Evaluation and Treatment of Enthesitis-Related Arthritis

Juvenile idiopathic arthritis (JIA) is the most common chronic pediatric rheumatic disease and the most common cause of acquired disability in childhood in the US [1–6]. Incidence and prevalence rates vary significantly by geographical location, with an estimated annual incidence of approximately one per 100 000 children in Japan [7], 90 per 100 000 children in the US [8], and 170 per 100 000 in Belgium [9]. These widely disparate rates may reflect differences in genetic susceptibility, environmental influences, and other unknown factors. The term JIA encompasses seven categories of disease that involve arthritis of at least one joint that begins before the age of 16 years and lasts ≥6 weeks. Distinct clinical features characterize each of the JIA categories during the first 6 months of disease. The enthesitis-related arthritis (ERA) category of JIA describes a heterogeneous group of children, including those with enthesitis and arthritis, inflammatory bowel disease (IBD)-associated arthropathy, and what is traditionally thought of as juvenile ankylosing spondylitis (AS). ERA accounts for approximately 15–20% of JIA cases and has a peak age of onset of 12 years [10–13]. Boys are affected more often than girls, accounting for 60% of cases [13]. Approximately half of children are human leukocyte antigen-B27 (HLA-B27)-positive and 20% have a family history of HLA-B27-associated disease, such as reactive arthritis, AS, or IBD with sacroiliitis. Tests for rheumatoid factor (RF) and anti-nuclear antibodies (ANA) are characteristically negative in patients with ERA. ERA is similar to but not interchangeable with juvenile spondyloarthropathy (SpA). Juvenile SpA includes not only children with ERA but also many who do not meet ERA criteria, including patients with the International League of Associations for Rheumatology (ILAR) undifferentiated and psoriatic arthritis (PsA) categories, IBD-related arthritis, juvenile AS, and reactive arthritis.

Clinical characteristics

According to ILAR criteria, ERA is defined as arthritis and enthesitis of ≥6 weeks’ duration in children aged <16 years, or arthritis or enthesitis plus two of the following: sacroiliac tenderness or inflammatory spinal pain, HLA-B27 positivity, onset of arthritis in boys aged >6 years, anterior uveitis associated with pain, redness, or photophobia and family history of HLA-B27-associated disease [14]. Clinical manifestations include arthritis, enthesitis, and symptomatic uveitis.

Arthritis

With the exception of hip and sacroiliac arthritis, articular involvement is primarily peripheral. During the first 6 months of disease, the arthritis is most often asymmetrical and oligoarticular (fewer than four joints) [13]. The most frequently affected joints are the ankles, knees, and hips [13,15]. Tarsitis, or inflammation of the small joints of the midfoot, is not uncommon and is highly suggestive of the diagnosis (Figure 1). As many as one-third of children with juvenile SpA, a condition that encompasses most children with ERA, have tarsitis at disease onset [16].

Figure 1.

Tarsitis in enthesitis-related arthritis. A: Sagittal fluid-sensitive image of the lateral aspect of the foot showing edema within the midfoot and forefoot bones with surrounding soft tissue swelling. The arrow depicts the cuboid with the arrowhead at the fifth tarsometatarsal joint. B: Coronal fluid-sensitive image through the midfoot demonstrating bone marrow edema within the middle cuneiform, lateral cuneiform, and the cuboid. The arrow depicts the middle cuneiform. There is edema within the dorsolateral aspect of the foot (arrowhead) as well as the deep plantar compartment of the midfoot (*).

Axial, or spinal, involvement is uncommon at diagnosis [17,18]. However, 30–50% of children with established ERA have clinical or radiographic evidence of sacroiliitis (inflammation of the sacroiliac joints [SIJs]) [13,15]. In one study, as many as 92% of children with juvenile SpA developed sacroiliitis within 5 years of diagnosis [19]. Sacroiliitis or spondylitis (inflammation of the vertebrae) may progress to ankylosis or fusion of the involved joints.

