Spondyloarthritis in children

INTRODUCTION — 

Spondyloarthritis (SpA) refers to a group of conditions that are characterized by peripheral and axial arthritis, enthesitis, and frequent human leukocyte antigen B27 (HLA-B27) positivity [1]. It encompasses multiple subtypes of arthritis that affect both children and adults, including ankylosing spondylitis (AS), psoriatic arthritis, reactive arthritis, and arthritis related to inflammatory bowel disease (IBD). (See "Overview of the clinical manifestations and classification of spondyloarthritis".)

SpA in children is considered to be a subtype of juvenile idiopathic arthritis (JIA) and can be further classified as enthesitis related arthritis (ERA), psoriatic juvenile idiopathic arthritis (psJIA), or undifferentiated arthritis based on commonly accepted criteria [2,3]. This topic will review the clinical manifestations, diagnosis, and treatment of SpA in children. Detailed information on psJIA and the classification of JIA are provided elsewhere. (See "Psoriatic juvenile idiopathic arthritis: Epidemiology, clinical manifestations, and diagnosis" and "Psoriatic juvenile idiopathic arthritis: Management and prognosis" and "Classification of juvenile idiopathic arthritis".)

IBD-related arthritis and reactive arthritis in children and adults are likewise discussed elsewhere. (See "Clinical manifestations and diagnosis of arthritis associated with inflammatory bowel disease and other gastrointestinal diseases" and "Reactive arthritis".)

Finally, the diagnosis and classification of SpA and other related types of arthritis in adults are covered separately:

●(See "Overview of the clinical manifestations and classification of spondyloarthritis".)

●(See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults".)

●(See "Clinical manifestations and diagnosis of peripheral spondyloarthritis in adults".)

●(See "Clinical manifestations and diagnosis of psoriatic arthritis".)

TERMINOLOGY AND CLASSIFICATION — 

Spondyloarthritis (SpA) refers to multiple subtypes of inflammatory arthritis that can affect adults and children and are often associated with human leukocyte antigen B27 (HLA-B27). These subtypes include the following:

●Ankylosing spondylitis (AS)

●Psoriatic arthritis

●Reactive arthritis

●Arthritis related to inflammatory bowel disease (IBD)

When SpA occurs in children who are under age 16 years old, it is considered to be a subtype of juvenile idiopathic arthritis (JIA). SpA can be further subcategorized depending on the classification criteria used for JIA:

●Using criteria developed by the International League of Organizations for Rheumatology (ILAR) [3], children with SpA may fall into one of three mutually exclusive categories (table 1) (see "Classification of juvenile idiopathic arthritis", section on 'Revised ILAR JIA classification criteria'):

•Enthesitis related arthritis (ERA)

•Psoriatic JIA (psJIA)

•Undifferentiated JIA

●Using a newer proposed classification system for JIA from the Pediatric Rheumatology International Organization (PRINTO) [2], children with SpA are categorized as having enthesitis/spondylitis-related JIA or "other" or "unclassified" JIA. PRINTO does not include a clear category for patients with psJIA, which has been criticized [4]. (See "Classification of juvenile idiopathic arthritis", section on 'Future directions'.)

An overview of the overlap between ILAR, PRINTO, and adult classification systems demonstrates the overlap between various categories of SpA (figure 1). (See "Classification of juvenile idiopathic arthritis", section on 'Overlap with the classification of adult arthritis'.)

Many other names have been used to refer to children with SpA, which are relevant to review given their use in the medical literature:

●Spondyloarthropathy – While this term is frequently used interchangeably with spondyloarthritis, the latter is preferred as it better encapsulates the inflammatory nature of these conditions [5].

●Juvenile SpA – This term may be used to refer to patients who were diagnosed with SpA in childhood and who have bilateral sacroiliitis based on imaging [6].

●Juvenile AS – In contrast with the term "AS" in adults [7], "juvenile AS" is less well defined and is occasionally and inappropriately used to indicate any child with axial arthritis.

●Seronegative enthesitis and arthritis (SEA) syndrome – This term may refer to patients with various types of SpA including AS, reactive arthritis, and IBD-related arthritis. This diagnosis is present in older literature but presently is rarely used, if ever.

EPIDEMIOLOGY — 

The incidence of all types of juvenile idiopathic arthritis (JIA) is approximately 1 in every 2000 children [8]. Among patients with JIA, approximately 10 to 15 percent are categorized as having enthesitis related arthritis (ERA) and another 5 percent as having psoriatic JIA (psJIA) [9-11]. However, the number of patients with psJIA who would be considered to have spondyloarthritis (SpA) may be lower, since approximately 30 percent of patients who are categorized as having psJIA develop arthritis at a young age (before age five or six years) and have features that better align with early-onset antinuclear antibody (ANA)-positive oligo- and polyarticular JIA [12,13].

Children with ERA typically have an older age of symptom onset (around 10 to 11 years old) compared with those who have other types of JIA [14]. While JIA in general affects females more than males, approximately 50 percent of children with SpA are male [14,15]. While historically SpA was thought to affect more males than females, subsequent studies have questioned this association [16,17].

SpA is associated with human leukocyte antigen B27 (HLA-B27) positivity. While HLA-B27 is more common in people from certain ethnic backgrounds (eg, White Americans and Northern Europeans) [18], SpA can affect people regardless of their heritage and country of origin.

More information on the epidemiology of JIA in general, psJIA, and certain other subtypes of SpA is provided separately:

●(See "Juvenile idiopathic arthritis: Epidemiology and immunopathogenesis", section on 'Epidemiology'.)

●(See "Psoriatic juvenile idiopathic arthritis: Epidemiology, clinical manifestations, and diagnosis", section on 'Epidemiology'.)

●(See "Reactive arthritis", section on 'Epidemiology'.)

●(See "Clinical manifestations and diagnosis of arthritis associated with inflammatory bowel disease and other gastrointestinal diseases", section on 'Prevalence'.)

PATHOGENESIS — 

The pathogenesis of juvenile idiopathic arthritis (JIA), including spondylarthritis (SpA), is unknown. More detail on the possible pathogenesis of JIA is provided elsewhere. (See "Juvenile idiopathic arthritis: Epidemiology and immunopathogenesis", section on 'Pathogenesis'.)

Presumably, the pathogenesis of enthesitis related arthritis (ERA) and psoriatic JIA (psJIA) are similar to that of their adult counterparts, which are described in detail separately. (See "Pathogenesis of spondyloarthritis" and "Pathogenesis of psoriatic arthritis".)

