalating or tapering therapy [57].
The optimal duration of therapy for CAU depends on multiple factors, most notably the control of related JIA disease activity. However, for patients who require DMARD therapy, we generally aim to maintain remission on DMARD therapy without topical or systemic glucocorticoids for at least two years prior to tapering the DMARD [57,60]. Patients require frequent monitoring when tapering off of therapy.
The rate of flare after withdrawal of therapy was evaluated in a double-masked trial that randomly assigned 87 patients with JIA-related uveitis who had been in remission for ≥1 year on adalimumab to continue adalimumab or switch to placebo [89]. Over the two-year trial, treatment failure was more common in the group taking placebo compared with the group continuing adalimumab (68 versus 14 percent, respectively; HR 8.7, 95% CI 3.6-21.2). Treatment failures in the placebo group occurred after a median of 119 days, and all affected patients were ultimately able to reestablish sustained disease control with reinitiation of adalimumab after a median of 105 days. Similarly, in an observational study of 114 children with inactive noninfectious uveitis who tapered adalimumab, 46 percent developed recurrent uveitis after a median of 30 weeks from the start of the taper [90]. In the subset of 52 patients with JIA, the risk of disease recurrence was lower for patients who had at least two years of inactive disease prior to tapering (HR 0.62, 95% CI 0.43-0.94). Slower tapering schedules (ie, increasing the interval between adalimumab administrations by no more than one week every four months) were also associated with a reduction in relapse risk (HR 0.50, 95% CI 0.34-0.72).
Prognosis — Approximately one-third to one-half of patients with uveitis will have ocular complications, highlighting the importance of frequent monitoring and aggressive therapy [52,53]. Complications include cataracts, glaucoma, synechiae, band keratopathy, and macular edema, which can be related to the disease and/or its therapies (ie, adverse effects of glucocorticoids). As an example, a systematic review of 26 studies published between 1980 and 2004 found that the most common complications included cataracts (21 percent), glaucoma (19 percent), band keratopathy (16 percent), and significant visual impairment (defined as visual acuity <20/40 in both eyes; 9 percent) [91]. In a subsequent study from 2007 that included 142 patients with JIA and uveitis, 53 (37 percent) developed complications including cataracts (n = 33) and synechiae (n = 31), followed by glaucoma (n = 22), band keratopathy (n = 20), and macular edema (n = 7) [52]. After a mean follow-up time of seven years, 10 patients developed legal blindness in 10 eyes, and 4 patients developed impaired vision in 6 eyes. Of note, these papers were published before bDMARDs were widely available. Another retrospective review found that male sex was a risk factor for a more complicated course of uveitis and poorer visual outcome [92,93]. However, severe disease is also seen in females.
The rate of complications from uveitis may be decreasing over time, possibly reflecting the increasing availability of bDMARDs as well as a relatively more aggressive approach to therapy in patients with JIA [94]. One study examined two cohorts of patients with JIA, the first diagnosed between 1990 and 1993 and the second diagnosed between 2000 and 2003 [95]. Compared with the second cohort, the first cohort had a higher frequency of uveitis (25 versus 18 percent, respectively) and related complications (35 versus 21 percent, respectively), despite having a higher rate of systemic glucocorticoid use and a similar rate of methotrexate use.
Acute anterior uveitis — Acute anterior uveitis (AAU) can also affect patients with spondyloarthropathy (SpA), especially those who are positive for human leukocyte antigen B27 (HLA-B27). Among the subtypes of JIA, ERA is most commonly associated with AAU, followed by psJIA.
Patients with AAU classically experience acute-onset eye pain and redness, photophobia, and sometimes vision loss. Physical examination may reveal scleral injection and photophobia. Most patients suspected of having AAU should have an ophthalmology examination as soon as possible. (See "Uveitis: Etiology, clinical manifestations, and diagnosis", section on 'When to refer to an ophthalmologist'.)
Treatment of AAU in patients with JIA is similar to the initial treatment of CAU and typically involves topical glucocorticoids and/or a short course of oral glucocorticoids. For patients who develop recurrent episodes of AAU, DMARD therapy may reduce the risk of subsequent AAU flares. Supportive data for this approach are presented elsewhere. (See "Uveitis: Treatment", section on 'Prevention of recurrent episodes'.)
