Version July 2025
INTRODUCTION —
Systemic juvenile idiopathic arthritis (sJIA) is a form of inflammatory arthritis that is characterized by systemic features such as fever, evanescent rash, and, less commonly, macrophage activation syndrome (MAS). sJIA was previously known as Still's disease or systemic-onset juvenile rheumatoid arthritis. The underlying disease process is primarily autoinflammatory [1,2] and is thought to be the same as that in adult-onset Still's disease (AOSD) [3], leading to recommendations from the European Alliance of Associations for Rheumatology (EULAR) and Pediatric Rheumatology European Society (PReS) in 2024 to refer to both conditions as "Still's disease" [4].
This topic reviews the treatment of sJIA. The treatment of certain complications of sJIA, including MAS and sJIA-associated lung disease (sJIA-LD), and treatment of AOSD are discussed separately. (See "Systemic juvenile idiopathic arthritis: Complications" and "Adult-onset Still's disease: Treatment".)
More information on other aspects of sJIA, including classification, pathogenesis, and diagnosis, is also provided elsewhere:
●(See "Classification of juvenile idiopathic arthritis", section on 'Systemic arthritis'.)
●(See "Juvenile idiopathic arthritis: Epidemiology and immunopathogenesis".)
●(See "Systemic juvenile idiopathic arthritis: Clinical manifestations and diagnosis".)
DISEASE COURSE —
The course of systemic juvenile idiopathic arthritis (sJIA) is highly variable, which has prompted many observers to suspect that sJIA is not a single entity. The disease course in a typical child with sJIA starts with a several-month period of spiking fevers and rash, with varying degrees of arthralgia and arthritis. This may be followed by a relative quiescence of the systemic manifestations. However, there are some children who have persistently active systemic disease, which is associated with worse outcomes [5,6]. (See 'Prognosis' below.)
There are three recognized disease-course patterns: monophasic, polycyclic, and persistent [6-8]. The actual proportions of sJIA patients who follow these disease course patterns are unclear, as available data come from single-center studies that used variable definitions of remission and inactive disease.
●Monophasic course – These patients present with active sJIA which becomes completely quiescent with time, usually after four to six months, and does not recur. Adequate treatment is still needed to prevent potential disease damage.
●Polycyclic course – These patients have recurrences of active sJIA episodes with long periods of disease inactivity in between. The inactive periods may range from a few months to years. This was the least common course in all series.
●Persistent course – These patients continue to have chronic disease. This is the most common course, and there are variable patterns of disease:
•Systemic manifestations (ie, fever and rash and sometimes recurrent macrophage activation syndrome [MAS]), with little to no arthritis.
•Persistent systemic manifestations (sometimes with recurrent MAS) and progressive arthritis.
•Persistent arthritis despite resolution of systemic manifestations. Some of these children have relentless destructive arthritis that progresses despite any therapy and requires joint replacements at an early age.
Arthritis resolves completely in approximately 40 to 50 percent of patients [9]. The child who appears well after six months has a substantial probability of remaining well, whereas the child who continues to have active disease six months after diagnosis typically has a more difficult and chronic course [6,10,11]. However, some children have recurrences even after years of apparent remission. Thus, some patients who present with adult-onset Still's disease (AOSD) may, with careful questioning, give a history that suggests that the actual initial onset was in childhood.
GENERAL APPROACH TO TREATMENT
Urgency of treatment — We treat patients with systemic juvenile idiopathic arthritis (sJIA) immediately upon diagnosis because these patients are often very ill. They can develop severe, potentially life-threatening complications, such as severe serositis and/or macrophage activation syndrome (MAS), that require urgent treatment. In addition, evidence suggests that there is a window of opportunity and that the early use of some therapies (eg, interleukin 1 [IL-1] inhibitors) within three months of disease onset may change the course of disease [12-15]. (See 'Disease remission' below.)
While it is imperative to treat patients with sJIA early, the diagnosis of sJIA can be difficult to make given the lack of specific diagnostic tests; providers must make a diagnosis based on the constellation of signs and symptoms, as well as the exclusion of other possible diseases. (See "Systemic juvenile idiopathic arthritis: Clinical manifestations and diagnosis".)
Goals of therapy — When treating patients with sJIA, the aggressive and often permanently disabling nature of unremitting disease must be carefully balanced against the potential toxicities of therapy. Therefore, the two main goals of therapy are:
●Rapid control of symptoms to minimize the risk of complications related to persistent uncontrolled disease (eg, joint damage, chronic MAS)
●Judicious use of immunosuppression to minimize the risk of adverse effects from therapy, especially those related to chronic systemic glucocorticoids
When to refer — All patients with sJIA should be referred to an experienced pediatric rheumatologist to confirm the diagnosis and guide therapy.
PRETREATMENT CONSIDERATIONS —
Patients with systemic juvenile idiopathic arthritis (sJIA) often require immunosuppression, such as systemic glucocorticoids, biologic disease-modifying antirheumatic drugs (bDMARDs), and/or conventional synthetic DMARDs (csDMARDs). Patients should be evaluated for possible latent infections, including tuberculosis (TB) and hepatitis B virus, prior to starting immunosuppressive medications. (See "Tuberculosis infection (latent tuberculosis) in children", section on 'Test selection'.)
In addition, patients with sJIA who are starting immunosuppressive medications should receive all of their indicated preventive vaccinations. While it is optimal to administer preventive vaccinations two to four weeks prior to starting a bDMARD, delaying treatment while awaiting vaccination is not feasible or safe for many children presenting with active sJIA. Detailed guidance on the administration of live-viral vaccines and other standard childhood immunizations in patients with JIA is discussed in detail separately. (See "Juvenile idiopathic arthritis: Immunizations and complications", section on 'Immunizations'.)
PATIENTS WITH MACROPHAGE ACTIVATION SYNDROME —
Patients with systemic juvenile idiopathic arthritis (sJIA) are predisposed to developing macrophage activation syndrome (MAS), which is a potentially life-threatening complication that is often accompanied by unremitting fever, profound ferritin elevations, cytopenias, and liver dysfunction. Patients may display features of MAS at any point in their disease course. When MAS occurs in a patient for reasons that are unrelated to a rheumatic disease (eg, genetic defects, viral infections), it is often called hemophagocytic lymphohistiocytosis (HLH). The clinical manifestations, diagnosis, and treatment of MAS in the setting of sJIA and an overview of MAS/HLH are discussed separately:
●(See "Systemic juvenile idiopathic arthritis: Complications", section on 'Macrophage activation syndrome'.)
●(See "Clinical features and diagnosis of hemophagocytic lymphohistiocytosis".)
PATIENTS WITHOUT MACROPHAGE ACTIVATION SYNDROME
Initial therapy
Approach to initial therapy — Our general approach to initial treatment of patients with systemic juvenile idiopathic arthritis (sJIA) who do not have macrophage activation syndrome (MAS) typically involves a nonsteroidal antiinflammatory drug (NSAID) for most patients, plus a systemic immunosuppressive medication for those with moderate to severe disease. Our approach is outlined in the algorithm (algorithm 1), with drug dosing provided in the table (table 1).
