Wrist and Ankle MRI of Patients With Juvenile Idiopathic Arthritis: Identification of Unsuspected Multicompartmental Tenosynovitis and Arthritis

Transcription

1 Pediatric Imaging Original Research Javadi et al. Wrist and nkle MRI of Patients With Juvenile Idiopathic rthritis Pediatric Imaging Original Research Sanaz Javadi 1 J. Herman Kan 1 Robert C. Orth 1 Marietta DeGuzman 2 Javadi S, Kan JH, Orth RC, DeGuzman M Keywords: arthritis, juvenile idiopathic arthritis, MRI, pediatrics, tenosynovitis DOI: /JR Received January 24, 2013; accepted after revision May 15, Department of Pediatric Radiology, Texas Children s Hospital, 6701 Fannin St, Ste 470, Houston, TX ddress correspondence to J. H. Kan (jhkan@texaschildrens.org). 2 Department of Rheumatology, Texas Children s Hospital, Houston, TX. JR 2014; 202: X/14/ merican Roentgen Ray Society Wrist and nkle MRI of Patients With Juvenile Idiopathic rthritis: Identification of Unsuspected Multicompartmental Tenosynovitis and rthritis OJECTIVE. The purpose of this study was to characterize the extent of joint and tendon involvement in patients with juvenile idiopathic arthritis referred for MRI of the wrist or ankle. MTERILS ND METHODS. Forty-five patients (32 female and 13 male patients; mean age, 10 years; age range, 1 19 years) with an established diagnosis of juvenile idiopathic arthritis were referred for MRI of the wrist or ankle between January 2000 and ugust 2012 (39 wrists and 33 ankles). ll MRI examinations and clinical notes were reviewed, and joint and tendon involvement was recorded. RESULTS. Tenosynovitis was present in 50% (36/72) of examinations. Tenosynovitis was not documented in clinical notes before MRI. When tenosynovitis was present, an average of 3.5 tendons were involved (range, 1 12 tendons). For the wrist, 59% (23/39) had tenosynovitis, and the extensor digitorum tendon (23% [9/39]) was most commonly involved. For the ankle, 39% (13/33) had tenosynovitis, and the tibialis posterior tendon (33.3% [11/33]) was most commonly involved. For the wrist, 89.7% (35/39) had active joint inflammation with an average of 3.1 joints involved (range, 0 6 joints), and the intercarpal joint was most commonly involved (69% [27/39]). For the ankle, 69.7% (23/33) had active joint inflammation, with an average of 2.4 joints involved (range, 0 5 joints), and the tibiotalar joint (52% [17/33]) was most commonly involved. CONCLUSION. Multicompartmental tenosynovitis and arthritis involvement is common in patients with juvenile idiopathic arthritis referred for MRI of the wrist or ankle and is underappreciated on clinical examination. International League of ssociations for Rheumatology subclassification and targeted intraarticular steroid injections guided by clinical examination alone may lead to undertreatment or incorrect treatment of active disease. J uvenile idiopathic arthritis is a heterogeneous group of nonpyogenic chronic inflammatory arthritides with persistent symptoms for at least 6 weeks that are present before the child is 16 years old [1]. In certain juvenile idiopathic arthritis subtypes, the disease is rapidly progressive and, if left untreated, may lead to structural joint damage and permanent impairment of physical function. The goal of treatment is to control synovial inflammation and thereby preserve joint function. Intraarticular corticosteroid administration is an established technique for obtaining local control of synovial inflammation in children with juvenile idiopathic arthritis and may potentially decrease the need for systemic therapy [2]. Injections commonly target the small joints of the wrist and ankle according to clinical assessment [3, 4]. However, given the close proximity of multiple joints and tendon sheaths, clinical assessment alone before targeted steroid injections may underestimate the extent of disease. The purpose of this study was to characterize the extent of joint and tendon involvement in patients with juvenile idiopathic arthritis referred for MRI of the ankle or wrist. Materials and Methods This retrospective study was approved by our institution s investigational review board, and the need for informed consent was waived. Patients Patients with an established clinical diagnosis of juvenile idiopathic arthritis who underwent MRI of the wrist or ankle from January 2000 to ugust JR:202, February