Leg-length discrepancy — Patients with JIA who have involvement of a major joint in the lower extremity (ie, knee, hip, ankle) may develop overgrowth in both the length and width of bones, leading to a leg-length discrepancy. While leg-length discrepancies <1 cm are common (in up to 70 percent of the population) and not clinically significant [96-98], greater discrepancies are associated with gait abnormalities and increased strain on the shorter leg [99,100].
●Pathophysiology – In patients with JIA, leg-length discrepancies are likely driven by the effects of proinflammatory cytokines on growth plates [101,102]. In young children, active arthritis can potentially cause neovascularization of the growth plate and increased growth of the affected bone. By contrast, active arthritis during early puberty may lead to premature fusion of the epiphysis and therefore impaired growth of the affected bone. While discrepancies of bone growth can occur in any affected joint, leg-length discrepancies are most commonly caused by knee arthritis since two-thirds of longitudinal growth occurs around this joint.
●Monitoring – We routinely monitor all patients with JIA for the development of leg-length discrepancy based on physical examination. This can be assessed in several ways, including looking for differences in knee height while lying supine with flexed knees, or in the height of the posterior superior iliac spine while standing.
●Prevention – Limited data suggest that the use of intraarticular glucocorticoids for patients with JIA who develop knee and/or ankle arthritis may prevent leg-length discrepancies. A retrospective study compared the development of leg-length discrepancy among 16 children with oligoarticular JIA who received intraarticular glucocorticoid injections of the knee or ankle with a similar group of 14 children who did not receive such injections [103]. None of the patients who received injections developed a significant leg-length discrepancy or required a shoe lift. By comparison, seven patients (50 percent) who did not receive injections developed leg-length discrepancies, averaging approximately 1 percent of total leg length.
●Management – Proper gait can be maintained in children with mild leg-length discrepancies (1 to 2 cm) by placing an appropriate lift in the opposite shoe. An absolute discrepancy of >2 to 2.5 cm should prompt orthopedic evaluation for possible surgical management [104]. Orthopedic consultation should be obtained before the child reaches skeletal maturity. By using radiographic standards to forecast the remaining growth, it is possible to perform a minor surgical procedure (stapling) to surgically close the distal tibial epiphysis in the leg that is longer and allow catch-up growth on the opposite side [104]. This procedure usually has a good outcome.
Short stature — Poor growth can affect patients with various subtypes of JIA, including an estimated 11 to 36 percent of patients with oligoarticular JIA [105-108]. In one study of 95 patients with persistent oligoarticular JIA, poor growth affected 36 percent of patients and was severe in 12 percent [105]. Poor growth was also more likely to affect children who required DMARD therapy compared with those who required intraarticular glucocorticoid injections alone, likely reflecting differences in disease severity rather than an effect of the specific treatments. Other risk factors for poor growth included younger age at onset and elevated erythrocyte sedimentation rate (ESR). (See "Causes of short stature", section on 'Rheumatologic disease'.)
Amyloidosis — The risk of developing secondary (AA) amyloidosis is low in patients with JIA and has dramatically decreased in the setting of expanding therapeutic options, to the point that most practicing pediatric rheumatologists report never having seen secondary amyloidosis. When it does occur, limited data suggest it is most common in those with sJIA (likely due to its robust autoinflammatory nature) and perhaps those with extended oligoarticular JIA [109]. (See "AA amyloidosis: Pathogenesis" and "AA amyloidosis: Causes and diagnosis".)
SOCIETY GUIDELINE LINKS —
Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Uveitis" and "Society guideline links: Juvenile idiopathic arthritis".)
INFORMATION FOR PATIENTS —
UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Juvenile idiopathic arthritis (The Basics)" and "Patient education: What you should know about vaccines (The Basics)" and "Patient education: Vaccines for babies and children age 0 to 6 years (The Basics)" and "Patient education: Vaccines for children age 7 to 18 years (The Basics)" and "Patient education: Measles, mumps, and rubella vaccine (The Basics)")
●Beyond the basics topics (See "Patient education: Why does my child need vaccines? (Beyond the Basics)" and "Patient education: Vaccines for infants and children age 0 to 6 years (Beyond the Basics)" and "Patient education: Vaccines for children age 7 to 18 years (Beyond the Basics)".)
SUMMARY AND RECOMMENDATIONS
●Immunizations – For patients with juvenile idiopathic arthritis (JIA) who are not taking immunosuppression, we maintain a routine immunization schedule. (See 'Patients not taking immunosuppression' above.)