Studies that are specific to the pathogenesis of SpA in children have highlighted the potential role of genetic factors, alterations in the intestinal microbiota, and intestinal and entheseal inflammation:

●Genetic factors – The most widely known genetic risk factor for SpA in both children and adults is the human leukocyte antigen B27 (HLA-B27) allele [19], which is present in 30 to 90 percent of children with SpA [14] and in 90 to 95 percent of patients with ankylosing spondylitis (AS) [20,21]. It is unclear how HLA-B27 influences the development of SpA, but various mechanisms have been proposed including antigen presentation to self-reactive T cells [22], activation of the innate immune system through endoplasmic reticulum misfolding or cell surface dimerization [19], and alteration of the intestinal microbiota [23]. (See "Pathogenesis of spondyloarthritis", section on 'Role of HLA-B27'.)

Multiple additional genetic risk factors have been identified using genome-wide association studies (GWAS) in adults [24]; endosomal receptor aminopeptidase 1 and interleukin 23 receptor (IL23R) have been associated with ERA and psJIA, respectively [25]. (See "Pathogenesis of spondyloarthritis", section on 'Non-HLA genes'.)

●Alterations in the gut microbiota and intestinal inflammation – The relationship between inflammatory arthritis and the gastrointestinal system remains under investigation, with potentially mediating factors including the gut microbiota, defective barrier function, genetics, and T helper type 17 (Th17) cells [26]. (See "Pathogenesis of spondyloarthritis", section on 'The gut mucosa, gut microbiome, and IL-17A'.)

Multiple observational studies have identified alterations in the gastrointestinal microbiota among children with ERA compared with healthy children [27-29]. Certain key bacteria have been positively or negatively associated with ERA in multiple different patient groups, raising the possibility of causal relationships [30]. Given the relatedness of the microbiota between family members [31], the microbiota could account for some of the "missing heritability" identified in AS [32], defined by the difference between the reported heritability of 97 percent calculated in a twin study [33] versus the 24 percent calculated through GWAS [34].

Large cross-sectional studies indicate that up to two-thirds of adults with SpA have subclinical intestinal inflammation [35]; smaller studies in children have replicated this finding [36,37], including one that noted an association between elevated fecal calprotectin levels and a higher incidence of active sacroiliitis in patients with ERA [38].

●Entheseal inflammation – Entheseal inflammation due to mechanical stress may contribute to the development and/or progression of SpA. Studies have demonstrated that excessive stress at the enthesis causes damage to the extracellular matrix and that the subsequent healing process contributes to chronic inflammation [39,40]. In addition, studies of adults with SpA have noted worsening disease activity or progression among those who have jobs that require mechanical labor or "dynamic flexibility" (eg, twisting, bending) [41,42].

CLINICAL PRESENTATION — 

Pediatric patients with spondylarthritis (SpA) may have a variety of articular manifestations including axial arthritis, peripheral arthritis, enthesitis, and, in certain subtypes, dactylitis. The peripheral arthritis is typically asymmetric and oligoarticular, with a predilection for large joints in the lower limbs more than the upper limbs. Extraarticular manifestations may include uveitis, psoriasis, and intestinal inflammation. The progression of symptoms is typically gradual, with the exception of acute anterior uveitis (AAU).

The clinical presentation of psoriatic juvenile idiopathic arthritis (psJIA) is discussed in more detail elsewhere. (See "Psoriatic juvenile idiopathic arthritis: Epidemiology, clinical manifestations, and diagnosis", section on 'Clinical manifestations'.)

Articular manifestations

●Axial arthritis – Patients with SpA may have inflammation in the spinal vertebrae (spondylitis) and/or the sacroiliac joint (sacroiliitis), which is referred to as axial arthritis. Axial arthritis is a less common presenting feature in children than in adults, with a prevalence of 10 to 24 percent at the time of diagnosis [43,44]. Children who develop axial arthritis often have peripheral arthritis at the onset of the disease.

Some patients with axial arthritis may be asymptomatic [45,46], while others experience a gradual onset of symptoms suggestive of inflammatory back pain (eg, pain that has a gradual onset and improves with activity but not with rest). Pain from sacroiliitis may also be referred to the hip or buttock.

●Peripheral arthritis – Peripheral arthritis in children with SpA is most often asymmetrical and oligoarticular (in four or fewer joints). The most commonly involved sites at diagnosis are the knee, ankle, and midfoot. Children with enthesitis related arthritis (ERA) typically present with arthritis involving the large joints of the lower extremities [14]. psJIA, like its counterpart in adults, has a more variable presentation and more frequently involves small joints of the lower and upper extremities [12].

●Enthesitis – Enthesitis is defined as inflammation at the site where a tendon inserts into the bone (ie, the enthesis) (figure 2). It is often symmetric and frequently involves the plantar fascial insertion at the calcaneus, the calcaneal (Achilles) tendon insertion into the calcaneus, and the patellar tendon insertion into the lower patellar pole and tibial tuberosity [43,47].

Enthesitis is part of the International League of Organizations for Rheumatology (ILAR) classification criteria for ERA [3], which may bias estimates of its frequency. However, enthesitis appears to affect approximately two-thirds of patients with children with SpA based on a case series that did not use ILAR classification [48] and an inception cohort of children with ERA or SpA that did [43].

●Dactylitis – Dactylitis refers to fusiform swelling of a digit (picture 1 and picture 2) and is characterized by flexor tenosynovitis, localized soft tissue inflammation, synovitis, and periostitis [49]. It can affect patients with psJIA or reactive arthritis. It is part of the ILAR criteria for psJIA [3], which complicates efforts to determine its true prevalence, and occurs in approximately one-third of patients [50].

Extraarticular manifestations — Patients with SpA may develop extraarticular manifestations, as discussed below.

Uveitis — Uveitis associated with JIA is characterized by two distinct phenotypes that are discussed in detail elsewhere:

●Acute anterior uveitis – AAU causes pain, scleral injection, and photophobia. It is typically unilateral. Children with ERA are at risk for AAU, especially if they are positive for human leukocyte antigen B27 (HLA-B27). (See "Juvenile idiopathic arthritis: Immunizations and complications", section on 'Acute anterior uveitis'.)

●Chronic uveitis – Chronic uveitis is typically asymptomatic until it begins to affect vision. This form is more common in patients with many types of JIA, including patients with psJIA [13]. (See "Juvenile idiopathic arthritis: Immunizations and complications", section on 'Chronic anterior uveitis'.)

