Polyarticular juvenile idiopathic arthritis: Treatment

INTRODUCTION — Polyarticular juvenile idiopathic arthritis (formerly called polyarticular-onset juvenile rheumatoid arthritis [RA]) is a subset of juvenile idiopathic arthritis (JIA) that is defined by the presence of more than four affected joints during the first six months of illness [1]. In the revised nomenclature, this disease, which comprises 20 to 30 percent of patients with JIA, is included in the group termed "childhood polyarthritis." (See "Classification of juvenile idiopathic arthritis".)

Therapy is directed toward treating the underlying inflammation of JIA and preventing complications associated with JIA (eg, joint damage) or the adverse effects of its treatment [2]. Medications used in the treatment of polyarticular JIA have various mechanisms of action, some of which are unknown, but all suppress inflammation. First-line agents include nonbiologic disease-modifying antirheumatic drugs (DMARDs) such as methotrexate. Second-line drugs referred to as biologic DMARDs (bDMARDs) include anti-tumor necrosis factor (TNF) alpha agents.

Although many parents/caregivers and clinicians voice fear of potential side effects from medications, years of experience make it clear that there is far more to fear from the damage known to be caused by ongoing joint inflammation. The belief statement that children often "grow out of JIA" is incorrect and may lead to delays in appropriate therapy and poor outcomes. It is essential that clinicians make parents/caregivers aware of the fact that children with polyarticular JIA are unlikely to outgrow their disease since this myth may heavily influence their understanding of the risk-to-benefit ratio of medications.

The treatment of polyarticular JIA is reviewed here. The clinical manifestations, diagnosis, complications, and prognosis of polyarticular JIA are discussed separately. (See "Polyarticular juvenile idiopathic arthritis: Clinical manifestations, diagnosis, and complications".)

GOALS OF TREATMENT — Advances in the therapeutic options available to treat polyarticular JIA have made remission a realistic goal for most children. An international task force recommended remission as the treatment goal with emphasis on assuring normal growth and development and avoidance of long-term systemic glucocorticoids [3]. In one single-center study, implementation of structured disease activity monitoring with the clinical Juvenile Arthritis Disease Activity Score (cJADAS-10; score = active joint count [0 to 10] + physician global disease activity score [0 to 10] + caregiver/patient global well-being score [0 to 10]) in the setting of clinical decision support resulted in significant improvements in disease activity in both incident and prevalent cases of disease [4].

MANAGEMENT APPROACH — The approach to the treatment of polyarticular JIA is reviewed here and is consistent with the approach outlined in the 2019 American College of Rheumatology (ACR)/Arthritis Foundation (AF) guideline for therapeutic approaches for JIA [2]. Treatment of polyarticular JIA is individualized based upon disease activity and severity, as well as patient preferences, after reviewing the risks, benefits, and costs of the therapies. The specific agents are discussed in detail separately or in the sections below. (See "NSAIDs (including aspirin): Pharmacology and mechanism of action" and 'Methotrexate' below and 'Tumor necrosis factor inhibitors' below.)

Pretreatment assessment — Patients are assessed for the presence of features associated with poor prognosis, which include [2]:

●Cervical or hip arthritis

●Rheumatoid factor [RF] positivity

●Cyclic citrullinated peptide [CCP] positivity

●Radiographic evidence of joint damage (erosions or joint space narrowing)

The updated ACR/AF guidelines do not differentiate initial treatment recommendations for those with low versus moderate/high disease activity (defined as a clinical Juvenile Arthritis Disease Activity Score [cJADAS] of ≤2.5 or >2.5, respectively), with the exception of the option of a limited course of oral glucocorticoids as adjunct therapy for those with moderate/high disease activity and significant impairment [2].

Initial management — Initial drug therapy for children with polyarticular JIA should be aggressive in order to control the inflammatory process and relieve symptoms as quickly as possible while minimizing drug side effects. A disease-modifying antirheumatic drug (DMARD) is recommended for all patients at the time of diagnosis since it is rare for nonsteroidal antiinflammatory drugs (NSAIDs) alone to control the inflammatory process of polyarticular disease, and chronic joint changes may occur if arthritis remains active [5]. NSAIDs are appropriate for children with suspected disease awaiting rheumatology evaluation or as adjunctive therapy for symptom management.

DMARDs used to treat polyarticular JIA include the nonbiologic agents (eg, methotrexate) and the biologic agents (eg, tumor necrosis factor [TNF] inhibitors). Methotrexate is the traditional DMARD of choice. Indications for using a biologic agent in addition to or rather than methotrexate include presence of severe polyarthritis, poor prognostic features, or factors associated with poor response to methotrexate (eg, predominantly axial arthritis).

For children with high or moderate disease activity (cJADAS of >2.5), adjunctive therapy with oral glucocorticoids (total duration of less than three months) is conditionally recommended for those with significant discomfort or mobility impairment. Chronic low-dose glucocorticoid use is not recommended for children, regardless of disease activity or risk factors.

Escalation of therapy is recommended in children with continued disease activity in at least one joint after three months of treatment [2] and may include any of the following:

●Intraarticular glucocorticoid injections.

●Changing the route of administration (to subcutaneous [SC] if receiving oral formulation) or dose (if not at maximal dose) of methotrexate.

●Adding a biologic agent (if only on methotrexate).

●Changing to an alternate biologic agent or a Janus kinase (JAK) inhibitor (if already on a biologic agent). If a patient has failed a TNF inhibitor, then switching to a non-TNF inhibitor biologic (eg, tocilizumab, abatacept) or a JAK inhibitor (eg, tofacitinib or possibly baricitinib) is preferred over trying a different TNF inhibitor.

Of the DMARDs, methotrexate has been the standard therapy for children with polyarticular disease [2,5-9]. The usual starting dose of methotrexate is 10 to 15 mg/m² body surface area (BSA) per week. Methotrexate is effective for many patients, but it does not work quickly, usually taking four to eight weeks before demonstrating its benefits. The trial periods for methotrexate are customary because of the higher risks and costs associated with the biologic agents compared with methotrexate. However, biologic agents have a more rapid onset of action and appear to be more efficacious than methotrexate. Etanercept is the first of the TNF inhibitors approved for clinical use, with a relatively large amount of data regarding long-term safety and efficacy. Standard dosing for etanercept is 0.8 mg/kg once a week or 0.4 mg/kg twice a week. Other TNF inhibitors such as adalimumab and golimumab have since been approved for use in polyarticular JIA. There are no data to indicate that one TNF inhibitor is more efficacious than another for polyarticular JIA. (See 'Tumor necrosis factor inhibitors' below and 'Methotrexate' below.)