The choice of initial therapy depends upon the severity of symptoms. For the purposes of choosing initial therapy, we define moderate to severe sJIA as having one or more of the following features:
●Moderate to severe symptoms or signs of systemic disease (eg, high fever, diffuse lymphadenopathy, hepatomegaly, splenomegaly)
●Disabling arthritis (eg, >5 joints, interfering with daily activities)
●Significant elevations in C-reactive protein (CRP; ie, >14 mg/dL)
●Signs or symptoms suggestive of evolving MAS (eg, hepatic dysfunction, high ferritin, relative cytopenias) (see "Systemic juvenile idiopathic arthritis: Complications", section on 'Macrophage activation syndrome')
If none of the above features are present, we consider patients to have mild sJIA. While NSAID monotherapy may be sufficient for rare patients with mild sJIA [16], patients with moderate to severe sJIA are also treated with an interleukin 1 (IL-1) or IL-6 inhibitor. Glucocorticoids are avoided if possible due to their adverse effects and the availability of more targeted immunosuppressive options; however, they may be required when it is essential to rapidly control certain symptoms and complications (eg, severe serositis). (See 'Mild disease' below and 'Moderate to severe disease' below and 'Indications for glucocorticoids' below.)
Evidence to support one type of treatment over another is extremely limited. Our approach is generally consistent with those outlined in the 2021 American College of Rheumatology (ACR) guidelines for the treatment of sJIA [17]. Additional research is needed on the comparative effectiveness of various therapeutic strategies [18,19] and the optimal selection of biologic agents for specific manifestations of sJIA [20].
Mild disease — In most patients with mild sJIA, we suggest a trial of monotherapy with an NSAID rather than a systemic glucocorticoid or a conventional synthetic disease-modifying antirheumatic drug (csDMARD). We prefer this over alternative options (ie, systemic glucocorticoids or DMARDs) since NSAIDs are generally well tolerated and are not immunosuppressive. This makes them especially useful for empiric initial treatment of sJIA when the diagnosis may not be clear and the evaluation is ongoing (see 'NSAIDs' below). If patients do not respond completely after one to two weeks or have worsening disease, then we escalate therapy. (See 'Persistent initial disease' below.)
Limited data suggest that NSAIDs may be effective as monotherapy for patients with mild disease. In a retrospective review that included 51 patients with new-onset sJIA who were initially treated with NSAIDs as monotherapy [21], clinically inactive disease was achieved in 13 patients and was associated with certain patient characteristics at presentation including a younger age (≤8 years old), a lower active joint count (≤5), and a lower CRP (≤13 mg/dL).
Moderate to severe disease — In most patients with moderate to severe sJIA, we suggest starting an IL-1 or IL-6 inhibitor rather than a csDMARD (see 'Interleukin 1 inhibitors' below and 'Interleukin 6 inhibitors' below). Patients may also receive adjunctive NSAIDs if needed for fever and pain relief. (See 'NSAIDs' below.)
In the absence of more definitive evidence to help choose between various agents, we select an IL-1 or IL-6 inhibitor through shared decision-making with the patient/family based upon the specific clinical scenario and differences in routes of administration, frequency of dosing, potential adverse effects, and regulatory issues, which are summarized in the table (table 1). As an example, when choosing between types of IL-1 inhibitors, anakinra may be favored for patients with unstable earlier disease as it is easier to adjust, while canakinumab may be preferable for patients with more stable disease who wish to minimize the frequency of subcutaneous injections.
We favor the use of an IL-1 or IL-6 inhibitor since these agents are more targeted to the specific underlying inflammatory pathway [22,23], have a more rapid onset of action than csDMARDs, and have less substantial and fewer cumulative adverse drug effects than systemic glucocorticoids. However, there is very little comparative effectiveness data for various DMARDs in sJIA. Key studies include the following:
●The Childhood Arthritis and Rheumatology Research Alliance (CARRA) has evaluated consensus treatment plans (CTPs) for the initial treatment of sJIA in two observational studies, which included options with biologic DMARDs (bDMARDs) and csDMARDs [24,25]. In the larger study, 63 of the 73 patients who were enrolled were treated with a bDMARD CTP (of these, 59 were treated with IL-1 inhibitor) and 10 received a nonbiologic CTP (glucocorticoids alone or methotrexate with or without glucocorticoids) [25]. While the imbalance in treatment groups limited the ability to test for meaningful differences between them, clinically inactive disease without the use of glucocorticoids was able to be measured in 53 patients at the nine-month visit and was achieved in 29 of the 47 patients on a biologic CTP and three of the six patients on a nonbiologic CTP.
●When comparing the various bDMARDs, the limited data available are relatively less supportive of rilonacept. A meta-analysis of five trials comparing an IL-1 or IL-6 inhibitor with placebo found low-quality evidence that patients who received rilonacept were less likely to improve on therapy compared with those who received canakinumab or tocilizumab (odds ratio [OR] 0.1, 95% CI 0.02-0.38 and OR 0.12, 95% CI 0.03-0.44, respectively) [26].
Indications for glucocorticoids — Systemic glucocorticoids act rapidly and are used together with other glucocorticoid-sparing agents to treat sJIA in the following scenarios:
●Patients with severe systemic symptoms, including severe serositis (eg, pericarditis), persistent fever, lymphadenopathy, hepatomegaly, splenomegaly, and/or severe anemia.
●Patients with relatively severe systemic symptoms who have suspected early MAS (eg, transaminitis, uptrending ferritin, downtrending cell counts and/or erythrocyte sedimentation rate [ESR] despite ongoing inflammation). (See "Systemic juvenile idiopathic arthritis: Complications", section on 'Macrophage activation syndrome'.)
●Patients who have progressive systemic symptoms despite initial therapy. (See 'Persistent initial disease' below.)
Glucocorticoids should not be used unless the possibility of malignancy, which may cause a similar presentation, has been excluded.
Intraarticular glucocorticoids may also be added to other therapies for patients with peripheral arthritis affecting joints that are amenable to injection. The procedure may require sedation, especially in younger children. Intraarticular glucocorticoids should not be given if there is concern for a possible septic joint. The dosing of glucocorticoids and approach to arthrocentesis, including complications, are described in detail elsewhere. (See "Joint aspiration or injection in children: Indications, technique, and complications".)
Dosing, adverse effects, and efficacy
NSAIDs — NSAIDs are commonly used to treat symptoms of fever and arthritis in patients with sJIA and may be used as monotherapy for those with mild disease. Any NSAID other than aspirin may be used at standard antiinflammatory doses (eg, ibuprofen 10 mg/kg/dose orally three to four times daily [maximum 2400 mg/day] or naproxen 5 to 10 mg/kg/dose orally twice daily [maximum 1000 mg/day]). Aspirin is not commonly used due to its potential toxicities and the need for more intensive monitoring when it is given at antiinflammatory doses. Adverse effects of NSAIDs, including hepatic and kidney dysfunction, and the primary prevention of gastroduodenal toxicity are described in detail elsewhere. (See "Nonselective NSAIDs: Overview of adverse effects" and "Overview of COX-2 selective NSAIDs" and "NSAIDs (including aspirin): Primary prevention of gastroduodenal toxicity".)
There are no clear data to favor the use of one NSAID over another for sJIA. Anecdotally, indomethacin may be more effective for control of recalcitrant fever. The safety profiles of various NSAIDs appear similar for patients with JIA. In an observational study of 274 patients with various types of JIA, there was a similar rate of adverse effects among those taking a nonselective NSAID (eg, naproxen, meloxicam, nabumetone) and those taking the cyclooxygenase 2 (COX-2) selective NSAID celecoxib [27]; of note, only three patients had sJIA and none of them took celecoxib.
Selected bDMARDs — Certain bDMARDs, specifically IL-1 and IL-6 inhibitors, are frequently used to treat sJIA. They require close monitoring for potential complications, including potential drug reactions. (See 'Drug therapy' below.)