Psoriasis — Based on ILAR exclusion criteria, patients with psoriasis must be categorized as having either psJIA or undifferentiated arthritis; therefore, patients with ERA by definition cannot have psoriasis. Patients with psJIA often develop arthritis several years before psoriasis [51], which is in contrast with adults with psoriatic arthritis who typically have psoriasis first. Since many of the medications used to treat JIA are also effective in psoriasis, it is possible that treatment of psJIA sine psoriasis can prevent development of the cutaneous disease.

Intestinal inflammation — The development of SpA is frequently associated with intestinal inflammation. Patients may be relatively asymptomatic or progress to having inflammatory bowel disease (IBD) with symptoms including abdominal pain, altered bowel habits, aphthous ulcers, weight loss, and decreased longitudinal growth. Rarely, pyoderma gangrenosum and erythema nodosum may be seen in patients with SpA related to IBD [52].

Aortic insufficiency — A rare extraarticular manifestation is aortic disease. In adults with ankylosing spondylitis (AS), approximately 6 to 10 percent develop aortic insufficiency [53]. Data in children with SpA are limited. In a prospective study, 3 of 31 children with SpA had "mild" aortic regurgitation on echocardiography compared with none of the 33 healthy controls [54]. Screening for aortic insufficiency in asymptomatic patients is not routinely performed in pediatric practice. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Cardiovascular disease'.)

EVALUATION AND DIAGNOSIS

When to suspect — The diagnosis of spondylarthritis (SpA) should be suspected in a child who presents with axial and/or peripheral arthritis or enthesitis, especially when the arthritis involves the large joints of the lower extremities.

Other suggestive clinical and demographic features include patients who are in their teenage years and those with a personal and/or family history of related conditions such as acute anterior uveitis (AAU), psoriasis, and/or inflammatory bowel disease (IBD). Patients often experience notable symptomatic relief with nonsteroidal antiinflammatory drugs (NSAIDs).

Approach to evaluation — The evaluation of children with suspected SpA includes a history and physical examination with attention to evidence of arthritis, enthesitis, and related conditions (eg, psoriasis). Laboratory testing is not specific for SpA but can help evaluate the extent of peripheral inflammation and evaluate for alternative causes of inflammatory arthritis. Imaging is required to evaluate for axial arthritis.

History and physical examination — We ask patients and/or their families about symptoms related to articular and extraarticular manifestations of SpA and take a family history focused on related conditions (ie, psoriasis, IBD, uveitis) (see 'Clinical presentation' above). In addition, we review growth charts for evidence of weight loss and/or decreased longitudinal growth. While specific questions to assess inflammatory back pain are often used for adults (eg, asking about the presence of morning stiffness, improvement with exercise but not with rest, pain in the second half of the night, and alternating buttock pain [55]), these measures have not been validated in children and are insufficient to exclude potential axial disease, especially since patients may be asymptomatic.

Physical examination should be comprehensive, including the following aspects:

●Joints – Peripheral arthritis can be detected by the presence of pain, tenderness, and limited range of motion. Joint swelling, warmth, and overlying redness may be appreciable in superficial joints. The presence of hip arthritis can be assessed through careful assessment of range of motion, since swelling will not be visible [3]. Patients with tarsitis will have pain when squeezing the midfoot.

Axial arthritis can be difficult to detect. There are numerous physical examination assessments for sacroiliac joint arthritis (eg, Patrick test (picture 3)) [56]; however, they appear to have limited sensitivity and specificity in children [46,57], and data supporting their use in adults are mixed [56,58]. The modified Schober test is commonly used to assess forward flexion of the lumbosacral spine (picture 4).

●Entheses – Entheses can be palpated at the insertion sites of ligaments and tendons onto bone (figure 2). We carefully examine the patient for signs of thickened superficial tendons (eg, Achilles tendon) and focal tenderness (eg, by palpating the plantar fascia and patellar and Achilles tendon insertion sites).

●Skin – Findings suggestive of psoriasis in children can be subtle, such as mild lesions behind the ears, on the scalp, and inside the intergluteal cleft (figure 3) [52]. In infants and toddlers, psoriasis can be present in atypical locations and mistaken for more common entities, such as diaper dermatitis and eczema [59]. Nail pitting and onycholysis may also be seen (picture 5 and picture 6) [52]. (See "Psoriasis in children: Epidemiology, clinical manifestations, and diagnosis", section on 'Clinical presentation'.)

We also look for evidence of aphthous ulcers (picture 7), as can be seen with underlying IBD. (See "Oral lesions", section on 'Aphthous ulcerations'.)

●Eyes – We look for signs of uveitis, including photophobia and scleral injection (picture 8). Synechiae, which are adhesions between the iris and the lens that cause irregularities in the pupil, may be a sign of chronic, minimally symptomatic uveitis (picture 9). (See "Uveitis: Etiology, clinical manifestations, and diagnosis", section on 'Clinical manifestations'.)

Laboratory testing — There is no single laboratory test that is diagnostic of SpA in children. We typically send the following studies:

●Inflammatory markers – We obtain an erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). These markers are not specific for inflammatory arthritis nor are they always elevated in those with active disease. However, baseline elevations have some prognostic value for patients with juvenile idiopathic arthritis (JIA) due to their association with lower rates of achieving remission [60] and increased risk of progression to ankylosing spondylitis (AS) [61].

●Complete blood cell count – A complete blood cell count (CBC) may be normal or demonstrate thrombocytosis and/or anemia of chronic disease. Iron deficiency anemia can be a sign of associated IBD.

●Autoantibodies – We occasionally check for the presence of rheumatoid factor (RF), anti-cyclic citrullinated peptide (CCP) antibody, and antinuclear antibody (ANA). Patients are typically negative for RF, anti-CCP, and ANA, although ANA may be positive in patients with psoriatic JIA (psJIA). Patients with JIA who are ANA positive also have a higher risk for developing chronic uveitis. High titers of autoantibodies may suggest an alternative or overlapping diagnosis, such as seropositive polyarticular JIA or a systemic autoimmune disease (eg, systemic lupus erythematosus).

●Human leukocyte antigen B27 (HLA-B27) – HLA-B27 is positive in anywhere between 30 to 90 percent of children with enthesitis related arthritis (ERA) [14]. Although there are variations in practice pattern, we routinely obtain HLA-B27 testing in children with known or suspected ERA in order to stratify the risk of associated complications, including AAU and sacroiliitis.

Additional studies may be indicated for patients suspected of having IBD (eg, fecal blood and calprotectin). (See "Clinical presentation and diagnosis of inflammatory bowel disease in children", section on 'Laboratory features'.)