The question as to whether TNF inhibitors, with their potentially faster onset of action and greater efficacy, should be used before methotrexate remains unresolved. There is also evidence that the combination of a TNF inhibitor and methotrexate is synergistic for the treatment of arthritis [5,10-15]. These considerations must be balanced against the potential increased risk of opportunistic infections with the use of TNF inhibitors, especially when used as part of multiple drug therapy. Neither TNF inhibitor monotherapy nor TNF inhibitor plus methotrexate combination therapy increased the risk of infections during or resulting in hospitalization compared with methotrexate monotherapy in a large administrative database study of children with JIA [16]. In addition, there is not yet a clear answer to the question of whether benefits of biologic agents justify their significantly higher costs [17,18]. Ongoing studies aimed at identifying markers of response and risk factors for adverse events should improve selection of appropriate candidates for the use of biologic agents and help answer and resolve some of these important issues. (See 'Methotrexate' below.)

Many pediatric rheumatologists start with a biologic agent before or concurrent with methotrexate in children with high disease activity or poor prognostic features (see 'Pretreatment assessment' above) if there are no insurance restrictions with regard to starting a biologic before a trial of methotrexate [2]. Of the biologics, a TNF inhibitor is typically started first, although there is no evidence to guide choice of one biologic over another early in the disease course. In addition, children with longer disease duration, negative antinuclear antibody (ANA) titers, higher level of disability, and arthritis present in both wrists are less likely to respond to methotrexate alone [19], and methotrexate is ineffective in predominantly axial arthritis.

Some centers are using biologic agents at the start of methotrexate therapy, with the goal of preventing additional joint damage during the period necessary for methotrexate to take effect. In a 2013 survey of 138 pediatric rheumatologists, 46 and 9 percent of providers stated that they would use a nonbiologic and a biologic DMARD as initial therapy for children with polyarticular arthritis with and without poor prognostic features, respectively, regardless of disease activity [20]. Less than 1 percent would start a biologic agent alone for polyarticular disease. Labeling indications for the approved anti-TNF therapies (adalimumab, etanercept, golimumab), abatacept, tocilizumab, and the kinase inhibitor tofacitinib do not require failure to respond to methotrexate before their use in children with JIA. All of the biologic agents can be used as monotherapy or in combination with a DMARD such as methotrexate. However, combination with other biologics, kinase inhibitors, or other immunosuppressants such as cyclosporine or azathioprine is not recommended. Ongoing trials should help clarify the optimal approach to new-onset polyarthritis in children [21]. (See 'Biologic agents' below.)

Additional therapies — Additional modes of treatment for polyarticular JIA include NSAIDs, intraarticular glucocorticoids, and physical/occupational therapy.

Nonsteroidal antiinflammatory drugs — An NSAID such as ibuprofen (10 mg/kg every eight hours, maximum dose 3200 mg/day), naproxen (10 to 15 mg/kg/day in two divided doses, maximum dose 1000 mg/day), or indomethacin (1 to 2 mg/kg/day in two to three divided doses, maximum dose 200 mg/day) may help to provide symptomatic relief [2,22,23]. NSAIDs reach full efficacy within two to three months but usually start to relieve symptoms within a few days [22].

Unfortunately, there is no way to predict which NSAID is most likely to benefit an individual patient. A process of trial and error must be used while keeping expectations appropriate. Patients should be aware that NSAIDs are useful for symptomatic relief of pain and stiffness, but they do not delay or prevent joint damage in JIA. Long-term safety data on NSAIDs are needed because potential adverse effects, especially cardiovascular events, may not appear until adulthood [24]. However, an email survey of the pediatric rheumatology community found no recognized cases of children treated with NSAIDs who developed premature myocardial infarctions.

Intraarticular glucocorticoids — Intraarticular injections of glucocorticoids are both safe and effective for children with polyarticular JIA, and their use is recommended as needed for continued disease activity [25-27]. However, the need to repeatedly inject the same joint, or to inject multiple joints simultaneously, suggests that more effective systemic therapy is required. With the availability of anti-TNF agents, non-TNF inhibitor biologics, or kinase inhibitors, the simultaneous injection of multiple joints is rarely necessary.

Injection of intraarticular glucocorticoids into a single inflamed joint, such as the hip or temporomandibular joint, is often dramatically effective [28-30]. The injections are performed under general anesthesia in young children, or local anesthesia in older children, with computed tomography or ultrasound-guided imaging, or without imaging by an experienced clinician, depending upon the joint injected. (See "Joint aspiration or injection in children: Indications, technique, and complications".)

Folic acid — Folic acid supplementation (400 micrograms daily) is recommended for all children with JIA receiving methotrexate, given the beneficial effects noted and the lack of convincing contrary data [5]. If the child experiences methotrexate side effects on 400 micrograms of folic acid daily, the dose of folic acid is typically increased to 1 mg/day. Some side effects such as glossitis and macrocytic anemia are often avoided by the addition of supplemental folic acid, started at the same time as methotrexate. Whether high doses of folic acid also diminish the effectiveness of methotrexate or cause side effects are topics of debate. Published articles report conflicting results regarding potential antagonism of the beneficial effects of methotrexate by folate supplementation [31]. Similarly, studies have shown both an increase [32] and a decrease [33] in the incidence of cancer among adults receiving folic acid supplementation. The effects appear to depend upon a variety of factors, including characteristics of the specific population studied [34]. However, no adverse effects have been demonstrated to date in children receiving supplemental folic acid to minimize methotrexate toxicity. Leucovorin (folinic acid; usual dose 5 mg once a week) is an alternative to folic acid.

Physical and occupational therapy — Physical therapy plays an important role in the rehabilitation of any child with polyarticular disease. However, range of motion and stretching exercises are of limited benefit in the face of ongoing joint inflammation. In addition, the efficacy and availability of the biologic agents have allowed many children to regain full function in a short period of time, limiting the need for rehabilitation. Nonetheless, any child with persistent weakness or flexion contractures should be referred for physical and/or occupational therapy. Physical therapy typically targets the larger joints and lower-extremity joints needed for movement while occupational therapy targets the ability to perform activities of daily living (eg, writing in school).