Interleukin 1 inhibitors — IL-1 is one of the predominant proinflammatory cytokines underlying the inflammatory symptoms of sJIA [22]. Thus, IL-1 inhibitors such as anakinra (recombinant IL-1 receptor antagonist), rilonacept (IL-1 trap), and canakinumab (anti-IL-1-beta monoclonal antibody) are beneficial in many children with sJIA. All available IL-1 inhibitors can be administered by subcutaneous injection, and, in addition, anakinra may be given intravenously. Anakinra has the shortest half-life and therefore the most frequent administration (daily), while rilonacept and canakinumab are less frequent (weekly and monthly, respectively). More detailed information on the dosing, adverse effects, and monitoring of IL-1 inhibitors is provided separately. (See "Interleukin 1 inhibitors: Biology, principles of use, and adverse events".)
Direct evidence to support the use of these agents for patients with sJIA consists of a small number of randomized trials and multiple observational studies, which are summarized below. Indirect evidence can also be extrapolated from their use in patients with adult-onset Still's disease (AOSD), which is described in detail elsewhere. (See "Adult-onset Still's disease: Treatment", section on 'Initial therapy: New-onset disease'.)
●Anakinra – Anakinra is a recombinant IL-1 receptor antagonist that blocks both IL-1-alpha and -beta.
•Dosing – The optimal dosing of anakinra remains undefined. The authors' practice is to start at a dose of 2 to 4 mg/kg/day for most patients [12,18,28-30], with subsequent dose escalation to 4 mg/kg/day or higher if patients have not improved after several days [18,31]. This may be given subcutaneously or intravenously. Younger children have different drug pharmacokinetics and may therefore require higher initial doses.
•Evidence – Several observational studies have examined the use of anakinra for the initial or maintenance treatment of sJIA and most, although not all, have noted a benefit [12,28,32-35]. Anakinra may be most effective if used early in the disease course, rather than as "rescue" therapy once other therapies have failed. Key studies include the following:
-In a multicenter case series of 46 patients who had received anakinra as part of initial therapy for sJIA, 60 percent of patients experienced a compete response without the need to add further therapy, including 8 of the 10 patients taking anakinra as monotherapy [12]. Furthermore, after at least six months of follow-up, only 11 percent of patients had active arthritis, compared with 36 to 63 percent among historical controls.
-In a subsequent prospective case series, anakinra was initiated in 42 patients with new-onset sJIA who had not responded to a week of NSAID therapy [14]. The initial dose of anakinra was 2 mg/kg/day; if fever persisted for three days, the dose was subsequently increased to 4 mg/kg/day, with the option to add prednisolone or switch to another biologic agent for refractory symptoms. After one month of therapy, inactive disease on anakinra monotherapy was achieved in 23 patients (55 percent); after three months, 35 patients (83 percent) had clinically inactive disease, including 30 patients on anakinra monotherapy. Ultimately, only a third of patients required concomitant glucocorticoids, compared with over 80 percent of patients in cohort studies prior to the availability of biologics. Patients who did not have arthritis on initial presentation still had a positive response to therapy and development of articular damage over a median of 5.8 years of follow-up was rare (5 percent).
●Canakinumab – Canakinumab is an anti-IL-1-beta monoclonal antibody.
•Dosing – The typical dose of canakinumab for sJIA is 4 mg/kg (maximum dose 300 mg) every four weeks, given subcutaneously.
•Evidence – Results from two randomized trials, in addition to observational data, indicate that canakinumab is an effective therapeutic option for patients with sJIA [36,37]:
-In the first trial, 84 children (aged 2 to 19 years old) with active sJIA were randomly assigned to a single dose of canakinumab (4 mg/kg subcutaneously) or placebo, with or without background therapy (eg, NSAIDs, systemic glucocorticoids, and/or methotrexate) [36]. The percentage of patients achieving an adapted JIA ACR 30 response was higher in the canakinumab group compared with the placebo group (84 versus 10 percent, respectively), with a low rate of serious adverse events (two in each group).
-The second trial had an initial open-label phase, where 177 patients (including canakinumab responders and placebo-group patients from the initial trial) were treated with canakinumab every four weeks for 12 to 32 weeks [36]. In the subsequent withdrawal phase, the 100 patients who had a positive sustained response (as defined by sustaining an adapted JIA ACR 30 response or better and discontinuing or tapering glucocorticoids to a low, stable dose) were randomly assigned to continue canakinumab or switch to placebo. Compared with the placebo group, those taking anakinra at the end of the withdrawal phase had a lower rate of disease flare (26 versus 75 percent, respectively) and a higher rate of inactive disease (62 versus 34 percent, respectively).
●Rilonacept – Rilonacept is an IL-1 receptor trap fusion protein that blocks IL-1-alpha and -beta, as well as IL-1 receptor antagonist.
•Dosing – The typical dosing of rilonacept for adolescents ages 12 to 17 years is an initial loading dose of 4.4 mg/kg subcutaneously in divided doses as necessary (maximum 160 mg per injection), followed one week later by a maintenance dose of 2.2 mg/kg (maximum dose 160 mg) once weekly.
•Evidence – Data to support the use of rilonacept in treating patients with sJIA come from one randomized trial and an exploratory study:
-A trial of 71 children with sJIA randomly assigned patients to initial therapy with rilonacept (loading dose of 4.4 mg/kg, then 2.2 mg/kg, given subcutaneously every week) versus placebo for four weeks, followed by an open-label phase for 20 weeks where all patients received rilonacept [38]. After the initial four weeks, the number of patients who had a favorable treatment response was higher among the rilonacept group compared with the placebo group (20 of 35 patients [57 percent] versus 9 of 33 patients [27 percent], respectively). In addition, those who initially received rilonacept had a shorter time to response compared with those who received placebo (median four versus eight weeks, respectively).
-During the blinded phase of an exploratory trial of 24 patients with refractory sJIA who were randomly assigned 2:1 to rilonacept (2.2 mg/kg in cohort 1 and 4.4 mg/kg in cohort 2, given subcutaneously on days 3, 7, 14, and 21) or placebo, there were no significant differences in treatment response [39]. However, during the subsequent open-label phase, all 23 patients had complete resolution of fever and rash by three months and sustained this response over the duration of their enrollment (up to 23 months), allowing over 90 percent of patients to reduce or discontinue glucocorticoids. A limitation of the study was a relatively high rate of withdrawal during the open-label phase (13 of 23 patients), most commonly due to lack of efficacy, adverse treatment effects, or withdrawal of consent (n = 3 for each).
Interleukin 6 inhibitors — Tocilizumab is a human recombinant monoclonal antibody against the IL-6 receptor.
●Dosing – Tocilizumab can be given subcutaneously for sJIA at a dose of 162 mg every two weeks for patients weighing <30 kg, or weekly for those weighing ≥30 kg [40]. Alternatively, it may be administered intravenously at a dose of 12 mg/kg every two weeks in patients weighing <30 kg, or 8 mg/kg every two weeks in patients weighing ≥30 kg. This dosing is notably higher than that used for polyarticular JIA. More detailed information on the dosing, adverse effects, and monitoring of IL-6 inhibitors is provided separately. (See "Interleukin 6 inhibitors: Biology, principles of use, and adverse effects".)
●Evidence – Results of one randomized, placebo-controlled trial [41], as well as a withdrawal phase trial [42] and multiple observational studies [43-45], support the use of tocilizumab in sJIA. As an example, a trial of 112 children ages 2 to 17 years old with active sJIA despite therapy with NSAIDs and glucocorticoids randomly assigned patients 2:1 to tocilizumab (12 mg/kg for children <30 kg or 8 mg/kg for children ≥30 kg, given intravenously every two weeks) or placebo [41]. After 12 weeks, rates of improvement as measured by ACR Pedi 30, 50, and 70 responses were higher in the treatment group compared with the placebo group (85, 85, and 71 percent, respectively, versus 24, 11, and 8 percent, respectively). Adverse events associated with tocilizumab included infections (eg, sinopulmonary infections, septic arthritis, varicella, gastroenteritis), MAS, neutropenia, and elevated liver function tests and cholesterol.