Imaging

Axial imaging for patients with back pain — In children with back pain and suspected SpA, we routinely obtain imaging to assess for axial involvement. Arguably, in children with peripheral SpA treated exclusively with conventional synthetic disease-modifying antirheumatic drugs (csDMARDs), a case could be made for routine screening of the sacroiliac joint to evaluate for axial SpA, as axial involvement may be clinically asymptomatic and would warrant use of biologic DMARDs (bDMARDs) [57].

The optimal imaging modality to assess the sacroiliac joint is magnetic resonance imaging (MRI) of the pelvis without gadolinium [62] because it is more sensitive and does not involve radiation. Some centers initially obtain radiographs of the sacroiliac joints and proceed to MRI if the radiographs are not diagnostic. There is limited value of obtaining plain films to evaluate for sacroiliitis in children [63], although they may be required for cost and regulatory reasons prior to obtaining a MRI study (eg, by health insurance companies in the United States).

MRI findings suggestive of active sacroiliitis include periarticular bone marrow edema [64]. However, isolated areas of bone marrow edema are not specific for SpA [65]. It may be difficult to differentiate inflammatory changes in the sacrum from the appearance of hypercellular bone marrow, which is normally present throughout much of childhood and is likewise periarticular (image 1). Chronic sacroiliitis may be visualized on MRI with sclerosis, fatty metaplasia of the bone, erosions, and fusion [66], although these findings are likely observed less frequently in children as compared with adults.

Computed tomography (CT) can identify chronic changes but is not the preferred diagnostic modality in children due to inability to detect active lesions and the burden of radiation. Notably, sacroiliitis is sometimes incidentally detected incidentally by cross-sectional imaging of the abdomen (eg, CT, MRI) in patients undergoing evaluation for IBD [67]; in this case, additional imaging is typically not required.

Peripheral imaging in patients with focal symptoms — When there is diagnostic uncertainty regarding the etiology of peripheral joint pain, advanced imaging studies may be helpful. The choice of imaging study depends on the area affected, the suspected underlying pathology (eg, arthritis versus enthesitis), and the availability of various imaging modalities. Synovitis and enthesitis are more easily visualized on ultrasound and MRI compared with radiographs [68,69]. However, radiographs can show bony abnormalities and abnormalities at the entheseal insertion sites later in disease. As an example, in a patient being evaluated for possible ERA who has wrist pain, ultrasonography may be a helpful initial test when available; otherwise, radiography can be done first and be followed by MRI if radiographs are normal.

Establishing the diagnosis — The diagnosis of SpA in children is made clinically based on a combination of clinical features (eg, arthritis, enthesitis, and/or sacroiliitis) in the setting of a personal or family history of related conditions (eg, psoriasis, AAU, AS, IBD). Most patients who have axial arthritis and/or enthesitis on imaging, dactylitis, and/or arthritis with a personal history of psoriasis, AAU, or IBD will be categorized as having SpA, although those with more psoriatic features will fall into the psoriatic arthritis category. HLA-B27 positivity is also suggestive of SpA, although it is not diagnostic.

Classification criteria for diagnosing axial disease in children with SpA were published in 2024 [70]. However, since these were created to be used in research rather than clinical practice, they are intentionally stringent and therefore exclude some patients who can still be diagnosed with axial SpA based on clinical expertise.

In patients who are diagnosed with SpA, it is important to evaluate for possible related conditions including IBD and psoriasis. These complications can also develop after arthritis. (See 'Psoriasis' above and "Clinical presentation and diagnosis of inflammatory bowel disease in children" and "Psoriasis in children: Epidemiology, clinical manifestations, and diagnosis".)

The diagnosis of psJIA is discussed in more detail separately. (See "Psoriatic juvenile idiopathic arthritis: Epidemiology, clinical manifestations, and diagnosis", section on 'Diagnosis'.)

Classification criteria for enthesitis related arthritis — The International League of Organizations for Rheumatology (ILAR) revised classification criteria for JIA provide additional guidance for diagnosing subtypes of SpA in children, specifically ERA and psJIA [3]. Patients with arthritis lasting at least six weeks with age of onset before their 16^th birthday are classified as having ERA if they fulfill the following inclusion and exclusion criteria:

●Inclusion criteria – Both arthritis and enthesitis or either arthritis or enthesitis accompanied by ≥2 of the following:

•The presence of or history of sacroiliac joint tenderness and/or inflammatory back pain

•HLA-B27 positivity

•Onset of arthritis in a male over six years of age

•AAU

•History in a first-degree relative of AS, ERA, sacroiliitis with IBD, reactive arthritis, or AAU

●Exclusion criteria – Patients have ≥1 of the following:

•A personal history of or first-degree relative affected by psoriasis

•Positive testing for RF (defined as having a positive immunoglobulin M [IgM] RF test on at least two occasions checked at least three months apart within the first six months of disease)

•Features consistent with systemic JIA (sJIA; eg, fever, evanescent rash, elevated ferritin) (see "Systemic juvenile idiopathic arthritis: Clinical manifestations and diagnosis")

Importantly, a diagnosis of ERA may be made in the absence of peripheral arthritis when a child has enthesitis or isolated sacroiliitis [2,3,71]. The diagnosis of sacroiliitis is challenging in both pediatric and adult subjects and requires imaging for confirmation. (See 'Axial imaging for patients with back pain' above.)

The classification of psJIA is discussed in more detail separately. (See "Psoriatic juvenile idiopathic arthritis: Epidemiology, clinical manifestations, and diagnosis", section on 'Diagnostic criteria'.)

DIFFERENTIAL DIAGNOSIS — 

The differential diagnosis for patients suspected of having spondyloarthritis (SpA) is broad and varies based on the clinical manifestations (eg, peripheral arthritis, axial arthritis, enthesitis). Key diagnoses to consider include the following:

●Other systemic rheumatic diseases causing peripheral arthritis – Inflammatory peripheral arthritis is common to many systemic rheumatic diseases, including other types of juvenile idiopathic arthritis (JIA), systemic lupus erythematosus (SLE), and sarcoidosis. In contrast to SpA, axial arthritis and prominent enthesitis are rarely seen in these other conditions.