Ongoing management — Ongoing management includes monitoring for drug side effects, trial off medications at some point after inactive disease is attained, and treatment of refractory or recurrent disease. Inactive disease is defined as meeting the following criteria [35]:

●No active arthritis

●No fever, rash, serositis, splenomegaly, or lymphadenopathy attributable to JIA

●No active uveitis

●Normal inflammatory markers

●Global assessment of disease activity by clinician indicates no activity

Monitoring on DMARD therapy — Monitoring of liver enzymes, creatinine, and complete blood counts (CBCs) is recommended prior to methotrexate initiation, four to eight weeks after initiation, and every three to four months thereafter [2,5]. Supplemental folic acid or leucovorin (folinic acid) is given to minimize methotrexate toxicity. (See 'Methotrexate' below and 'Folic acid' above.)

A CBC and comprehensive metabolic panel (CMP) are recommended prior to initiation of all biologics and kinase inhibitors. A baseline lipid panel is also recommended for tocilizumab and kinase inhibitors. Labs (CBC and CMP) should be checked again four to eight weeks after therapy starts and then every three months thereafter. Lipids should be rechecked four to eight weeks after starting tocilizumab or a kinase inhibitor and then periodically. Skin testing for tuberculosis prior to initiation of biologics and kinase inhibitor therapy is standard. (See 'Tumor necrosis factor inhibitors' below.)

Duration of therapy — The amount of time that methotrexate [5] and biologics should be continued in children with polyarticular disease is unclear since clinical outcomes following medication withdrawal have not been studied prospectively. Medications are typically tapered off after a period of clinical inactivity between six months and two years while on medication. The decision regarding which medication to stop first is individualized depending upon response to the medication, tolerance of the administration regimen, and other patient factors.

In a large, retrospective study of 171 children with juvenile arthritis, the median time to biologic discontinuation after attainment of inactive disease was six months (range 0 to 69 months) [36]. In this cohort, one-half and two-thirds had relapse of arthritis by 6 and 12 months, respectively. There was also no significant association in this study between the duration of therapy after inactive disease was achieved or total duration of biologic therapy prior to discontinuation and time to disease relapse.

In a 2015 survey of North American pediatric rheumatologists, 75 percent of participants stopped methotrexate or a biologic 6 to 12 months after attainment of inactive disease [37]. Sixty-three percent stop methotrexate before a biologic. RF positivity made clinicians less likely to withdraw therapy.

Refractory disease — Patients are considered to have refractory disease if they have persistent disease activity following four months of treatment with a TNF inhibitor and methotrexate. A trial of an alternate TNF inhibitor agent or another biologic agent (tocilizumab or abatacept) or kinase inhibitor (tofacitinib or possibly baricitinib) is indicated in these patients [2,38,39].

Recurrent disease — Relapse is common within 6 to 12 months of withdrawal of therapy after remission. Patients are restarted on their previous regimen. The regimen is altered if they fail to recapture disease inactivity on medication.

Immunizations — The administration of live-viral vaccines and other standard childhood immunizations in patients with JIA is discussed in detail separately. (See "Oligoarticular juvenile idiopathic arthritis", section on 'Immunizations'.)

Rheumatoid factor-positive patients — Adolescents with RF-positive polyarthritis probably represent the early onset of adult-type disease and are at significant long-term risk for progressive arthritis and the other complications normally associated with adult-onset rheumatoid arthritis (RA) [40,41]. Thus, these children should be treated similarly to adults with moderate-to-severe RA. (See "Initial treatment of rheumatoid arthritis in adults".)

Joint damage in RF-positive polyarticular JIA occurs earlier, leading many investigators to believe that the therapeutic pyramid should be inverted, with early and aggressive treatment of these patients. Studies of early, aggressive therapy in adults with RA confirm better responses and improved joint preservation [42-44]. Ongoing therapeutic trials in children are attempting to determine whether a similar approach involving treatment with disease-modifying agents should be followed in JIA patients at high risk for progressive joint destruction.

DISEASE-MODIFYING ANTIRHEUMATIC DRUGS — The primary disease-modifying antirheumatic drugs (DMARDs) used for polyarticular JIA are methotrexate and the tumor necrosis factor (TNF) inhibitors. The other biologic agents and kinase inhibitors are usually reserved for patients with refractory disease who have failed therapy with a TNF inhibitor. Several of the older agents, such as cyclosporine, azathioprine, and systemic glucocorticoids, are rarely used with the advent of more effective agents.

Methotrexate — Methotrexate is an immunomodulator and acts as an inhibitor of purine synthesis, although other mechanisms of action appear to account for its effects at the doses used to treat JIA [5,45]. At these doses (typically 10 to 15 mg/m²/week, maximum dose 25 mg), methotrexate acts as an antiinflammatory agent rather than as a cytotoxic drug.

Efficacy — The question of whether treatment of early polyarticular JIA with either methotrexate alone or methotrexate in combination with a biologic agent and systemic glucocorticoid is effective was examined in a randomized trial of 85 children aged 2 to 16 years of age [10]. Children were randomly assigned to combination therapy (n = 42) with methotrexate (0.5 mg/kg/week subcutaneously [SC], maximum dose 40 mg); etanercept (0.8 mg/kg/week SC, maximum dose 50 mg); and prednisolone (0.5 mg/kg/day orally, maximum dose 60 mg, tapered off by 17 weeks) or methotrexate alone (n = 43) with etanercept and prednisolone placebos. Both regimens were moderately effective, although there was a trend toward greater efficacy in the initial combination group compared with methotrexate monotherapy that was significant only at the six-month time point. At four months, 30 (71 percent) children in the combination-therapy group had inactive disease compared with 19 (44 percent) in the methotrexate-only group. By six months, only 17 (40 percent) and 9 (21 percent) had inactive disease in the combination and methotrexate groups, respectively. At 12 months, 9 of the 17 children in the combination group had attained clinical remission, and the remainder had inactive disease. In comparison, three children were in clinical remission and four had inactive disease at 12 months in the methotrexate group. Five children in the combination group withdrew over the course of the study: three due to adverse events, one because of an inadequate response, and one due to a disease flare. Eleven children withdrew from the methotrexate group: four due to an adverse event, five because of an inadequate response, one due to a disease flare, and one due to a fear of needles.