Indirect evidence to support tocilizumab in sJIA can be derived from its use in AOSD and other forms of JIA, as described elsewhere. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'IL-6 inhibitors' and "Adult-onset Still's disease: Treatment".)
Systemic glucocorticoids
●Dosing – Due to the multiple adverse effects of systemic glucocorticoids, patients should receive the lowest dose that controls symptoms for the shortest duration possible. Ideally, we try to limit the dose to ≤0.5 mg/kg/day of prednisone (or its equivalent) and the duration to ≤6 months. Patients with severe manifestations may require high-dose glucocorticoids (ie, 1 to 2 mg/kg/day, to a maximum dose of 60 mg) or, more rarely, pulse doses (ie, 30 mg/kg [maximum dose 1 g] intravenously daily, for up to three days). More information on the general use of glucocorticoids, including pharmacology and considerations in choosing a specific regimen, is provided elsewhere. (See "Overview of the pharmacologic use of glucocorticoids".)
●Adverse effects – The adverse effects and monitoring of systemic glucocorticoids are summarized in the table (table 2) and described in detail elsewhere. (See "Major adverse effects of systemic glucocorticoids".)
Systemic glucocorticoids may be used together with NSAIDs; however, this combination can increase the risk of gastrointestinal toxicity and primary prophylaxis (eg, a proton pump inhibitor) may be required. (See "NSAIDs (including aspirin): Primary prevention of gastroduodenal toxicity".)
●Evidence – While there are no randomized trials of glucocorticoids for the treatment of sJIA, they are commonly used and are included in the ACR guidelines for the treatment of JIA [17] as well as the CARRA consensus treatment plan for new-onset sJIA [18].
Subsequent therapy
Persistent initial disease — Patients are considered to have persistent acute disease when they have persistent or worsening symptoms or laboratory abnormalities despite one to two weeks of initial therapy (see 'Disease activity and complications' below). The severity of the disease will influence how quickly therapy is adjusted. As an example, most pediatric rheumatologists would escalate therapy within a week for patients with persistent polyarthritis, fever, and rash, and sooner if there was concern for evolving MAS or severe serositis.
Some patients will have persistent systemic features (eg, fever, serositis), while others have persistent arthritis (which can be progressive and destructive), and still others have a mixture of both. The pathway to escalate therapy depends upon these refractory clinical manifestations as well as the current treatment being used, as outlined in the algorithm (algorithm 1) and summarized below, with drug dosing provided in the table (table 1):
●Patients taking NSAID monotherapy – For most patients with mild sJIA who have persistent initial disease despite a short trial (ie, one to two weeks) of NSAID monotherapy, we suggest adding IL-1 or an IL-6 inhibitor with or without a systemic glucocorticoid, rather than adding a systemic glucocorticoid alone or a csDMARD. This approach is similar to that used for the initial treatment of patients with moderate to severe disease, including the choice of IL-1 or IL-6 inhibitor, drug dosing and monitoring, and reservation of systemic glucocorticoids for severe symptoms. (See 'Moderate to severe disease' above and 'Indications for glucocorticoids' above.)
●Patients taking an IL-1 or IL-6 inhibitor – For most patients with sJIA who have persistent initial disease despite a trial of an IL-1 or IL-6 inhibitor, we suggest switching to an alternative IL-1 or IL-6 inhibitor, rather than adding a systemic glucocorticoid alone or adding a csDMARD. (See 'Interleukin 1 inhibitors' above and 'Interleukin 6 inhibitors' above.)
Alternative approaches for patients with persistent arthritis include switching to a different type of bDMARD (eg, a tumor necrosis factor [TNF] inhibitor, abatacept, or Janus kinase [JAK] inhibitor), or, for those who have otherwise had a good response to the initial IL-1 or IL-6 inhibitor, adding intraarticular glucocorticoid injections or a csDMARD (eg, methotrexate or leflunomide). (See 'TNF inhibitors' below and 'Abatacept' below and 'JAK inhibitors' below and 'csDMARDs' below.)
As with other clinical scenarios, we add systemic glucocorticoids only for patients with severe disease manifestations. (See 'Systemic glucocorticoids' above.)
There are no published data directly comparing treatment options for initial treatment failure for sJIA. Our approach is consistent with the 2021 ACR guidelines for the treatment of sJIA [17] and is based on the cumulative toxicity of systemic glucocorticoids, demonstrated efficacy and steroid-sparing effects of bDMARDs [36,41], and relatively limited efficacy of csDMARDs and bDMARDs (other than IL-1 and IL-6 inhibitors) for systemic manifestations in sJIA.
Refractory chronic disease — We consider patients to have refractory chronic disease when after ≥6 months they have persistent, worsening, or recurrent systemic and/or arthritic symptoms despite separate trials of an IL-1 and an IL-6 inhibitor, and/or if they have been unable to taper off of systemic glucocorticoids [46].
The optimal type of therapy for refractory chronic sJIA will depend on the predominant clinical features (ie, predominately systemic features, predominant arthritis, or a mix of both):
●Predominant systemic features – Therapies that appear to have a prominent effect on the systemic features in sJIA include JAK inhibitors and cyclosporine (see 'JAK inhibitors' below and 'csDMARDs' below). The treatment of patients with acute or chronic MAS is described elsewhere. (See "Systemic juvenile idiopathic arthritis: Complications", section on 'Macrophage activation syndrome'.)
●Predominant arthritis – Patients with prominent arthritic symptoms may benefit from adding or switching to DMARDs that commonly treat other forms of JIA, including methotrexate, leflunomide, TNF inhibitors, and abatacept [47,48]. Some of these therapies may be less effective for treating the systemic features of disease.
The therapeutic regimen for refractory chronic disease will also depend on the tolerance of and response to prior therapies, as often different agents must be combined. We generally avoid combining two bDMARDs or a bDMARD with a JAK inhibitor, but we consider such combinations for patients with highly refractory disease. We do often combine a bDMARD with a csDMARD (eg, an IL-1 inhibitor plus methotrexate). Systemic or intraarticular glucocorticoids may be added to any immunosuppressive regimen. (See 'Indications for glucocorticoids' above.)
Recurrent or polyphasic disease — Some patients have a polyphasic disease course, with the disease becoming completely quiescent for months or years without treatment and then recurring episodically [6,7]. The approach to treating these recurrent episodes is the same as that used for initial therapy.
Dosing, adverse effects, and efficacy — The use and efficacy of various csDMARDs and bDMARDs for patients with ongoing arthritis are summarized below.
csDMARDs — Conventional synthetic disease-modifying antirheumatic drugs (csDMARDs) that are used for the treatment of sJIA include methotrexate, leflunomide, and cyclosporine.
●Methotrexate – Methotrexate is an antimetabolite and is a type of csDMARD.
•Dosing and monitoring – Methotrexate is given either orally or subcutaneously once weekly at a dose of up to 25 to 30 mg/m2 (0.5 to 1 mg/kg) (maximum dose 25 mg). Oral absorption is unreliable at doses above 15 to 20 mg/m2 (roughly 0.5 mg/kg); thus, whenever possible, we give such doses subcutaneously or in divided oral doses (ie, split into two doses, given twice daily once a week) [49]. Methotrexate is administered together with folic acid (0.4 to 1 mg orally once daily) or leucovorin (2.5 to 5 mg orally once a week) to minimize adverse effects such as nausea and mouth sores. It should be avoided in patients with liver disease and requires close laboratory monitoring. More detailed information on the potential adverse effects and monitoring of methotrexate is provided separately. (See "Major adverse effects of low-dose methotrexate".)