●Chronic nonbacterial osteomyelitis (CNO)/chronic recurrent multifocal osteomyelitis (CRMO) – CNO/CRMO is a systemic disease that is associated with human leukocyte antigen B27 (HLA-B27) and other associated conditions, including SpA, psoriasis, and inflammatory bowel disease (IBD). Inflammatory bone lesions from CNO/CRMO may cause widespread or more focal bone pain and may involve the spine and pelvis, mimicking axial arthritis. However, synovitis is rare. It may be distinguished from SpA based on the typical appearance on MRI. (See "Chronic nonbacterial osteomyelitis (CNO)/chronic recurrent multifocal osteomyelitis (CRMO) in children".)

●Apophysitis – Traction apophysitis results from stress on a growth plate and the nearby ossification centers, which can occur in the setting of growth, overuse, and injury. Symptoms including focal pain and tenderness over tendon insertion sites can mimic enthesitis. Specific types include Osgood-Schlatter disease (tibial tuberosity apophysitis), Sinding-Larsen-Johansson syndrome (patellar apophysitis), and calcaneal apophysitis (Sever disease). The presence of multiple supposed areas of apophysitis in one child, or the presence of any of them in a skeletally mature child, is suggestive of enthesitis secondary to SpA. (See "Osgood-Schlatter disease (tibial tuberosity avulsion)" and "Approach to chronic knee pain or injury in children or skeletally immature adolescents", section on 'Sinding-Larsen-Johansson disease (patellar apophysitis)' and "Heel pain in the active child or skeletally immature adolescent: Overview of causes", section on 'Calcaneal apophysitis (Sever disease)'.)

●Amplified musculoskeletal pain syndrome (fibromyalgia) – The presence of generalized joint and back pain, especially in the absence of any obvious peripheral inflammation, should prompt consideration of amplified musculoskeletal pain syndrome (AMPS; also known as fibromyalgia). Pain and tenderness in AMPS are widespread, unlike the more focal findings related to arthritis and enthesitis in SpA. The response to nonsteroidal antiinflammatory drugs (NSAIDs) also differs, being generally less favorable for patients with AMPS and more favorable for those with SpA. Importantly, the presence of central pain is not mutually exclusive with SpA, as 10 to 45 percent of cases of ankylosing spondylitis (AS) are complicated by chronic widespread pain or fatigue [72,73]. (See "Fibromyalgia in children and adolescents: Clinical manifestations and diagnosis" and "Overview of chronic widespread (centralized) pain in the rheumatic diseases".)

●Malignancy – The presence of bone pain with or without limited areas of swelling can also signify malignancy. In a comparative study of 680 children with juvenile idiopathic arthritis (JIA) and 1277 with malignancy, features predictive of malignancy included bone pain, weight loss, thrombocytopenia, monoarticular involvement, hip disease, and male sex [74]. Of note, many of these findings can be seen in SpA. In such cases, imaging and consideration of a potential biopsy are imperative.

●Infection – Septic arthritis must always be on the differential diagnosis for patients presenting with joint pain and swelling, especially when there is monoarticular involvement. In patients presenting with isolated knee arthritis, Lyme arthritis needs to be excluded in endemic areas. (See "Musculoskeletal manifestations of Lyme disease".)

The sacroiliac joint may rarely be infected by organisms including tuberculosis and brucellosis; patients typically present acutely and have high fevers, unlike patients with SpA. Presence of extracapsular fluid collections, localized myositis, and/or apparent abscesses on imaging all support infectious rather than inflammatory sacroiliitis [75-77]. When there is high suspicion for infection, contrast administration is indicated.

Bacterial osteomyelitis in the spine and pelvis can also mimic SpA; fever and difficulty bearing weight may be present. (See "Hematogenous osteomyelitis in children: Clinical features and complications".)

An overview of the causes of back pain in children and adolescents is presented separately. (See "Back pain in children and adolescents: Causes".)

TREATMENT OF ENTHESITIS RELATED ARTHRITIS — 

Treatment of children with most types of juvenile idiopathic arthritis (JIA) is similar. This section will focus on enthesitis related arthritis (ERA) (algorithm 1). Treatment of children with other types of spondyloarthritis (SpA; eg, psoriatic JIA [psJIA], arthritis related to inflammatory bowel disease [IBD]) and certain extraarticular manifestations (eg, uveitis) is discussed separately:

●(See "Psoriatic juvenile idiopathic arthritis: Management and prognosis".)

●(See "Arthritis related to inflammatory bowel disease: Treatment and prognosis".)

●(See "Juvenile idiopathic arthritis: Immunizations and complications", section on 'Chronic anterior uveitis'.)

Data to support the use of specific therapies in patients with ERA are very limited and include a small number of randomized controlled trials, which are summarized below. The American College of Rheumatology (ACR) and Arthritis Foundation (AF) issued guidance for the treatment of JIA in 2019 that includes recommendations for patients with polyarthritis, sacroiliitis, and enthesitis who do not have systemic JIA (sJIA) [78]. Our approach is consistent with these guidelines. In addition, many clinicians extrapolate from studies of adults with SpA in light of the clinical and genetic similarities between pediatric and adult SpA [79,80].

Initial treatment — Initial treatment of patients with ERA typically includes nonsteroidal antiinflammatory drugs (NSAIDs) and, for those with peripheral arthritis, a disease-modifying antirheumatic drug (DMARD) (algorithm 1). Intraarticular glucocorticoids may be added when there is moderate to severe peripheral arthritis affecting ≤2 peripheral joints that are large enough for injection.

NSAIDs for most patients — In most children with ERA, we offer NSAIDs as bridging therapy, especially if the diagnosis is unclear. NSAIDs can provide symptomatic relief and reduce inflammation for patients with various manifestations of ERA, including axial disease, peripheral arthritis, and enthesitis, although they are rarely used as long-term monotherapy in pediatric patients.

The dosing of NSAIDs for treatment of ERA is the same as that used for other types of JIA; examples include:

●Ibuprofen 10 mg/kg/dose orally three to four times a day (maximum 2400 mg/day)

●Naproxen 5 to 10 mg/kg/dose orally twice a day (maximum 1000 mg/day)

●Celecoxib 50 mg orally twice a day for patients weighing 10 to 25 kg, or 100 mg orally twice a day in patients weighing >25 kg

Topical diclofenac is an alternative option for patients who do not tolerate oral NSAIDs and is often most helpful for patients with small joint arthritis (eg, fingers) and/or enthesitis in areas that are close to the overlying skin (eg, wrist). Contraindications to and adverse effects of NSAIDs are discussed in detail elsewhere. (See "Nonselective NSAIDs: Overview of adverse effects" and "Overview of COX-2 selective NSAIDs", section on 'Toxicities and possible toxicities'.)