Enthesitis

Enthesitis is a distinct pathological feature of ERA and is an independent risk factor for increased pain in children with the disease [20]. It is defined as inflammation at the site where a tendon, ligament, or joint capsule attaches to bone. Enthesitis can also be seen in other categories of JIA, including PsA, undifferentiated JIA, and extended oligoarticular JIA [20]. Overall, 45–80% of children with ERA have at least one tender enthesis [13,21]. Enthesitis in ERA is often symmetrical and predominantly affects the lower limbs [15]. In one study, tenderness was most common at the patellar ligament insertion at the inferior pole of the patella (50% of patients), the plantar fascia insertion at the calcaneus (38%), and the Achilles tendon insertion at the calcaneus (22%) [13]. In another study, the hip extensor insertion at the greater trochanter was the most common site [15]. Manifestations of chronic enthesitis may include erosions, calcifications, osteopenia, and new bone formation [22,23]. Enthesitis may not respond as well to therapy as peripheral arthritis [24]. In an inception cohort of ERA patients, the odds ratio for having persistent enthesitis at 6 months after diagnosis increased significantly with the number of tender entheses at disease onset (odds ratio 2.18, 95% confidence interval 1.11–4.28) [13].

Acute anterior uveitis

The eye inflammation associated with ERA is typically acute in onset and characterized by conjunctival injection, pain, and photophobia. This is in contrast to the uveitis seen in the other JIA categories, which is typically asymptomatic. Uveitis in ERA is often unilateral and recurrent. The cumulative incidence of uveitis in children with ERA over the course of the disease is approximately 30% [25]. An estimated 10% of children develop uveitis in the first 6 months of the disease [13]. Persistent uveitis may cause complications including iris scarring, corneal calcium deposition, glaucoma, cataracts, macular edema, and visual loss (Figure 2).

Figure 2.

Enthesitis-related arthritis-associated uveitis. The image shows cataract and anterior lens deposits secondary to recurrent human leukocyte antigen-B27-associated uveitis.

Evaluation

Laboratory testing

The diagnosis of ERA is predominantly clinical as there is no confirmatory laboratory test. HLA-B27 is strongly associated with but not diagnostic for ERA. Only half of children with ERA are HLA-B27-positive [13]. ANA, RF, and anti-cyclic citrullinated peptide antibodies are characteristically absent. Inflammatory marker levels may be normal or elevated.

Methods to detect enthesitis

Enthesitis in children is typically defined as localized pain, tenderness, or swelling on physical examination. However, physical examination misses many cases of enthesitis that are confirmed on ultrasound [15,26,27]. Normal children may also have tender entheses on examination, particularly at the plantar fascia insertions along the metatarsal heads [28]. The positive and negative predictive values of physical examination for enthesitis at the Achilles tendon in adults, using ultrasonography with power Doppler (USD) as the gold standard, are 0.55 and 0.73, respectively [26]. Indices that have been validated in adults to assess enthesitis include the Mander index [29], a modified Mander index [30], the SpA Research Consortium of Canada Enthesitis Index [31], the Maastricht AS Enthesitis Score [32], the Leeds Enthesitis Index [33], and the Major Enthesitis Index [34] (Table 1). A standard enthesitis index for use in children has not yet been developed.

Table 1.

Standard enthesitis indices for adults.

	Spondyloarthritis Research Consortium of Canada Enthesitis Index	Maastricht Ankylosing Spondylitis Enthesitis Score	Leeds Enthesitis Index	Major Enthesitis Index
Sites per index	14	13	6	12
Supraspinatus insertion at greater tuberosity of humerus	√
Common flexor insertion at medial epicondyle of humerus	√			√
Common extensor insertion at lateral epicondyle of humerus	√		√	√
First costochondral joint		√
Seventh costochondral joint		√
Posterior superior iliac spine		√
Sartorius insertion at anterior superior iliac spine		√
Iliac crest		√		√
Fifth lumbar spinous process		√
Posterior superior iliac spine		√
Hip extensor insertion at greater trochanter of femur				√
Quadriceps insertion at superior border of patella	√
Medial femoral condyle			√
Patellar ligament insertion at inferior pole of patella	√
Achilles tendon insertion at calcaneus	√	√	√	√
Plantar fascia insertion at calcaneus	√			√

USD and magnetic resonance imaging (MRI) are increasingly being used for the early detection and monitoring of disease activity in adults with SpA. These modalities may help to differentiate inflammatory enthesitis from other non-inflammatory conditions such as mechanical injuries, apophysitis, and fibromyalgia [26,35,36]. The most common ultrasound abnormalities include increased power Doppler signal (Figure 3), enthesophytes, calcifications, tendon thickening, and hypoechogenicity [37]. USD and MRI seem to be promising in the detection of disease activity in children with ERA. In one study, 70% of subjects with ERA had enthesitis detectable by USD [27]. However, half of the sites with positive USD findings were normal on clinical examination. Whole-body MRI has also been used in children for the assessment of both joint and enthesis imaging (Figure 4). Whole-body MRI results suggest that physical examination overestimates inflammatory enthesitis in children [15]. In one study, although 52% of the children had tenderness on physical examination, enthesitis was confirmed by whole-body MRI in only 26% [15].