•Evidence – Based on limited direct evidence, methotrexate appears to be effective for treating sJIA, especially for the arthritic symptoms. Key studies include the following:
-In one trial, 45 children <16 years old with sJIA were randomly assigned to a four-month treatment period of weekly oral methotrexate (15 mg/m2, with the potential to increase to 20 mg/m2 after two months in the absence of improvement) or placebo, followed by a two-month washout period and then a crossover to the other treatment arm for an additional four months of treatment [50]. While there were greater improvements in the physician and parent global assessment of disease activity for patients taking methotrexate compared with those taking placebo, improvements in other core measures (number of active joints, range of joint motion, and ESR level) and in systemic symptoms were similar between the two treatment groups.
-In an observational cohort study that included 25 patients with sJIA who were initially treated with methotrexate, after three months 84 percent had at achieved an ACR pediatric 30 response, and after 12 months 77 percent had achieved an ACR pediatric 70 response [51].
Indirect evidence to support the use of methotrexate in AOSD and other forms of JIA is discussed elsewhere. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Methotrexate'.)
●Leflunomide – Leflunomide is a type of csDMARD that affects many components of immune and inflammatory responses. It is given orally at a dose and frequency determined by the child's weight. Patients require regular monitoring for the development of cytopenias and hepatotoxicity.
Leflunomide has not specifically been studied in sJIA. It is effective for treating other forms of JIA, although it is less effective in comparison with methotrexate [52], as is described elsewhere. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Leflunomide'.)
●Cyclosporine – Cyclosporine is a calcineurin inhibitor that has been used with some frequency in sJIA, mainly in patients with chronic refractory systemic activity and/or acute or chronic MAS [53,54], which is described elsewhere. (See "Systemic juvenile idiopathic arthritis: Complications" and "Treatment and prognosis of hemophagocytic lymphohistiocytosis".)
The adverse effects and monitoring of cyclosporine are also discussed in detail separately. (See "Pharmacology of calcineurin inhibitors".)
TNF inhibitors — Tumor necrosis factor (TNF) inhibitors are a type of bDMARD; while they are not specifically approved by the US Food and Drug Administration (FDA) or the European Medicines Agency (EMA) for sJIA, they are commonly used to treat other forms of JIA.
●Choice of TNF inhibitor and dosing – There are multiple types of TNF inhibitors, which vary slightly in their form and frequency of administration as well as the ability to adjust dosing (table 3). The initial dosing is generally similar to that used for other forms of JIA; however, higher doses of TNF inhibitors may be required to treat sJIA due to the autoinflammatory nature of this condition [55].
●Adverse effects and monitoring – TNF inhibitors are immunosuppressive and are associated an increased risk of fungal infections (eg, histoplasmosis) and reactivation of latent tuberculosis (TB). While the FDA has issued a warning about the incidence of lymphoma and other malignancies in children treated with TNF inhibitors, subsequent research suggests that this risk is most likely due to the underlying disease and not the type of treatment [56]. More information of the risk of malignancy, other adverse effects, and drug monitoring is provided elsewhere. (See "Tumor necrosis factor-alpha inhibitors: Risk of malignancy" and "Tumor necrosis factor-alpha inhibitors: An overview of adverse effects".)
●Evidence – Data to support the efficacy of TNF inhibitors are all observational, and the results suggest limited and unpredictable benefits [57-62]. Multiple studies have also found that TNF inhibitors are less beneficial for children with sJIA compared with other forms of JIA [57,60-62] but may be of benefit in children who have developed chronic arthritis. The following examples are illustrative:
•In an observational study of patients in the Dutch national register with various types of JIA who were initiated on etanercept as their first biologic therapy, the rate of remission over a median of 2.5 years of follow-up was similar across the various subtypes of JIA, including 38 percent of patients with sJIA [60]. However, in follow-up studies, sJIA was identified as a risk factor for a poor response to etanercept [62]. Almost half (24 of 51 patients) of the patients with sJIA ultimately switched to an alternative therapy, most often due to lack of efficacy [63].
•Another observational study examined the response to multiple types of TNF inhibitors in 45 patients with sJIA who were on therapy for >6 months [61]. Over a median follow-up of 24 months, 24 percent of patients experienced remission (eight on etanercept, two on infliximab, and one on adalimumab). However, the remission was only sustained in 13 percent of patients.
Indirect evidence to support the use of TNF inhibitors in AOSD and other forms of JIA is discussed separately. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Tumor necrosis factor inhibitors' and "Adult-onset Still's disease: Treatment", section on 'Refractory to standard therapies'.)
Abatacept — Abatacept is a T-cell costimulatory blocker and a type of bDMARD. It is approved for the treatment of other forms of JIA. It is dosed subcutaneously once a week based on weight.
Direct data to support the use of abatacept in sJIA are limited to small observational studies. In a systematic review that identified seven observational studies, 80 percent of the 668 patients achieved a modest improvement and 40 percent had a more robust improvement (as defined by the ACR pediatric 30 and 70 responses, respectively) [64]. Indirect evidence to support the use of abatacept in other forms of JIA is discussed separately. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Abatacept'.)
JAK inhibitors — Janus kinase (JAK) inhibitors are a type of targeted synthetic DMARD (tsDMARD) that may be added to a csDMARD (eg, methotrexate) with or without glucocorticoids.
●Dosing and monitoring – We typically use tofacitinib, which is approved by the FDA and EMA for patients with several other forms of JIA. Tofacitinib is given by mouth at a weight-based dose twice daily. Alternative JAK inhibitors include upadacitinib, which is approved by the FDA for the treatment of polyarticular JIA, and, for those with concomitant MAS, ruxolitinib. More information on the mechanism of action, adverse effects, and drug monitoring of JAK inhibitors is provided elsewhere. (See "Overview of the Janus kinase inhibitors for rheumatologic and other inflammatory disorders".)
●Evidence – Data to support the use of JAK inhibitors in sJIA are very limited. There are ongoing clinical trials of JAK inhibitors in sJIA, and anecdotally they are used for refractory cases and lung disease associated with sJIA (sJIA-LD) [65,66]. There are case reports of potential efficacy in patients with sJIA [67,68] and a case series in AOSD [69]. (See "Adult-onset Still's disease: Treatment", section on 'Refractory to standard therapies'.)
Indirect evidence to support their use can also be derived from their efficacy in treating multiple other types of inflammatory arthritis in children and adults, including polyarticular JIA. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Small molecule inhibitors'.)
Other therapies — Other treatment strategies may be used in patients who fail the more standard therapies above, including the following:
●Rituximab – Rituximab is a type of bDMARD, specifically an anti-CD20 antibody that depletes B cells. It is sometimes used to treat other types of refractory inflammatory arthritis, most commonly seropositive rheumatoid arthritis in adults. Data to support its use in sJIA are extremely limited, although there is one successful case series [70]. More information about the mechanism, dosing, adverse effects, and monitoring of rituximab and its use in other forms of JIA is provided separately. (See "Rituximab: Principles of use and adverse effects in rheumatologic disease" and "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Rituximab'.)
●Cyclophosphamide – Cyclophosphamide is a cytotoxic drug that very rarely may be used for short periods of time to treat severe refractory sJIA that does not respond to extensive use of less toxic medications. This approach became even less common once bDMARDs were available. Cyclophosphamide has also been used for certain complications of sJIA, including MAS, pulmonary hypertension, and interstitial lung disease. (See "Systemic juvenile idiopathic arthritis: Complications", section on 'Treatment of refractory disease'.)