Otherwise, for those who are taking immunosuppression, we take the following measures (See 'Patients taking immunosuppression' above.):
•Avoidance of live vaccines – Live vaccines (table 1) should not be given to children who are receiving systemic immunosuppression. (See 'Avoidance of live vaccines' above.)
•Adjustment of non-live vaccines – Due to the potential for decreased vaccine immunogenicity, we consider adjusting the timing of immunosuppressive medication or vaccine administration for patients taking methotrexate, rituximab, or higher doses of glucocorticoids. Selected patients may benefit from additional doses of certain non-live vaccines beyond the routine childhood immunization schedule (eg, pneumococcal, COVID-19, and herpes zoster vaccines). (See 'Adjustment of non-live vaccines' above.)
●Temporomandibular joint (TMJ) arthritis – TMJ arthritis is common in JIA and in refractory cases can lead to permanent changes in facial appearance. (See 'Temporomandibular joint arthritis' above.)
•Clinical presentation and evaluation – Patients with active TMJ arthritis may be asymptomatic or describe pain in their jaw or inner ear, morning stiffness, and/or difficulty fully opening their mouth. We monitor all patients with JIA for the development of TMJ arthritis based on history and physical examination (eg, evaluation of jaw asymmetry and opening). (See 'Clinical presentation and evaluation' above.)
•Diagnosis – If TMJ arthritis is suspected and its presence will impact treatment decisions, then MRI with contrast to confirm the diagnosis is paramount. Otherwise, we make a clinical diagnosis of TMJ arthritis and monitor patients closely. (See 'Diagnosis and differential diagnosis' above.)
•Treatment – Our approach to the initial treatment of TMJ arthritis is outlined in the algorithm (algorithm 1). For patients with active TMJ arthritis, we add or switch disease-modifying antirheumatic drug (DMARD) therapy to include methotrexate and/or a biologic DMARD (bDMARD), rather than using nonsteroidal antiinflammatory drugs (NSAIDs) alone (Grade 2C). The exact choice of DMARD is guided by baseline therapies and other active clinical manifestations of JIA. (See 'Treatment and prognosis' above.)
●Chronic anterior uveitis (CAU) – CAU is the most common extraarticular complication of JIA, affecting approximately 10 to 20 percent of all patients with JIA. (See 'Chronic anterior uveitis' above.)
•Clinical presentation and routine monitoring – While CAU is often initially silent, those with refractory disease can develop severe vision impairment. We refer all patients with JIA to an ophthalmologist for regular monitoring for CAU, which includes performing a complete slit lamp examination. The optimal approach to monitoring for CAU in patients with JIA varies depending on the patient's risk factors (algorithm 2); high-risk patients should be evaluated by ophthalmology within four weeks of JIA diagnosis. (See 'Clinical presentation and routine monitoring' above.)
•Treatment
-Initial therapy based on disease severity – We treat all patients with topical glucocorticoids (algorithm 3). For patients with features suggestive of severe disease (eg, profound visual impairment, macular edema, band keratopathy, cataracts, posterior synechiae, and/or glaucoma), we recommend starting or intensifying baseline DMARD therapy (Grade 1C). If patients are not taking immunosuppression at baseline, we start subcutaneous methotrexate; otherwise, patients may need to switch or add to their baseline DMARD(s) depending on extraocular JIA manifestations and response to therapy. (See 'Initial therapy based on disease severity' above.)
-Persistent or refractory disease – For most patients with JIA on topical glucocorticoid monotherapy for three months who have persistent CAU and/or who cannot discontinue glucocorticoids, we add subcutaneous methotrexate (algorithm 4) rather than other DMARDs (Grade 2C). If such patients have developed features of severe disease or if patients have refractory disease despite a three-month trial of methotrexate, then we also add a monoclonal tumor necrosis factor (TNF) inhibitor. (See 'Persistent or refractory disease' above.)
●Other complications – Other potential complications include acute anterior uveitis (AAU) in patients with spondyloarthropathy (SpA), leg-length discrepancy in those with involvement of a major joint in the lower extremity (ie, knee, hip, ankle), and poor growth with subsequent short stature. AA amyloidosis is exceedingly rare due to advances in therapy. (See 'Acute anterior uveitis' above and 'Leg-length discrepancy' above and 'Short stature' above and 'Amyloidosis' above.)