Figure 3.

Power Doppler ultrasound of entheses. A: Sagittal grayscale ultrasound image of the patellar tendon demonstrates a normal appearance with the expected fibrillary pattern (arrows). The patella (*) appears normal with intact cartilage (arrowhead). B: Sagittal power Doppler ultrasound image of the patellar tendon demonstrates increased hyperemia of the patellar tendon (arrow) at the insertion along the inferior patellar pole (*) consistent with inflammation.

Figure 4.

Magnetic resonance image of entheses. A: Axial fluid-sensitive image demonstrates increased abnormal signal intensity at the tendon insertions at the greater tuberosities (arrows), left greater than right. B: The coronal fluid-sensitive image demonstrates increased, abnormal signal intensity at the insertions of the hamstrings at the ischial tuberosities (arrows), right greater than left.

Methods to detect sacroiliitis

Inflammatory back pain, defined as lower back pain that starts insidiously, improves with exercise, and is associated with >30 min of morning stiffness or alternating buttock pain, should raise suspicion of sacroiliitis [38,39]. The physical examination findings characteristic of sacroiliitis are tenderness on direct compression over the SIJ, reduced lumbar flexion, reduced lateral flexion, and a positive Patrick test (a maneuver that causes pain in the contralateral SIJ with downward pressure on the knee when it is flexed and the hip is externally rotated, flexed, and abducted). Lumbar and sacroiliac flexion is measured with the modified Schober test [40]. To perform this test, the patient stands erect while horizontal lines are drawn between the dimples of Venus, 10 cm above and 5 cm below the dimples. The patient then bends forward and the distance between the upper and lower lines is recorded. A modified Schober result <6 cm (e.g. an increase from 15 cm to <21 cm) is considered abnormal [21]. To measure lateral spinal flexion, the patient stands erect with the heels and back against the wall. The examiner places a mark on the thigh at the end of the third finger while the patient is erect and places a second mark after the patient leans sideways without bending the knees or lifting the heels. The result is recorded as abnormal if the change is <20 cm on either side [41].

In adults, the positive and negative predictive values of physical examination to detect sacroiliitis are poor [42,43]. Consequently, the European SpA Study Group and Assessment in SpA International Society classification criteria and the Amor criteria for axial disease in adult SpA require conventional radiographs or MRI to confirm the diagnosis of sacroiliitis [44]. Multiple modalities have been used to assess sacroiliitis in adults. Conventional radiographs and computed tomography can detect chronic bony changes but cannot detect active inflammation, and involve a lot of radiation exposure [45]. USD has been proposed as an alternative and less expensive method to evaluate sacroiliitis; however, it lacks sufficient resolution and penetration depth to adequately and reliably visualize the SIJs [46–49]. The current gold standard for sacroiliac imaging in adults and children is MRI with short-tau inversion recovery (STIR) [15,44,50–54]. Recent studies have also demonstrated that STIR sequences without gadolinium are sufficient for establishing a diagnosis of sacroiliitis [55]. Characteristic findings on MRI include bone marrow edema within the sacrum and/or adjacent ilium with or without capsulitis (inflammation of the joint capsule) or enthesitis [44] (Figure 5).

Figure 5.

Magnetic resonance images of the sacroiliac joints in enthesitis-associated arthritis. A: The fluid-sensitive coned-down image of the sacroiliac joints shows bone marrow edema within the right and left sacral alae (arrows) with a small amount of fluid within the sacroiliac joints (arrowheads). B: The axial T1-weighted image of the sacroiliac joints shows bone marrow hypointensity within the right sacral ala consistent with bone marrow edema (arrow).