The dosing and duration of cyclophosphamide is described separately. The main reason cyclophosphamide is not more widely used is the associated toxicity (eg, leukopenia, infection) and risks of future malignancy and infertility. Adverse effects of cyclophosphamide as well as monitoring strategies are described in more detail elsewhere. (See "Cyclophosphamide in rheumatic diseases: General principles of use and toxicity".)
Data to support the use of cyclophosphamide in sJIA are very limited. In one observational study, four patients with severe sJIA initially received 6 to 10 monthly infusions of intravenous cyclophosphamide (500 to 1000 mg/m2/dose) and methylprednisolone (30 mg/kg/dose, maximum 1 g); subsequent treatments were spaced out to be once every two to three months [71]. After between 12 and 20 infusions, all patients improved and three experienced disease remission.
●Hematopoietic stem cell transplantation – Hematopoietic stem cell transplantation (HCT) historically has been used in children with relentless sJIA that is uncontrolled despite the use of multiple drugs [72-76]. However, HCT was primarily used prior to the advent of effective biologic therapies. Given its high mortality rate [77], the use of HCT is restricted to only the most severely affected children refractory to other, less risky therapies. Several series of sJIA patients who underwent allogeneic transplants reported some success [76,78]. The use of HCT is described in more detail separately. (See "Overview of hematopoietic stem cells", section on 'Therapeutic uses of HSCs'.)
MONITORING —
The disease course in systemic juvenile idiopathic arthritis (sJIA) can be unstable, especially during the first few months after diagnosis. Patients with sJIA who are starting initial therapy must be monitored closely for their response to therapy as well as potential adverse effects of therapy. Anecdotally, the initiation of any therapy in patients with sJIA, including nonsteroidal antiinflammatory drugs (NSAIDs), has been associated with a transiently elevated risk of developing macrophage activation syndrome (MAS).
Disease activity and complications
●Clinical evaluation – Timing and frequency of clinical reassessment will vary depending on the patient's acuity and care setting. For most patients who are not hospitalized, we typically arrange follow-up within one to two weeks of initiating therapy. While subsequent follow-up will vary based on the therapeutic response, we generally see patients every two to six weeks until achieving remission. If patients are taking a glucocorticoid, we also continue this frequency of monitoring until patients are able to discontinue the glucocorticoid. Once glucocorticoids have been discontinued and the patient is stable, we see them approximately every three months.
At follow-up evaluations, we assess symptomatic improvement (eg, fever, arthritis, rash) and tolerance of therapy. We also look carefully for any signs of evolving complications, such as MAS (eg, new hepatosplenomegaly) or sJIA-associated lung disease (sJIA-LD; eg, clubbing, abnormal pulmonary examination). (See "Systemic juvenile idiopathic arthritis: Complications", section on 'Clinical presentation'.)
●Laboratory monitoring – The frequency of laboratory monitoring will depend on the severity of the disease. As an example, in patients with moderate to severe disease who are hospitalized, we trend labs daily, while those with mild disease who are being treated as outpatients might have a repeat evaluation after a week. In most patients, we obtain the following laboratory tests to monitor for ongoing disease activity and complications:
•Complete blood count (CBC) with differential
•Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP)
•Ferritin
•Kidney function (ie, blood urea nitrogen [BUN] and creatinine)
•Aspartate aminotransferase (AST) and alanine aminotransferase (ALT)
Other laboratory studies may be indicated depending on the clinical presentation. Many laboratory tests can take weeks to months to normalize once inflammation is controlled (eg, leukocytosis, anemia, ESR elevations). However, CRP levels typically rise and fall within a few days of acute inflammation. Thrombocytosis can sometimes worsen for a few weeks once inflammation is controlled. Interleukin 18 (IL-18) and CXCL9 levels may also be helpful, especially when assessing disease course, remission, and monitoring for development of MAS. However, these tests are not widely available and may take several weeks to result, limiting their clinical utility. More information about the assessment of the acute phase response and alterative biomarkers in MAS is provided separately. (See "Acute phase reactants", section on 'Laboratory evaluation' and "Systemic juvenile idiopathic arthritis: Complications", section on 'Evaluation and diagnostic testing'.)
When interpreting laboratory tests, it is important to recognize the potential for certain treatments to cause or mask certain laboratory abnormalities. As examples, systemic glucocorticoids commonly cause a neutrophilic leukocytosis, while IL-6 inhibitors can lower CRP and, to a lesser extent, ESR values [79]. In addition, some features of MAS may be more or less pronounced in patients treated with IL-1 or IL-6 inhibitors [80]. (See "Systemic juvenile idiopathic arthritis: Complications", section on 'Evaluation and diagnostic testing'.)
●Disease Activity Scores – Disease activity scores, such as the Juvenile Arthritis Disease Activity Score (JADAS), are useful for monitoring activity in clinical practice and research, and they can also help define treatment targets [81]. Researchers have developed a JADAS for sJIA and validated cutoff scores that define various levels of disease activity [82,83].
Drug therapy — Regular laboratory testing is required to monitor for treatment-related adverse effects. These are described in detail in the respective drug topics and summarized for the most common agents below:
●NSAIDs – Kidney function tests (ie, urinalysis, BUN, and creatinine), liver function tests, and hemoglobin at regular intervals (at least every three months).
●Methotrexate – AST, ALT, albumin, and a CBC every six to eight weeks until reaching the goal dose, then every three months or more frequently if abnormalities are identified. (See "Hepatotoxicity associated with chronic low-dose methotrexate for nonmalignant disease".)
●Biologic disease-modifying antirheumatic drugs (bDMARDs) – Annual rescreening for tuberculosis (TB) in patients with moderate to high risk of being exposed to TB. (See "Tuberculosis infection (latent tuberculosis) in children", section on 'Increased risk of new infection'.)
●IL-6 inhibitors – A lipid panel every four to eight weeks after initiation, then every six months. (See "Interleukin 6 inhibitors: Biology, principles of use, and adverse effects", section on 'Monitoring'.)
●JAK inhibitors – A CBC with differential, creatinine, and liver function tests every month for three months, then every three months; and a lipid panel six to eight weeks after starting, then periodically. (See "Overview of the Janus kinase inhibitors for rheumatologic and other inflammatory disorders", section on 'Monitoring'.)
In addition to the usual adverse effects associated with bDMARDs, there are several reports of patients with sJIA who develop features consistent with delayed-type hypersensitivity reactions and/or drug reaction with eosinophilia and systemic symptoms (DRESS) associated with IL-1 and IL-6 inhibitors. Clinicians should carefully monitor for signs of potential drug reaction, including significant eosinophilia, cutaneous reactions, and hepatitis, and if persistent consider alternative treatment strategies.
TAPERING THERAPY —
Approximately one third of patients with systemic juvenile idiopathic arthritis (sJIA) have a monophasic course of illness, which will remit completely after a period of months [6,7,32]. We therefore attempt to taper therapy once patients have achieved a period of remission based on their clinical symptoms and laboratory findings. (See 'Monitoring' above.)
The approach to tapering therapy depends upon what agents are being used:
●Patients taking nonsteroidal antiinflammatory drugs (NSAIDs) as monotherapy – If NSAID therapy alone is effective in controlling all of the sJIA manifestations for a period of several months, then we attempt a gradual taper of the NSAID (eg, naproxen can be decreased from twice daily to once a day for several weeks) before discontinuing it.
●Patients taking concomitant glucocorticoids – Whenever a patient is taking a systemic glucocorticoid with a steroid-sparing therapy (ie, a disease-modifying antirheumatic drug [DMARD]), the priority should be first tapering and then discontinuing the systemic glucocorticoid. This is because of the inevitable toxicity associated with chronic glucocorticoid exposure. The pace of the glucocorticoid taper will vary depending on how long the glucocorticoid-sparing therapy takes to become fully effective as well as the severity of sJIA, with more gradual tapers for patients taking conventional synthetic DMARDs (csDMARDs) versus biologic DMARDs (bDMARDs) and for those with severe versus mild disease.