Methotrexate is generally ineffective for the treatment of axial arthritis, as may be found in nonradiographic axial spondyloarthritis or ankylosing spondylitis [46]. Thus, it is important to recognize that children with spondyloarthritis are distinct from children with polyarthritis so that they may be advanced to appropriate biologic therapy without the delay and possible toxicity associated with a trial of methotrexate. (See "Spondyloarthritis in children".)

Duration of treatment — Treatment is usually withdrawn 6 to 12 months after attaining inactive disease. However, the relapse rate is high, and there is risk of ongoing joint damage with each relapse. This raises the question of whether the appropriate response is to continue therapy rather than withdraw it.

An open, randomized trial of methotrexate withdrawal after remission found no difference in relapse rates between 12- versus 6-month withdrawal [47]. Both groups experienced a roughly 60 percent rate of relapse, although patients with higher levels of the phagocyte activation marker myeloid-related proteins 8 and 14 heterocomplex (MRP8/14) were at greater risk of relapse after discontinuation of methotrexate. These tests are not routinely available and hence of limited utility. Further studies are needed to identify biomarkers predicative of future relapse.

In a multicenter prospective study of TNF inhibitor withdrawal in children with polyarticular JIA and inactive disease for at least six months, 40 percent of children flared within eight months, with a median time to flare of approximately seven months [48]. Significant predictors of flare included longer duration of disease at enrollment, younger age at diagnosis, and longer disease duration prior to the first episode of inactive disease. In another study evaluating etanercept withdrawal in children with a rheumatoid factor (RF) negative polyarticular course and inactive disease for at least six months, 60 percent of patients flared a median of 4.3 months after the last dose of medication [49]. In this study, only male sex and elevated baseline C-reactive protein (CRP) were significantly associated with flare.

Adverse effects — Potential side effects of methotrexate are many. Any evidence of significant toxicity requires a dose decrease or discontinuation of methotrexate. The more common side effects include leukopenia or thrombocytopenia and elevation of hepatic transaminases (especially if ethanol is consumed). Methotrexate is teratogenic, so safe-sex counseling is recommended for sexually active adolescents. Methotrexate may also cause nausea and vomiting within 24 to 48 hours of administration. Folic acid or leucovorin (folinic acid) supplementation (see 'Folic acid' above), administration of methotrexate on an empty stomach, and/or dose adjustment can help with the gastrointestinal side effects.

Rarer adverse effects include the development of pulmonary hypersensitivity reactions, nodulosis [50], and, possibly, malignancies [51]. In practice, however, dose-limiting side effects are rare [52]. Similarly, although long-term data in children are limited, meta-analyses in adults suggest that the chronic use of methotrexate in rheumatoid arthritis (RA) is associated with a minimally increased risk of infections, no increase in the incidence of opportunistic infections, no clear increase in the risk of malignancies, and an apparent advantage in longevity and decreased cardiovascular mortality [53].

A retrospective study examined the safety and efficacy of methotrexate regimens at two medical centers in the same geographic region, one of which started patients on an initial dose of >0.5 mg/kg/week (similar, but not identical to 15 mg/m²) and the other which began patients on ≤0.5 mg/kg/week (similar to 10 mg/m²) [54]. The high-dose group was more likely to have elevated liver enzymes but was no more likely to have a decreased active joint count.

Genotyping may help predict the risk of methotrexate toxicity. Single-nucleotide polymorphisms (SNPs) in genes within the methotrexate metabolic pathway, such as the C677T allele of the methylenetetrahydrofolate reductase gene, are associated with increased methotrexate-adverse effects [55,56]. However, these tests are not routinely available.

Biologic agents — The most common biologic agents used to treat polyarticular JIA are the TNF inhibitors (eg, etanercept, adalimumab, infliximab, golimumab). Other biologic agents act by blocking the interleukin (IL) 6 receptor, T cell costimulation, or CD20.

Tumor necrosis factor inhibitors — TNF inhibitors, including etanercept, adalimumab, golimumab, and infliximab, often provide rapid control of otherwise disabling arthritis, restoring quality of life and preventing further joint destruction. However, they are associated with an increased risk of infections [57,58]. There were initial concerns about a possible association between treatment with TNF inhibitors and malignancy, but subsequent data have suggested that the risk of malignancy is associated with underlying JIA rather than the medications [59]. These potential side effects are reviewed in greater detail separately. (See "Overview of biologic agents in the rheumatic diseases" and "Tumor necrosis factor-alpha inhibitors: Risk of malignancy" and "Systemic juvenile idiopathic arthritis: Treatment", section on 'Biologic disease-modifying antirheumatic drug (DMARD) therapy' and "Systemic juvenile idiopathic arthritis: Treatment", section on 'Tumor necrosis factor alpha inhibitors'.)

In adults, TNF inhibitors increase the risk of opportunistic infections [60]. In children, there is also an increased risk of fungal infections, especially in endemic areas of histoplasmosis such as the Ohio and Mississippi River valleys, with the use of anti-TNF agents [61]. Clinicians in areas with endemic fungal infections should monitor their patients appropriately. (See "Pathogenesis and clinical manifestations of disseminated histoplasmosis" and "Pathogenesis and clinical features of pulmonary histoplasmosis".)

Etanercept — Etanercept, a fusion protein containing TNF receptor 2 and the Fc domain of immunoglobulin G1 (IgG1), is effective in many children with resistant polyarticular disease, including those under four years of age [11,25,62-64]. Dramatic improvements may be seen within weeks of starting therapy, with benefits persisting during years of use. Most patients respond promptly, although some may have a delayed clinical response [65]. The available data favor the use of biologics that block TNF compared with other therapeutic regimens in children with refractory polyarticular JIA [66-68]. Standard SC dosing is 0.8 mg/kg once a week or 0.4 mg/kg twice a week.

As the first of the TNF inhibitors in clinical use, there is a relatively large amount of data regarding the long-term safety and efficacy of etanercept [67]. In patients who remain on therapy, the clinical benefits are substantial, and significant toxicity is uncommon. Symptoms improve in a sizable proportion of patients with refractory disease [62]. Growth also improves [69]. No cases of malignancy have been reported in these follow-up studies of children treated with etanercept for JIA. There have been case reports of inflammatory bowel disease (IBD) in JIA patients treated with etanercept, although these children might well have had unrecognized IBD at the time they were diagnosed with polyarthritis rather than bowel disease caused by etanercept [70,71].