Data to support the use of NSAIDs in JIA is very limited. In a double-blinded trial, 225 children (ages 2 to 16 years) with JIA were randomly assigned to one of three groups: lower-dose meloxicam (0.125 mg/kg daily), higher-dose meloxicam (0.25 mg/kg daily), or naproxen (10 mg/kg twice a day) [81]. After 12 months, symptoms improved in all three groups compared with baseline, and there were no significant differences between treatment groups. Regarding medication adverse effects, a prospective observational study of patients with JIA found similar safety profiles with celecoxib and other types of NSAIDs (most commonly naproxen, meloxicam, and nabumetone) [82].

csDMARDs for patients with peripheral arthritis — In patients with ERA who have peripheral arthritis, we also start a conventional synthetic DMARD (csDMARD), specifically methotrexate or sulfasalazine. The choice between agents should be individualized based on contraindications, patient and family preference, and provider experience. csDMARDs are not effective for axial SpA, which is reflected in both pediatric and adult treatment guidelines [83,84].

●Methotrexate – The dosing of methotrexate for ERA is the same as that for other types of JIA, given once a week as 10 to 15 mg/m² or 0.5 to 1 mg/kg (maximum 25 mg per week), either subcutaneously or orally [85]. It should be given with daily folic acid supplements (1 mg/day, optionally skipping the day that methotrexate is given) [85]. Common adverse effects include nausea and fatigue, which may be especially prominent in children. More information on contraindications, adverse effects, and monitoring of methotrexate is provided separately. (See "Use of methotrexate in the treatment of rheumatoid arthritis" and "Major adverse effects of low-dose methotrexate".)

Methotrexate is commonly used for other types of JIA and is preferred over other csDMARDs for treatment of polyarticular arthritis in the ACR/AF 2019 treatment guidelines [78]. Data to support the use of methotrexate in the treatment of ERA are extrapolated from its use in other forms of JIA. (See "Polyarticular juvenile idiopathic arthritis: Treatment and prognosis", section on 'Methotrexate'.)

●Sulfasalazine – The dosing of sulfasalazine for ERA is the same as that for other types of JIA, specifically 15 to 30 mg/kg/dose orally twice a day (maximum 3 g/day). Potential adverse effects include bone marrow suppression and, rarely, Stevens-Johnson syndrome, which are described in more detail elsewhere. (See "Sulfasalazine: Pharmacology, administration, and adverse effects in the treatment of rheumatoid arthritis".)

Data to support sulfasalazine specifically for patients with ERA are limited. In a trial of 33 patients with childhood-onset seronegative enthesopathy and arthropathy (SEA) syndrome or ankylosing spondylitis (AS), both those who were randomly assigned to receive sulfasalazine and those assigned to placebo experienced improvement; the only significant differences were in patient and clinician assessments of treatment efficacy, which were more favorable for patients receiving sulfasalazine [86]. Other evidence can be extrapolated from studies of patients with JIA, where there have been several randomized trials [87,88]. For adults with SpA, sulfasalazine is the preferred csDMARD [84].

Intraarticular glucocorticoid injections for selected patients — Patients with ERA who have arthritis in ≤2 peripheral joints that are amenable to injection may benefit from intraarticular glucocorticoid injections in addition to or in lieu of NSAIDs. This approach is often used when arthritis is severe enough to interfere with the patient's mobility. Intraarticular glucocorticoid injections are also sometimes used for patients with symptomatic sacroiliitis.

Potential benefits of local glucocorticoid injection include relatively fast symptomatic control compared with other treatments and a lesser degree of systemic immunosuppression. However, such injections can be challenging, especially in younger children, and may require sedation or even general anesthesia, especially in the case of sacroiliac joint injections. More information about performing intraarticular glucocorticoid injections in children and the dosing of medication is provided separately. (See "Joint aspiration or injection in children: Indications, technique, and complications" and "Intraarticular and soft tissue injections: What agent(s) to inject and how frequently?".)

Data to support the use of intraarticular glucocorticoids in ERA are limited. In an observational cohort of 220 patients with JIA (including four with ERA) who received intraarticular glucocorticoids for peripheral arthritis, one-third of patients experienced sustained remission over a median of 0.9 years; disease flares were more common in patients with polyarticular (versus oligoarticular) arthritis, those who were not concomitantly taking methotrexate, and those who had a negative antinuclear antibody (ANA) and/or an elevated C-reactive protein (CRP) [89]. Data to support sacroiliac joint injections come from another retrospective study of 50 patients with sacroiliitis related to JIA or IBD, which noted clinical improvement in two-thirds of patients by three months [90]. However, the procedures required imaging guidance and general anesthesia or sedation, and 94 percent of patients still required other immunosuppressive treatments after two years.

Assessing response to treatment — Patients with ERA should be followed closely for improvement in their symptoms and activities, typically with a follow-up visit after four to eight weeks. The frequency of subsequent visits depends on the severity of disease and the treatments being used but is typically every 8 to 12 weeks until patients have achieved remission. Children may not complain of pain even when they have active arthritis, underscoring the importance of careful physical examination [91].

Inflammatory markers (ie, erythrocyte sedimentation rate [ESR] and CRP) are imperfect measures of disease activity as they may be normal despite ongoing inflammation; however, in patients who have elevations at baseline, we continue to trend them as one measure of disease activity. Patients may also require laboratory testing for drug monitoring, depending on the type and duration of therapy (eg, checking electrolytes and kidney function in a patient on prolonged NSAIDs).

Imaging is not routinely used to assess response to treatment. However, it is a useful tool when patients have symptoms suggestive of ongoing inflammation without objective findings on physical examination. The choice of imaging study is guided by the type of symptom, as discussed elsewhere in this topic. (See 'Imaging' above.)

Treatment of persistent or refractory disease — In general, we consider patients to have refractory disease if they have axial arthritis that does not respond to an NSAID within four to six weeks, peripheral arthritis that does not respond to a csDMARD within three to four months, and/or require long-term use of NSAIDs. We add a biologic DMARD (bDMARD) for most patients who have refractory disease (algorithm 1).

bDMARD for most patients — In patients with ERA who have persistent axial arthritis despite initial therapy with NSAIDs, we recommend a bDMARD, specifically a tumor necrosis factor (TNF) or interleukin 17 (IL-17) inhibitor, rather than starting a csDMARD. This is due to data in adults indicating that csDMARDs are ineffective for axial SpA [92,93]. Similarly, we add a TNF or IL-17 inhibitor for patients with persistent peripheral arthritis who cannot tolerate or do not respond to methotrexate or sulfasalazine.