Treatment

Therapeutic options for ERA, which are primarily based on adult studies, include monotherapy or combination therapy with non-steroidal anti-inflammatory drugs (NSAIDs), disease-modifying anti-rheumatic drugs (DMARDs), and biological agents. NSAID monotherapy may be appropriate for children with low disease activity and without features of a poor prognosis [56]. Naproxen is often tried first for the treatment of enthesitis as it has a better toxicity profile than other NSAIDs [21]. Indomethacin or other NSAIDs in the indoleacetic acid group, such as diclofenac, sulindac, and tolmetin, are often tried if naproxen is ineffective. However, toxicity with indomethacin is common and side-effects may include headache and abdominal pain. NSAID toxicity monitoring should include a complete blood count, creatinine measurement, liver function test (LFT), and urinalysis 4–6 weeks after initiation and then every 6–12 months thereafter.

The efficacy and safety of methotrexate, a commonly used DMARD, has been established in patients with JIA [57,58]. The efficacy of methotrexate for enthesitis and sacroiliitis in children with ERA has not been specifically assessed. Several studies support the use of sulfasalazine in patients with ERA. In a randomized, double-blind, placebo-controlled study of juvenile SpA, there was no significant difference between the sulfasalazine and placebo groups in terms of active joint count, tender enthesis count, pain visual analogue scale, or spinal flexion [59]. However, children who received sulfasalazine had significant improvement in the physician and patient assessments. A second randomized trial demonstrated that sulfasalazine significantly improved articular severity scores, global disease assessments, and laboratory markers of inflammation [60]. Furthermore, this effect was sustainable [61]. The typical dose is 30–50 mg/kg/day and is titrated over several weeks. Clinical improvement is expected 6–8 weeks after initiation. Side-effects include gastrointestinal upset, cytopenias, hepatotoxicity, hypogammaglobulinemia, and Stevens–Johnson syndrome. Toxicity monitoring should include a complete blood count, LFT, and creatinine measurement prior to initiation, 4–6 weeks after initiation, and then every 3–4 months thereafter if the prior results are normal and the dose stable.

The biological agents currently approved for the treatment of the non-systemic forms of JIA are abatacept and the anti-tumor necrosis factor (anti-TNF) drugs etanercept and adalimumab. Anti-TNF agents are efficacious in adults for the treatment of arthritis, enthesitis, and axial disease. One randomized, placebo-controlled trial demonstrated that children with juvenile SpA who were treated with infliximab had significantly improved active joint counts, active enthesis counts, pain, and function [62]. Analyses from an observational pediatric Dutch registry demonstrated that anti-TNF agents were effective and safe in patients with ERA who were previously unresponsive to one or more DMARDs [63]. However, sustained remission was infrequent and no subject attained remission off medication. The typical etanercept dose is 0.8 mg/kg/week subcutaneously (max 50 mg). The typical adalimumab dose is 20 mg subcutaneously every other week in patients weighing <30 kg, and 40 mg every other week in those weighing >30 kg. Testing for latent tuberculosis should be performed prior to initiating anti-TNF therapy and yearly thereafter. Side-effects of anti-TNFs include infection, cytopenias, hypersensitivity reactions, psoriasis, demyelination, and malignancy.

Prognosis

Observational studies suggest that children with ERA are less likely than those with other JIA categories to attain inactive disease 1 year after treatment initiation [64]. Furthermore, in comparison with other JIA categories, ERA is associated with worse function, poorer quality of life, and increased pain intensity [20]. HLA-B27 positivity, tarsitis, hip arthritis, and older age at disease onset are associated with a poorer prognosis [65,66]. Five years after diagnosis, <20% of children with ERA attain disease remission [67].

Summary

The JIA category of ERA encompasses a heterogeneous group of children and accounts for approximately 15–20% of JIA cases. The ankles, knees, and hips are the most frequently affected joints. Tarsitis and axial arthritis are not uncommon and are highly suggestive of the diagnosis. In contrast to the uveitis that is seen in other JIA categories, eye inflammation associated with ERA is typically acute in onset and characterized by conjunctival injection, pain, and photophobia. Therapeutic options for ERA include monotherapy or combination therapy with NSAIDs, DMARDs, and biological agents. ERA is associated with worse function, poorer quality of life, and increased pain intensity compared with the other JIA categories. As the pathophysiology of ERA is elucidated, more targeted treatment options will hopefully enable better disease control and improved functional and patient-reported outcomes.