In one open-label study of 69 children who failed therapy with methotrexate, for example, the administration of etanercept at a dose of 0.4 mg/kg twice a week resulted in a response (defined as >30 percent improvement) in nearly 75 percent of patients within three months [72]. A flare subsequently occurred in 28 percent of those continuing to receive etanercept versus 81 percent in whom placebo was substituted in a randomized and blinded fashion. Thus, in most children, the effects of etanercept cease within a few weeks of its discontinuation.

Subsequent studies reviewed the safety and efficacy of long-term etanercept therapy in the 58 patients who elected to continue therapy after the initial phase of the study [73-75]. At the four-year follow-up, 34 of the 69 original patients remained on therapy, whereas only 20 remained at the eight-year follow-up. Not surprisingly, those who remained on therapy had a significant improvement in symptoms and disease activity. Most patients who still required treatment with glucocorticoids were able to taper to a lower dose. Severe infections were the most common serious adverse event reported in the first four years, while disease flares were the most common in the second four years of treatment.

In another small pilot study, bone mineral status improved in patients who had responded to and continued etanercept treatment for more than one year compared with those who failed to respond [76]. In addition, anti-TNF therapy appears to improve growth velocity in patients with delayed growth [77,78]. Improved linear growth, bone mineralization, and body composition were seen in another small study [79]. These effects are probably due to decreased inflammation resulting from control of arthritis. In general, similar effects to those of etanercept should be expected from other TNF antagonists, such as adalimumab and infliximab.

Adalimumab — Adalimumab is a fully humanized monoclonal anti-TNF antibody that is administered either weekly or every other week as a single SC injection. Standard dosing is 10 mg, 20 mg, or 40 mg every other week for children less than 15 kg, 15 to 30 kg, and 30 kg or greater, respectively. Adalimumab is effective in some children who have responded inadequately to etanercept.

In 2008, adalimumab was approved by the US Food and Drug Administration (FDA) for the treatment of moderate-to-severe JIA in patients ≥4 years of age. Approval was based upon data from a trial of 171 patients who were stratified into two groups: those treated with methotrexate and those not treated with methotrexate [80]. After an open-label lead-in phase of 16 weeks with adalimumab therapy, patients with an American College of Rheumatology Pediatric (ACR Pedi) 30 response were randomly selected to receive adalimumab at a dose of 24 mg/m² every other week or placebo for an additional 32 weeks. In the second phase of the study, patients who received adalimumab, compared with those who received placebo, experienced fewer disease flares regardless of whether they received methotrexate (37 versus 65 percent) or not (43 versus 71 percent). In this trial, severe adverse effects included neutropenia and infections (herpes simplex and zoster, urinary tract infection, pneumonia, and pharyngitis). No malignancies were reported. In 2014, the indication was extended to include children ages two to four years of age with moderate-to-severe polyarticular JIA [81].

In the 360-week long-term extension study, adalimumab remained effective and well tolerated with or without concomitant methotrexate therapy [82]. Of the 171 patients enrolled into the phase III trial [80], 62 (36 percent) completed the long-term extension study. Thirty-seven percent attained clinical remission for more than six months during the study. Seventy-three percent (observed) achieved a 27-joint Juvenile Arthritis Disease Activity Score (JADAS27) low disease activity assessment (44 percent by nonresponder imputation). Clinical response was typically sustained through week 312. Two-hundred thirty-five adverse events and 3.2 serious adverse events per 100 person-years were reported as possibly related to adalimumab exposure; incidence of adverse events and serious adverse events were similar among those receiving and not receiving concomitant methotrexate. Serious adverse events included herpes zoster, pharyngitis, genital herpes, and pneumonia. No malignancies, tuberculosis, or lupus-like syndromes were reported.

Golimumab — Golimumab is a fully human monoclonal antibody with affinity for TNF-alpha that is available in both intravenous (IV) and SC formulations. The IV formulation (80 mg/m² dosed at weeks 0 and 4 and then every eight weeks) was approved by the US FDA in 2020 for the treatment of polyarticular JIA based upon extrapolation of efficacy data from adults with RA and a small, open-label study in children with polyarticular JIA who had not responded to methotrexate alone [83]. One-hundred twenty-seven children who had persistently active disease despite at least two months of methotrexate monotherapy were treated with open-label golimumab (80 mg/m²) at weeks 0 and 4, and then every eight weeks thereafter through week 52. At week 28, 47, 70, and 80 percent achieved a JIA ACR 90, 70, or 50, respectively. Serious adverse events include infection (6 percent of patients) and death (one due to septic shock).

SC golimumab has also been studied in children with polyarticular JIA who are refractory to methotrexate [84]. During the 16-week open-label golimumab lead-in period, subjects demonstrated a rapid improvement in disease activity, with 34 percent of subjects attaining inactive disease. At the end of this period, patients were randomly assigned to continue golimumab or start placebo. At week 48, the treatment and placebo groups had comparable rates of disease flares and clinical remission. The reasons for failure to reach the primary endpoint remain unclear but may have been related to low mean baseline CRP in both groups, which, in turn, may have been associated with fewer disease flares regardless of treatment.

Infliximab — Infliximab is a chimeric mouse-human monoclonal antibody with affinity for TNF-alpha. It has been evaluated in combination with methotrexate in clinical trials in patients with recent-onset polyarticular JIA who are refractory to methotrexate monotherapy. It is not approved by the US FDA for the treatment of JIA due to failure to achieve the primary efficacy endpoint at three months in a randomized, placebo-controlled, double-blind study that used standard 3 mg/kg dosing [85]. However, infliximab is still used by some experts in combination with methotrexate because clinical experience suggests that a higher dose (10 mg/kg) is effective [86]. In addition, some patients and their parents/caregivers prefer infusions over injections.

Most centers report an increased incidence of side effects in children treated with infliximab relative to etanercept [85,87,88]. In general, etanercept is better tolerated and more easily administered than infliximab. High doses of infliximab (>5 mg/kg) may be more effective than the standard dose of infliximab (3 mg/kg/dose), but they may also carry increased risks. Further trials are required to characterize the efficacy and safety of various dosing regimens of infliximab in JIA.