●Patients taking DMARDs as monotherapy – While approaches to tapering DMARDs vary, most pediatric rheumatologists consider a taper once patients have been in remission on DMARDs and off of glucocorticoids for six months. For csDMARDs, the dose can be decreased and then discontinued over the course of several months. For bDMARDs, the dosing interval can be gradually increased before eventually stopping the medication.
There are no published guidelines to guide the tapering and withdrawal of DMARDs in sJIA. There is limited evidence from several observational series of patients treated with anakinra at disease onset [14,32] and clinical trials of canakinumab [84] and tocilizumab [85], with key findings summarized below:
•Anakinra – In one observational trial of 20 patients with sJIA who initiated treatment with anakinra, patients with inactive disease tapered the anakinra by changing from daily to alternate-day dosing over four weeks and then stopped therapy [14]. At one year, 52 percent had inactive disease off of medication; factors associated with this included a rapid complete response to therapy (within one month) and a high baseline neutrophil count (> 9x109/L). For the 42 percent of patients who flared off of therapy, the median time to flare after discontinuation of anakinra was five weeks, underscoring the importance of close follow-up when tapering treatment.
•Canakinumab – In a two-part study, 182 patients with sJIA received canakinumab 4 mg/kg every four weeks and, if needed, tapered off of other therapies (eg, methotrexate, glucocorticoids) [84]. The 75 patients who achieved inactive disease for ≥24 weeks on canakinumab monotherapy were then randomly assigned to a gradual three-step reduction in either the canakinumab dose (ie, from 4 mg/kg to 2 mg/kg, then to 1 mg/kg) or frequency (ie, from every 4 weeks to every 8 weeks, then to every 12 weeks) prior to discontinuing the medication. While 68 to 84 percent of patients met the primary endpoint (clinical remission for 24 weeks) after the first and second steps, only 33 percent remained in remission after discontinuing canakinumab. The percentages of patients meeting the primary endpoint during each step appeared to be similar between the two groups.
•Tocilizumab – In a clinical trial of tocilizumab in sJIA, patients who had undergone ≥2 years of treatment and who had achieved remission for ≥3 months could choose to gradually increase interval between tocilizumab infusions (from every two weeks to every three weeks, and then every four weeks) and discontinue therapy [85]. During the follow-up period, 20 of the 39 patients who participated had a disease flare; the time from initiating the tocilizumab taper to the disease flare was variable (1 to 28 months).
SPECIAL POPULATIONS
Comorbid sJIA lung disease — Patients who have systemic juvenile idiopathic arthritis-associated lung disease (sJIA-LD) may benefit from adding or substituting certain immunosuppressive therapies, particularly if there is evidence of persistent active sJIA and/or concern for a drug reaction. The approach to diagnosis and management of sJIA-LD is described elsewhere. (See "Systemic juvenile idiopathic arthritis: Complications", section on 'Lung disease'.)
Limited availability of biologic therapies — If patients do not have access to a biologic disease-modifying antirheumatic drug (bDMARD), initial therapy often involves a nonsteroidal antiinflammatory drug (NSAID) and, for those with moderate to severe symptoms, systemic glucocorticoids [17] and methotrexate.
Patients with refractory acute disease may benefit from a trial of thalidomide, as this medication is widely available and relatively inexpensive. Thalidomide suppresses the production of cytokines (eg, tumor necrosis factor [TNF]-alpha and interleukin-6 [IL-6]) [86,87] and blocks angiogenesis [88]. In children who are ≥3 years old, the initial dose of thalidomide is 2 mg/kg/day rounded to the nearest 50 mg dose; subsequently, the dose may be increased as needed at two-week intervals to a maximum of 3 to 5 mg/kg/day [89]. Common adverse effects include sedation and neuropathy, which are often dose-limiting. In addition, thalidomide is an infamous teratogen, which has led to strict regulation of its clinical use; effective contraception and monitoring for pregnancy are prerequisites for use in females of childbearing age.
Evidence to support the use of thalidomide in sJIA is limited to a small, prospective study of 13 patients with refractory sJIA; most patients improved within four weeks and 11 had a sustained response with adequate disease control [89]. Lenalidomide is an alternative, related agent that may have a lower incidence of side effects; however, there are no published data to support its use in sJIA.
In addition, there is a case report of the successful use of sirolimus in a patient with sJIA [90].
PROGNOSIS
Disease remission — Many patients with systemic juvenile idiopathic arthritis (sJIA) do respond well to therapy, especially when it is initiated early. Significant advances in the treatment of sJIA, particularly the use of biologic disease-modifying antirheumatic drugs (bDMARDs; eg, interleukin 6 [IL-6] and IL-1 antagonists) are thought to have dramatically improved the prognosis in children with sJIA. However, some patients continue to have persistent and resistant disease activity and may develop disease-related complications, including recurrent macrophage activation syndrome (MAS), joint damage and subsequent disability.
Based on multiple retrospective studies and two prospective studies, approximately 50 to 70 percent of patients with sJIA achieve remission in the one to five years following the diagnosis. [14,25,91-93]. As an example, in a large Canadian cohort of children diagnosed between 2005 and 2010, 71 percent of patients with sJIA attained clinically inactive disease within one year and 47 percent obtained remission within five years [91]. In a subsequent study of 73 patients with new-onset sJIA, most patients (86 percent) were taking a bDMARD at treatment induction and 62 percent had clinically inactive disease off of glucocorticoids after 12 months [25]. However, despite advances with bDMARDs, a significant proportion of patients have ongoing disease. In a study of 255 patients with sJIA who had been diagnosed at least two years prior, 56 percent had achieved remission, but almost a third had persistent arthritis despite the use of bDMARDs [93].
Multiple retrospective and prospective studies have suggested that there is a "window of opportunity" in which early, effective treatment may greatly improve the chances of remission [12,13,15,32]. One single-center prospective study of new-onset sJIA patients who were all treated with anakinra early in the disease showed that many (>70 percent) had medication-free remission at five years [94]. In addition, patients who have persistently active disease after six months from disease onset have a lower rate of disease remission, including those with persistent fevers, arthritis, elevated platelet counts, and/or continued requirements for glucocorticoid therapy [6,11].
Mortality — The overall mortality rate is low in children with sJIA, although higher rates are seen in children with more severe disease and in adults with a history of sJIA.
One study examined mortality outcomes in nearly 50,000 children with rheumatic disease [95]; the subset of 962 patients with sJIA did not have a significantly increased mortality rate, but this study may have missed hospitalized patients at highest risk of mortality. In two clinical trials of sJIA patients who mostly had severe, refractory disease, there were three deaths in 117 patients (two of which occurred after the trial period) and four deaths in 190 patients (two of which were after the trial period) in the trials of tocilizumab (IL-6 inhibitor) and canakinumab (IL-1 inhibitor), respectively [36,41].
In contrast, a high rate of mortality (up to 60 percent) is seen in patients with sJIA who develop lung disease (SJIA-LD) [5,66,96,97]. One retrospective series of 61 patients in 2019 reported a five-year survival rate of 43 percent [96], while a prospective single-center series from 2024 found a 7 percent mortality rate with a median follow-up of 2.9 years [66].
Long-term morbidity — Adults whose disease began during childhood may have residual problems, including musculoskeletal problems, early need for joint replacement therapy, and severe disability. The rates of long-term morbidities are expected to decrease with improvements in therapy, especially if treatment is started early in the disease course [13,32].