In the absence of evidence to strongly support one agent over the other, the choice of bDMARD agent is often dependent on a combination of patient and family preferences, provider experience, and occasionally regulatory or payor requirements.

●TNF inhibitors – There are various TNF inhibitors, including biosimilars, that vary somewhat in their structure, available formulations, and frequency of dosing (table 2). While they are not specifically approved for ERA by the US Food and Drug Administration (FDA), they are approved for patients with polyarticular JIA. Dosing is the same as that used to treat other types of JIA (table 3). Etanercept is not effective for the treatment of uveitis or IBD [94,95]. Important adverse effects of TNF inhibitors include an increased risk of infection, potential induction of anti-drug antibodies and autoantibodies, and a possible increased malignancy risk prompting a warning label from the FDA; these are discussed in detail elsewhere. (See "Tumor necrosis factor-alpha inhibitors: An overview of adverse effects" and "Tumor necrosis factor-alpha inhibitors: Risk of malignancy".)

Several randomized trials support the use of various TNF inhibitors in the treatment of ERA. In a double-blinded trial of 46 children (ages 6 to 17 years old) with ERA, patients who were randomly assigned to adalimumab had a greater improvement in the active joint count after 12 weeks compared with those who received placebo (-63 percent versus -12 percent) [96]. The efficacy of etanercept was evaluated in a trial in which 38 patients with ERA who responded to open-label treatment with etanercept for 24 weeks were randomly assigned to continue etanercept or to switch to placebo for another 24 weeks [97]. The rate of flare was significantly lower among those receiving etanercept compared with those who received placebo (15 versus 50 percent, respectively).

●Secukinumab – Secukinumab is an IL-17 inhibitor. Dosing is the same as that used for psJIA (table 3). The adverse effects of secukinumab are generally similar to those of other bDMARDs but do include higher rates of candida infections as well as new-onset IBD. In the United States, it is the only FDA-approved therapy for ERA.

In a trial of 86 patients with ERA or psJIA, patients received secukinumab for 12 weeks; those who responded were then randomly assigned to continue secukinumab or switch to placebo for up to 100 weeks [98]. The number of flares was lower (27 versus 55 percent) and the time to disease flare significantly longer in patients who continued secukinumab compared with those who received placebo.

Bridging therapy with oral glucocorticoids — Oral glucocorticoids may occasionally be used for short-term management (<1 month) of moderate to severe peripheral arthritis, sacroiliitis, and/or enthesitis when patients are escalating to systemic immunosuppression (ie, csDMARDs or bDMARDs).

There is no established dosing regimen for glucocorticoids; however, we typically use 10 to 20 mg/day of prednisone for no more than two weeks. Treatment guidelines recommend against long-term use of systemic glucocorticoids in children with JIA due to the cumulative risk of adverse effects and the availability of better-tolerated alternative therapies [83]. Adverse effects of glucocorticoids, including some that are specific to their use in children, are discussed in detail elsewhere. (See "Major adverse effects of systemic glucocorticoids", section on 'Young children'.)

Adjunctive physical therapy — Physical therapy may be a helpful adjunctive therapy for patients with ERA who are at risk of developing functional limitations due to their disease manifestations, especially when there are comorbid conditions such as hypermobility or pes planus.

Evidence to support physical therapy is very limited, but it is recommended by both the adult and pediatric treatment guidelines [83,84] and further supported by the reported association between mechanical stress and an increased risk of worsening disease activity or progression. (See 'Pathogenesis' above.)

Investigative approaches — Several therapies are actively being investigated for the treatment of JIA and/or ERA, including the following:

●Janus kinase (JAK) inhibitors – Tofacitinib and other JAK inhibitors appear to be effective in adult SpA, but data in pediatric SpA are limited [99,100]. A trial of tofacitinib in patients with JIA did include a prespecified exploratory analysis of 21 patients with ERA and noted a reduction in disease activity during the 18-week open-label period in which all patients received tofacitinib [101]. During the subsequent 44-week period, the rate of flare was 44 percent in patients randomly assigned to continue tofacitinib versus 57 percent for patients who received placebo. Tofacitinib is approved by the FDA for the treatment of polyarticular JIA [102], but further studies are needed to understand its use in other subtypes of JIA. In particular, very limited data suggest that tofacitinib might be relatively less effective to treat polyarticular arthritis in patients with JIA who are positive for human leukocyte antigen B27 (HLA-B27) versus those who are negative [103].

●IL-12/23 inhibitors – IL-12/23 inhibitors are likely to be effective for peripheral arthritis in patients with ERA based on their use in patients with psJIA. However, these agents are not effective in the treatment of axial SpA in adults [104,105], and their role in the management of pediatric axial SpA is uncertain. They should be considered first over IL-17 inhibitors in children with peripheral SpA and known or suspected IBD.

IMMUNIZATIONS — 

Many children with spondyloarthritis (SpA) require systemic immunosuppression, which both makes them more vulnerable to infection and decreases the response to immunizations. The impact of various types of immunosuppression on vaccine immunogenicity in children is presumably similar to that seen in adults (table 4). If disease activity is well controlled, it may be appropriate to temporarily hold immunosuppression while patients are receiving vaccinations. Further discussion of vaccination in patients with juvenile idiopathic arthritis (JIA) and other types of rheumatic diseases, including the use of live vaccinations, is provided separately. (See "Juvenile idiopathic arthritis: Immunizations and complications", section on 'Immunizations' and "Immunizations in autoimmune inflammatory rheumatic disease in adults".)

MONITORING FOR ASSOCIATED CONDITIONS — 

Children with spondyloarthritis (SpA) may develop related conditions such as uveitis, psoriasis, and/or inflammatory bowel disease (IBD). Children with SpA may develop acute anterior uveitis (AAU), which is typically associated with human leukocyte antigen B27 (HLA-B27) positivity, or clinically asymptomatic uveitis. Monitoring for these conditions is discussed separately. (See "Psoriatic juvenile idiopathic arthritis: Management and prognosis", section on 'Complications and associated conditions'.)

Some patients with SpA develop juvenile primary fibromyalgia syndrome (JPFS). The clinical manifestations and diagnoses of JPFS, as well as the relationship between chronic rheumatic conditions and fibromyalgia, are discussed elsewhere. (See "Fibromyalgia in children and adolescents: Clinical manifestations and diagnosis" and "Overview of chronic widespread (centralized) pain in the rheumatic diseases".)