In a multicenter, randomized trial, 122 children with polyarticular JIA who had not responded to methotrexate therapy were randomly assigned to one of two treatment regimens [85]. Patients received methotrexate plus infliximab 3 mg/kg at 0, 2, 6, and 14 weeks; placebo at week 16; and then infliximab 3 mg/kg at week 20 and then every 8 weeks through week 44 or methotrexate plus placebo for 14 weeks followed by methotrexate plus infliximab 6 mg/kg every 8 weeks through week 44. The following findings were noted:

●During the first 14 weeks, there was a nonsignificant trend towards greater improvement in the ACR Pedi criteria among patients who initially received infliximab 3 mg/kg compared with those who received placebo (64 versus 49 percent).

●By week 16, when the placebo-treated patients crossed over to infliximab 6 mg/kg, 73 percent of all patients had achieved a 30 percent improvement in the ACR Pedi criteria.

●By week 52, 70 and 52 percent of all patients had achieved a 50 and 70 percent improvement in the ACR Pedi criteria, respectively.

●Antibodies to infliximab developed in 25 percent of the 102 patients who were evaluated. A higher proportion of antibody formation was seen in the 3 mg/kg compared with the 6 mg/kg infliximab groups (38 versus 12 percent). Similarly, overall adverse events were more frequent among those who received 3 mg/kg infliximab versus those receiving 6 mg/kg.

●Adverse events were greater in patients who received infliximab compared with those who received placebo. They included infection (67 versus 47 percent) and infusion-associated reactions (35 versus 8 percent), such as vomiting, fever, headache, and hypotension. There were two deaths. One patient died during the placebo phase, and one patient receiving the lower dose of infliximab also died.

Seventy-eight patients (64 percent) from this study entered an open-label treatment extension [87]. Eight children discontinued infliximab due to lack of efficacy, and 19 patients stopped the drug due to reasons unrelated to efficacy or safety. By week 204, 40 and 33 percent had a 50 and 70 percent improvement in the ACR Pedi criteria, respectively. Infusion reactions continued to occur at a similar rate (32 percent) as in the randomized phase of the study. These reactions occurred more frequently in those positive for antibodies to infliximab.

In a randomized, open-label study, three regimens were compared for the treatment of recent-onset polyarticular JIA: infliximab (3 to 5 mg/kg) plus methotrexate (n = 19), methotrexate alone (n = 20), and a DMARD combination (methotrexate, sulfasalazine, and hydroxychloroquine; n = 20) [89]. ACR Pedi 75 percent improvement (ACR Pedi 75) was achieved in all patients on infliximab plus methotrexate, 65 percent on the DMARD combination, and 50 percent on methotrexate alone. Inactive disease was attained in 68, 40, and 25 percent of the same treatment groups, respectively.

Other biologic agents — Other biologic agents that are options in patients with refractory disease include tocilizumab, abatacept, rituximab, and intravenous immune globulin.

Tocilizumab — Tocilizumab, a humanized monoclonal antibody against the IL-6 receptor, is approved for the treatment of moderate-to-severe polyarticular JIA in children ages two years and older. It has been used in clinical trials in children who are refractory or intolerant to methotrexate [38,39]. Both IV and SC administration of tocilizumab are US FDA approved for polyarticular JIA.

In part I of a multicenter, randomized withdrawal study, 188 children with polyarticular JIA who had not responded to methotrexate therapy received tocilizumab (<30 kg: 10 mg/kg or >30 kg: 8 mg/kg IV every four weeks) [38]. In part II, the 163 patients who achieved an ACR Pedi 30 response in part I were randomized to receive tocilizumab or placebo. The following findings were noted:

●In part I, the open-label lead-in period, 89, 62, and 26 percent of children achieved an ACR Pedi 30, 70, or 90 response, respectively.

●In part II, the randomized withdrawal study, there were significantly more flares in the placebo group versus the treatment group (38 percent versus 26 percent, p = 0.002).

●Use of methotrexate statistically improved the ACR Pedi 70 and 90 response rates in both the placebo and treatment groups.

●There were 16 severe, adverse events in the treatment group, 7 of which led to study withdrawal. Commonly reported adverse events included pneumonia (n = 4), bronchitis (n = 2), and cellulitis (n = 2). There were no deaths and no malignancies. Laboratory abnormalities included transaminitis, neutropenia, and thrombocytopenia. None of the noted laboratory abnormalities were associated with infection or bleeding. Increased low-density lipoprotein (LDL) and total cholesterol levels were noted in 11 and 34 percent of patients, respectively; none required lipid-lowering therapy.

One-hundred sixty of 166 patients in part II entered part III, during which all children received open-label tocilizumab [90]. By week 104, 59.6 and 80.3 percent achieved the ACR 90 and 50, respectively, and 51 percent achieved inactive disease. No new safety signals were revealed.

In an open-label, multicenter study in Japan, 19 children with polyarticular JIA, ages 2 to 19 years, who failed prior therapy or were intolerant to other medications were given three IV doses of tocilizumab (8 mg/kg) every four weeks [39]. The following findings were noted:

●Sixty-three, 90, and 95 percent of patients achieved an ACR Pedi 30 at weeks 4, 8, and 12, respectively.

●Nine and 58 percent attained an ACR Pedi 50 and 70 at week 12, respectively.

●In the 48-week open-label extension period, the ACR Pedi responses were sustained.

In two 52-week phase Ib trials, 52 patients ages 1 to 17 years with polyarticular JIA were given SC tocilizumab every two weeks (≥30 kg) or every three weeks (<30 kg) [91]. Pharmacokinetics, pharmacodynamics, and efficacy assessments were comparable between the SC and IV formulations. By week 52, 31 percent of patients attained clinical remission.

Abatacept — Abatacept blocks T cell costimulation, which interferes with proinflammatory messages being passed to and from T cells. SC abatacept is approved by the US FDA for the treatment of moderate-to-severe polyarticular JIA in children ages two years and older. IV abatacept is approved for use in moderate-to-severe polyarticular JIA in children age six years and older.

The pivotal study demonstrated a decrease in the rate of arthritis flares with abatacept compared with placebo in a trial of 190 patients between 6 and 17 years of age who had a history of active disease in five joints and an inadequate response to at least one DMARD [92]. In this study, 170 patients who completed the open-label lead-in period of four months of abatacept therapy and had a positive response to the drug were randomly selected to continue to receive 10 mg/kg of abatacept every 28 days or placebo for an additional six months or until their arthritis flared. Patients who received abatacept, compared with those who received placebo, experienced fewer disease flares (20 versus 53 percent). There was no difference in the frequency of adverse events between the two groups. Greater improvements were also seen in health-related quality-of-life measures, particularly pain/discomfort and sleep, in the abatacept group compared with the placebo group [93].