●Impact of bDMARDs – With more common and earlier usage of bDMARDs such as IL-1 and IL-6 inhibitors, complications like growth retardation and osteoporosis are less likely to occur, but long-term outcomes in the era of bDMARD treatment still need to be studied. As an example, the actual incidence of lung disease is unknown and should be studied systematically, such as in the context of a large disease registry.
Patients with sJIA who were diagnosed and managed prior to the availability of bDMARDs often have substantial disease- and treatment-related morbidity. In one study of 52 such adults with sJIA and a median disease duration of approximately 29 years, complications included severe disability (62 to 65 percent, depending on the measure used), restricted jaw opening (54 percent), leg-length discrepancies (31 percent), and amyloidosis (nearly 20 percent) [98].
Subsequent studies of patients diagnosed and treated after widespread availability of bDMARDs have found improved outcomes, although approximately 15 percent still had articular damage and some degree of disability [99,100]. These results illustrate the importance of initiating effective disease-modifying therapy in order to minimize long-term morbidity and disability for children with sJIA.
●Joint damage – Historically, more than one in three children with sJIA developed permanent joint damage [101], especially in the hip joint, and more than one-half required substantial prolonged systemic glucocorticoid therapy (often for years) [102]. Two subsequent long-term outcome studies of children from Canada [99] and from Nordic countries [100] diagnosed and treated in the bDMARD era found that approximately 15 percent of children with sJIA had articular damage at last follow-up.
Several cohort studies have noted higher rates of destructive arthritis in patients who have persistent fevers, arthritis, elevated platelet counts, and/or continued requirements for glucocorticoid therapy after six months from their initial diagnosis [10,11]. In addition, one study suggested that children who develop sJIA at a very early age (before 18 months) may have worse outcomes with respect to destructive arthritis and disability [103].
●Osteoporosis – Osteoporosis in children with sJIA due to disease and treatment was also more common prior to recent therapeutic advances, which have allowed for reduced cumulative dosing of systemic glucocorticoids. The optimal way to minimize this complication is to attain disease control with a minimum of glucocorticoid exposure. (See "Major adverse effects of systemic glucocorticoids", section on 'Bone and muscle effects'.)
In patients with sJIA who develop severe osteoporosis, bisphosphonates may be an option for treatment to try to prevent recurrent insufficiency fractures. However, the safety of bisphosphonate use by children remains uncertain, especially as these medications are stored in and slowly released in the bones for years [104]. Bisphosphonates have been successfully used in children with rare bone disorders, including osteogenesis imperfecta. Other measures to optimize bone health in adolescents are discussed separately. (See "Bone health and calcium requirements in adolescents".)
Other complications that may occur in various forms of JIA, such as leg-length discrepancy and short stature, are discussed in detail separately. (See "Juvenile idiopathic arthritis: Immunizations and complications", section on 'Complications'.)
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: Juvenile idiopathic arthritis".)
SUMMARY AND RECOMMENDATIONS
●Urgency of treatment – We treat patients diagnosed with systemic juvenile idiopathic arthritis (sJIA) immediately upon diagnosis as they can develop severe, potentially life-threatening complications (eg, severe serositis and/or macrophage activation syndrome [MAS]). In addition, there may be a "window of opportunity" in which early, effective treatment may greatly improve the chance of remission. (See 'Urgency of treatment' above and "Systemic juvenile idiopathic arthritis: Complications", section on 'Macrophage activation syndrome'.)
●Goals of therapy – The goals of therapy in sJIA include rapidly control symptoms to minimize the risk of complications from persistent uncontrolled disease, as well as judicious use of immunosuppression to minimize the risk of adverse effects from therapy, especially those related to chronic systemic glucocorticoids. (See 'Goals of therapy' above.)
●When to refer – All patients with sJIA should be referred to an experienced pediatric rheumatologist to confirm the diagnosis and start treatment. (See 'When to refer' above.)
●Initial treatment of patients without MAS – In patients with sJIA who do not have concomitant MAS, the choice of initial therapy depends upon the severity of symptoms. Our general approach is outlined in the algorithm (algorithm 1) and drug dosing is provided in the table (table 1). (See 'Patients without macrophage activation syndrome' above and 'Approach to initial therapy' above.)
•Nonsteroidal antiinflammatory drugs (NSAIDs) for mild disease – In most patients with mild sJIA, we suggest a trial of monotherapy with an NSAID rather than a systemic glucocorticoid or disease-modifying antirheumatic drug (DMARD) (Grade 2C). Any NSAID other than aspirin may be used at standard antiinflammatory doses. (See 'Mild disease' above.)
•Interleukin 1 (IL-1) or IL-6 inhibitors for moderate to severe disease – In most patients with moderate to severe sJIA, we suggest starting an IL-1 or an IL-6 inhibitor rather than other types of DMARDs (Grade 2C). Patients may also receive an adjunctive NSAID if needed for fever and pain relief. (See 'Moderate to severe disease' above.)
•Concomitant glucocorticoids in selected patients – We add a systemic glucocorticoid to a glucocorticoid-sparing agent in certain clinical scenarios (eg, severe systemic symptoms, relatively severe systemic symptoms with suspected early MAS, or progressive symptoms despite initial therapy). Ideally, the glucocorticoid is limited to a dose of ≤0.5 mg/kg/day of prednisone (or its equivalent) for ≤6 months; however, patients with severe manifestations may require higher doses. (See 'Indications for glucocorticoids' above.)
●Subsequent therapy
•Persistent acute disease – We escalate therapy when patients have persistent or worsening symptoms or laboratory abnormalities despite one to two weeks of initial therapy. The approach to escalating therapy depends upon the initial treatment and specific disease manifestations (algorithm 1). (See 'Persistent initial disease' above.)
-Patients taking NSAID monotherapy – For such patients, we suggest adding an IL-1 or IL-6 inhibitor, rather than adding a systemic glucocorticoid alone or adding a conventional synthetic DMARD (csDMARD) (Grade 2C).
-Patients taking IL-1 or IL-6 inhibitors – For such patients, we suggest switching the type of IL-1 or IL-6 inhibitor, rather than adding a systemic glucocorticoid alone or adding a csDMARD (Grade 2C). An alternative option in patients with persistent arthritis is to add a csDMARD.
Adjunctive systemic glucocorticoids may be required for severe disease manifestations. (See 'Indications for glucocorticoids' above.)
•Refractory chronic disease – We consider patients to have refractory chronic disease if after ≥6 months they have persistent, worsening, or recurrent systemic and/or arthritic symptoms despite separate trials of an IL-1 and an IL-6 inhibitor and/or if they have been unable to taper off of systemic glucocorticoids. The optimal type of therapy will depend on the predominant clinical features (ie, systemic symptoms and/or arthritis). (See 'Refractory chronic disease' above.)
●Tapering therapy – The approach to tapering therapy varies based on the specific therapies being used. We prioritize tapering and discontinuing systemic glucocorticoids over DMARDs. (See 'Tapering therapy' above.)
●Prognosis
•Disease remission – Approximately 50 to 70 percent of patients with sJIA achieve remission in the one to five years following the diagnosis. However, despite advances with bDMARDs, a significant proportion of patients have ongoing disease and may develop disease-related complications, including recurrent MAS, joint damage, and subsequent disability. (See 'Disease remission' above and 'Long-term morbidity' above.)
•Mortality – The overall mortality rate is low in children with sJIA, although higher rates are seen in children with more severe disease and in adults with a history of sJIA. By contrast, a high rate of mortality (up to 60 percent) is seen in patients who develop sJIA-associated lung disease (sJIA-LD). (See 'Mortality' above.)
ACKNOWLEDGMENT —
The UpToDate editorial staff acknowledges Thomas JA Lehman, MD, who contributed to earlier versions of this topic review.