COURSE AND PROGNOSIS — 

Before biologic disease-modifying antirheumatic drugs (bDMARDs) became widely available, children with spondyloarthritis (SpA) had worse outcomes in general than children with other forms of seronegative and nonsystemic juvenile idiopathic arthritis (JIA). As an example, a 2006 study of adults with longstanding JIA demonstrated poorer function among patients with enthesitis related arthritis (ERA) compared with those who had other subtypes of JIA [106]. Even among patients with SpA, those who experience the onset of disease in childhood may have a worse prognosis than those who manifest in adulthood, including having a greater risk for hip replacement [107].

Although outcomes overall have improved with the introduction of advanced therapies, observations of less optimal outcomes among children with SpA have remained. A study at a single center showed that children with ERA were less likely to attain remission following initiation of tumor necrosis factor (TNF) inhibitors compared with children with other diseases treated with TNF inhibitors [108]. This observation was confirmed by analysis of the Childhood Arthritis and Rheumatology Research Alliance (CARRA) database, which demonstrated that, after adjustment for multiple factors including medication use, a diagnosis of ERA was a predictor of higher pain intensity and poor self-reported health status [109]. Likewise, a Canadian cohort including over 1000 children with JIA found that children with ERA were less likely than those with seronegative JIA to enter remission [9]. Further studies are needed to better define the long-term outcomes for children with SpA in the modern era.

The prognosis of psoriatic JIA (psJIA) and complications that can affect patient with multiple forms of JIA are discussed separately. (See "Psoriatic juvenile idiopathic arthritis: Management and prognosis" and "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: Spondyloarthritis" and "Society guideline links: Juvenile idiopathic arthritis" and "Society guideline links: Uveitis".)

SUMMARY AND RECOMMENDATIONS

●Terminology and classification – Spondyloarthritis (SpA) refers to multiple subtypes of inflammatory arthritis that affect patients across the age spectrum, including ankylosing spondylitis (AS), psoriatic arthritis, reactive arthritis, and arthritis related to inflammatory bowel disease (IBD). SpA in children age <16 years old is considered a subtype of juvenile idiopathic arthritis (JIA) and is further classified as enthesitis related arthritis (ERA), psoriatic JIA (psJIA), or undifferentiated JIA. (See 'Terminology and classification' above.)

●Clinical presentation – Patients with SpA may have axial arthritis involving the spinal vertebrae (spondylitis) and/or the sacroiliac joint (sacroiliitis). Peripheral arthritis in children with SpA is most often asymmetrical and oligoarticular (in four or fewer joints), commonly affecting the lower limbs. Enthesitis (inflammation at the site where a tendon inserts into the bone) is usually symmetric and frequently occurs in the ankle and knee. Extraarticular manifestations of SpA include uveitis, psoriasis, intestinal inflammation (including IBD), and, rarely, aortic insufficiency. (See 'Clinical presentation' above.)

●Evaluation and diagnosis

•When to suspect – SpA should be suspected in a child who presents with axial and/or peripheral arthritis or enthesitis, especially arthritis of the large joints of the lower extremities. (See 'When to suspect' above.)

•History, physical examination, and laboratory testing – Evaluation for suspected SpA includes a history and physical examination with attention to evidence of arthritis, enthesitis, and related conditions (eg, psoriasis). Laboratory testing is not specific for SpA but can help evaluate the extent of peripheral inflammation and evaluate for alternative causes of inflammatory arthritis. (See 'History and physical examination' above and 'Laboratory testing' above.)

•Axial imaging for patients with back pain– If there is suspicion of axial involvement, we obtain MRI of the sacroiliac joints without gadolinium. We do not obtain plain radiographs to screen for sacroiliitis, unless mandated for cost or regulatory reasons. (See 'Axial imaging for patients with back pain' above.)

•Establishing the diagnosis – The diagnosis of SpA in children is made clinically based on a combination of clinical features (eg, arthritis, enthesitis, and/or sacroiliitis) and a personal or family history of related conditions (eg, psoriasis, acute anterior uveitis [AAU], AS, IBD). Most patients who have axial arthritis and/or enthesitis on imaging, dactylitis, and/or arthritis with a personal history of psoriasis, AAU, or IBD will be categorized as having SpA. Human leukocyte antigen B27 (HLA-B27) positivity is also suggestive of but not specific for SpA. (See 'Establishing the diagnosis' above.)

●Treatment of ERA – Our approach to therapy is outlined in the algorithm (algorithm 1).

•Initial treatment

-Nonsteroidal antiinflammatory drugs (NSAIDs) for most patients – We start most patients with ERA on an NSAID. In patients with ERA with axial involvement who do not have peripheral involvement, we suggest using an NSAID as the initial therapy rather than a conventional synthetic disease-modifying antirheumatic drug (csDMARD) or biologic DMARD (bDMARD) (Grade 2C). NSAIDs are rarely used as long-term monotherapy in pediatric patients. (See 'NSAIDs for most patients' above.)

-csDMARDs for patients with peripheral arthritis – In most patients with ERA who have peripheral arthritis, we add a csDMARD, specifically methotrexate or sulfasalazine, rather than using an NSAID alone or adding a bDMARD (Grade 2C).

-Intraarticular glucocorticoid injections for selected patients – In most patients with ERA who have arthritis in one or two peripheral joints that are amenable to injection, we suggest intraarticular glucocorticoid injections in addition to NSAIDs and a csDMARD, rather than NSAIDs and a csDMARD alone (Grade 2C). This approach is often used when arthritis is severe enough to interfere with the patient's mobility. (See 'Intraarticular glucocorticoid injections for selected patients' above.)

•Treatment of persistent or refractory disease – In general, we consider patients to be candidates for systemic immunomodulatory therapy if they do not respond to NSAIDs within four to six weeks or csDMARDs within three to four months, or require long-term use of NSAIDs. (See 'Treatment of persistent or refractory disease' above.)

-bDMARDs for most patients – In patients with ERA who have persistent axial arthritis despite initial therapy with NSAIDs, we recommend a tumor necrosis factor (TNF) or interleukin 17 (IL-17) inhibitor rather than starting a csDMARD (Grade 1B). Similarly, in patients with ERA who have persistent peripheral arthritis who cannot tolerate or do not respond to a csDMARD, we add a TNF or IL-17 inhibitor. (See 'bDMARD for most patients' above.)

ACKNOWLEDGMENT — 

The UpToDate editorial staff acknowledges Thomas JA Lehman, MD and Sarah Ringold, MD, MS, who contributed to earlier versions of this topic review.