In a subsequent phase III, open-label trial, 173 children aged 6 to 17 years and 46 children aged 2 to 5 years with polyarticular JIA who had failed therapy with at least one DMARD were treated with weight-tiered SC abatacept weekly (50 mg if 10 to ≤25 kg, 87.5 mg if 25 to ≤50 kg, and 125 mg if ≥50 kg) for four months and then continued up to 24 months if they had reached at least an ACR Pedi 30 [94]. Approximately 80 percent of the children in both age groups were receiving concomitant methotrexate at baseline. In an intention-to-treat analysis, JIA-ACR30/50/70/90/100/inactive disease responses at four months were 83/73/53/28/14/30 percent in the older children and 89/85/74/59/41/50 percent in the younger children, with responses sustained to month 24. The Juvenile Arthritis Disease Activity Score 71-C-reactive protein (JADAS71-CRP) median scores improved from baseline to month 4 and were maintained to month 24, at which time 16 percent of the older children and 50 percent of the younger children had achieved remission based upon this score. Serious adverse events were reported in 6 to 8 percent of the children.

Rituximab — Rituximab, a chimeric anti-CD20 monoclonal antibody, has been approved for the treatment of adult-onset RA [95,96]. Although most subsets of JIA are not the same disease as adult RA, in limited experience, adolescents with RF-positive and anticyclic citrullinated peptide (CCP) positive disease unresponsive to other agents have benefited from rituximab. The utility of this agent in other settings is unknown. There are anecdotes of its successful use in other subtypes of JIA, but there are no reliable data to date regarding its efficacy in patients with refractory JIA. (See "Rituximab: Principles of use and adverse effects in rheumatoid arthritis".)

Intravenous immune globulin — IV immune globulin has been studied in polyarticular JIA. While it appears to have some efficacy, its effects are short lived and insufficient to justify its cost and potential toxicity [97].

Small molecule inhibitors — Small molecule inhibitors that are options in patients with active disease include Janus kinase (JAK) inhibitors such as tofacitinib and baricitinib.

Tofacitinib — Tofacitinib, a JAK inhibitor, was approved by the US FDA in 2020 for the treatment of polyarticular JIA based upon a small randomized trial [98]. Patients aged 2 to 17 years with polyarticular JIA not responsive or intolerant to methotrexate or a biologic agent were treated with oral tofacitinib (3.2 mg twice daily for body weight 10 kg to <20 kg, 4 mg twice daily for body weight 20 kg to <40 kg, and 5 mg twice daily for body weight ≥40 kg) for 18 weeks. At the end of this run-in phase, patients with at least a 30 percent improvement in the ACR JIA response criteria (ACR30) were randomly assigned to tofacitinib or placebo. In both phases, approximately two-thirds of patients were on concomitant methotrexate, and one-third were on oral glucocorticoids. At week 44, significantly fewer treated patients had a disease flare compared with the placebo group (31 versus 55 percent, respectively; difference in proportion -25 percent [95% CI -39 to -10 percent]).

Baricitinib — The safety and efficacy of baricitinib was studied in a randomized trial of 220 patients aged 2 to 17 years old with JIA (66 percent with polyarticular JIA) who had inadequate response or intolerance to standard therapy [99]. After a 2-week safety period and a 10- to 12-week open-label period during which all patients received baricitinib, 163 patients who had at least an ACR30 were randomly assigned to placebo (n = 81) or baricitinib (n = 82; dosing 2 mg daily for patients 2 to 8 years old and 4 mg daily for patients 9 to 17 years old) for an up to 32-week double-blind withdrawal period. Time to disease flare was shorter in the placebo group than baricitinib group (adjusted hazard ratio 0.241, 95% CI 0.128-0.453), and more patients in the placebo group had a flare during this period (disease flare rate 51 percent compared with 17 percent in the baricitinib group). A greater number of infections were reported in the baricitinib group than placebo during the withdrawal period (38 versus 19 percent, respectively), but rates of serious adverse events were similar in both groups. Pulmonary embolism reported in one patient on baricitinib was felt to be study drug related.

Additional agents — Additional DMARDs that may be useful in patients with polyarticular JIA who have contraindications or intolerance to methotrexate include sulfasalazine and leflunomide. Long-term systemic glucocorticoids are generally not recommended, but short-term, low-dose glucocorticoids and/or intraarticular glucocorticoids may provide rapid symptom relief for arthritis that is particularly disabling in one or more joints. Cyclosporine and azathioprine are rarely used, and gold compounds are no longer used now that more effective drugs (biologic agents) are available.

Sulfasalazine — Sulfasalazine has been shown to be beneficial for many children with polyarticular JIA [100-103].

This utility was illustrated in a randomized 24-week trial of 69 patients with oligoarticular and polyarticular JIA [101]. Sulfasalazine resulted in significant improvement compared with placebo, as determined by overall articular severity score, global assessments, and laboratory parameters. Adverse events, which were found in one-third of patients, were transient or reversible upon cessation of treatment [101,102].

A subsequent study reported the clinical status of this cohort of patients at a median follow-up of nine years [103]. Patients who were in the sulfasalazine treatment group compared with those treated with placebo appeared to require less intensive DMARD treatment based upon a shorter duration of sulfasalazine (2.5 versus 5.2 years) and a trend to less use of methotrexate and other DMARDs. At follow-up assessment, outcome scores were better in the group treated with sulfasalazine compared with the control group. However, more than one-third of patients in both groups reported long periods of noncompliance with therapy, which significantly impacted patient outcomes.

In using sulfasalazine, parents/caregivers must be warned of the possible development of rare, severe reactions seen with sulfa drugs (eg, Stevens-Johnson syndrome). White blood cell (WBC) count, platelet count, and liver enzymes must be periodically monitored as well. Sulfasalazine does not prevent chronic joint changes and, therefore, should not be relied upon in erosive disease. For those with erosive disease, methotrexate or anti-TNF therapies are preferable.

Leflunomide — Leflunomide, an immunomodulator that inhibits pyrimidine synthesis, has been shown to be safe and effective in adults with RA. Although not US FDA approved for pediatric use, efficacy in JIA has been reported but has not been shown to be superior to methotrexate [9,104,105].

In a randomized study of 94 children with JIA, both methotrexate and leflunomide resulted in clinically meaningful improvements [9]. Patients were treated initially with either methotrexate or leflunomide for 16 weeks, with blinded treatment extended for an additional 32 weeks. Both medications resulted in improved articular severity score, global assessments (clinician and patient), and erythrocyte sedimentation rate. Methotrexate appeared to be more effective than leflunomide since a greater percentage of children demonstrated clinical improvement, although the dose of leflunomide used in this study was somewhat low. The most common adverse events for both medications included gastrointestinal symptoms (abdominal pain, nausea, vomiting, or diarrhea), headache, and nasopharyngeal symptoms (eg, pharyngitis). Children on either medication need to be carefully monitored for evidence of hepatic toxicity.

Leflunomide was added to the regimen in one observational study of 32 children with polyarticular JIA who failed methotrexate therapy [106]. At six months, 85 percent achieved an ACR Pedi 30 response. At last follow-up (range: 0.3 to 3 years), 12 of 18 (66 percent) met the ACR Pedi 30 criteria, and 9 of 18 (50 percent) were in clinical remission on medication.

Systemic glucocorticoids — Long-term, high-dose oral glucocorticoids should be avoided in children with polyarticular JIA because of the high incidence of side effects, particularly growth retardation and osteoporosis. Some have advocated pulsed doses of glucocorticoids as being effective in controlling inflammation without causing side effects [25]. Others, however, have noted that over 20 percent of patients receiving pulsed-dose glucocorticoids had one or more adverse reactions [25].

Short-term use of low-dose glucocorticoids (less than 0.25 mg/kg per day of prednisone or its equivalent) may provide substantial benefits for the child who is otherwise incapacitated by arthritis without causing excessive complications. However, most clinicians have found anti-TNF agents to provide an equally rapid clinical response in children with severe JIA. Children with JIA severe enough to warrant consideration of glucocorticoids may ultimately require anti-TNF therapy. Thus, it may be preferable to begin with anti-TNF therapy in such patients.

Cyclosporine and azathioprine — For children with continuing active disease despite appropriate treatment with NSAIDs and methotrexate, azathioprine and cyclosporine have been used with varying success in the past [25]. Cyclosporine is effective in children with dactylitis who have not responded to sulfasalazine. However, these drugs carry significant potential toxicity and appear to be less effective than the newer biologic agents. As a result, most clinicians favor the use of the biologic agents over these drugs.

COURSE AND PROGNOSIS — Children with the early onset of polyarticular JIA, regardless of the presence or absence of rheumatoid factor (RF), have a low likelihood of developing spontaneous remission. Despite the availability of additional medications to treat polyarticular JIA (eg, methotrexate and tumor necrosis factor [TNF] inhibitors), a significant number of patients continue to have some disease activity. The majority of children who "grow out of it" probably represent misclassified cases of reactive arthritis.

The following examples are illustrative:

●In a prospective study of 137 children with polyarticular JIA, clinically inactive disease was an unstable state with 18 percent of children experiencing a flare while on TNF inhibition [48]. In that same study, 37 percent of children experienced a major flare within eight months of TNF inhibitor withdrawal. Levels of serum S100 proteins (immunohistochemical markers) were not predictive of maintenance of remission while on TNF inhibition or flare after withdrawal of TNF inhibition [107].

●In a large Canadian cohort of children diagnosed with JIA from 2005 to 2010, at one, three, and five years from the time of diagnosis, 34, 78, and 97 percent of children with RF-negative disease achieved inactive disease, respectively [108]. These numbers are slightly lower in children with RF-positive disease (22, 67, and 93, respectively). Only 14 percent of RF-negative and none of the children with RF-positive disease attained remission off medication within five years of diagnosis.

●In another observational cohort study from Taiwan, 48 and 13 percent of children with RF-negative disease and 33 and 11 percent of children with RF-positive disease achieved inactive disease and remission off medication over the course of 15 years [109]. Differences in these estimates likely result from slight differences in definitions of inactive disease and remission and follow-up times.

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".)

SUMMARY AND RECOMMENDATIONS

●Goals of treatment – Treatment of polyarticular juvenile idiopathic arthritis (JIA) is directed toward treating the underlying synovitis and associated inflammation. Prompt initiation of therapy that provides relief of arthritic symptoms and preservation of function is critical to an improved outcome. (See 'Introduction' above.)

●Considerations in choosing a treatment regimen – Treatment regimens are chosen according to prognostic factors, as well as clinician, parent/caregiver, and patient preferences, and are adjusted based upon clinical response. (See 'Management approach' above.)

●Initial therapy – We recommend initial therapy with a disease-modifying antirheumatic drug (DMARD) in all patients with polyarticular JIA (Grade 1B). We use either methotrexate at a dose of 10 mg/m² body surface area (BSA)/week or a biologic agent that inhibits tumor necrosis factor (TNF) in combination with methotrexate. Indications for using a TNF inhibitor in addition to or rather than methotrexate include presence of severe polyarthritis, poor prognostic features, or factors associated with poor response to methotrexate (eg, predominantly axial arthritis). Nonsteroidal antiinflammatory drugs (NSAIDs) are not appropriate as monotherapy but are suitable as adjunct therapy for symptom management. (See 'Initial management' above and 'Methotrexate' above and 'Tumor necrosis factor inhibitors' above and 'Nonsteroidal antiinflammatory drugs' above.)

●Supplemental therapy – Folic acid or leucovorin (folinic acid) supplementation is used in all children with JIA receiving methotrexate, given the beneficial effects noted and the lack of convincing contrary data. (See 'Folic acid' above.)

●Refractory disease – Individualized therapeutic decisions are necessary in patients who fail to respond to methotrexate and a TNF inhibitor. Options include the use of combination DMARDs, another biologic agent, or a small molecule inhibitor under the supervision of providers with expertise in pediatric rheumatology. (See 'Refractory disease' above and 'Other biologic agents' above and 'Small molecule inhibitors' above and 'Additional agents' above.)

●RF- and CCP-positive disease – Adolescents with rheumatoid factor (RF) positive and anticyclic citrullinated peptide (CCP) positive polyarthritis probably represent the early onset of rheumatoid arthritis (RA). As such, these children should be treated similar to adults with moderate-to-severe RA. (See 'Rheumatoid factor-positive patients' above and "Initial treatment of rheumatoid arthritis in adults".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Thomas JA Lehman, MD, who contributed to earlier versions of this topic review.