2 Javadi et al. Twenty-five patients underwent MRI of more than one body part. For instance, one patient may have had a wrist and an ankle MRI, and this was counted as two separate studies. Fig. 1 8-year-old girl with systemic juvenile idiopathic arthritis and active joint inflammation of wrist without tenosynovitis. and, STIR coronal images in posterior () and anterior () views show effusions within radiocarpal (arrowhead, ), intercarpal (arrow, ), and first carpometacarpal (arrow, ) joints. TLE 1: Positive MRI Examinations in Patients With an Established Diagnosis of Juvenile Idiopathic rthritis Referred for Imaging natomic Location Examinations Percentage (No.) With rthritis 2012 were included in the study. For patients who underwent multiple MRI studies, only the first was evaluated. Patients were excluded from the study if joint or tendon involvement was related to other causes, including septic arthritis, osteomyelitis, and trauma. Patients with psoriatic arthritis, enthesitis-related arthritis, and undifferentiated arthritis were also excluded. Percentage (No.) With Coexisting rthritis and Tenosynovitis Percentage (No.) With Tenosynovitis Only Without rthritis nkle (23) 33.3 (11) 6.1 (2) Wrist (35) 51.3 (20) 7.7 (3) Total (58) 43.1 (31) 6.9 (5) MRI MRI was performed on a 1.5-T system (chieva, Philips Healthcare). The majority of patients had triplanar fluid-sensitive (T2-weighted with fat saturation, proton density-weighted with fat saturation, or STIR) and T1-weighted coronal or sagittal images of the area of interest using a dedicated wrist or ankle coil. In a minority of cases, a small surface coil was used (Flex-M coil, Philips Healthcare). Fifty-six percent of the patients (40/72) received gadolinium-based IV contrast agent (gadopentetate dimeglumine; Magnevist, ayer HealthCare) at 0.1 mmol/kg. Standard imaging included a slice thickness of 3 mm with a 0.3-mm skip. In a minority of patients, the study included hand or foot imaging in addition to images of the wrist or ankle. For patients who received IV contrast agent, T1-weighted fat-saturated images in at least two orthogonal planes were obtained. Clinical Data and Image nalysis Demographic information, including age and sex, was collected. The pediatric rheumatology notes were reviewed, and the clinical description and impression of disease extent before MRI were documented. Patients were subcategorized as having oligoarticular juvenile idiopathic arthritis, polyarticular (rheumatoid factor positive or negative) juvenile idiopathic arthritis, or systemic arthritis according to the pediatric rheumatology clinical notes. Information about temporal relationships between rheumatology visits and MRI examinations, as well as whether the child received any rheumatologic pharmacologic treatment before MRI, was also gathered. MRI examinations were analyzed by a Certificate of dded Qualification pediatric radiologist with 7 years of experience, with an additional year of pediatric musculoskeletal fellowship training. The locations of joint and tendon inflammation were recorded for each study. The criteria for active joint inflammation in the context of a symptomatic region based on a pediatric rheumatology examination included joint effusion, juxtaarticular soft-tissue or bone-marrow edema, or thick synovial enhancement (determined subjectively) if IV contrast agent was administered. For the wrist, joint compartments were individually counted for a total of five compartments. The second through fifth carpometacarpal, first carpometacarpal, intercarpal, radiocarpal, and distal radioulnar joints were counted as individual joints following a compartmental approach similar to that of Damasio et al. [5] (Fig. 1). For the ankle, joint compartments were individually counted, for a total of six compartments. The tibiotalar, subtalar, calcaneocuboid, talonavicular, naviculocuneiform, Fig. 2 5-year-old girl with polyarticular juvenile idiopathic arthritis, active joint inflammation, and tenosynovitis affecting both wrists., STIR axial image shows bilateral extensor compartment and left wrist flexor pollicis longus tenosynovitis., STIR coronal image of left wrist shows effusions in distal radioulnar, radiocarpal, and intercarpal joints with marrow edema involving capitate (arrow). 414 JR:202, February 2014

3 Wrist and nkle MRI of Patients With Juvenile Idiopathic rthritis and the tarsometatarsal joints were each counted as individual joints. The criteria for tenosynovitis were circumferential high signal intensity around the tendon on fluid-sensitive sequences, or enhancement of a thickened tendon sheath if IV contrast agent was administered. For the wrist, each tendon was counted individually (abductor pollicis longus, extensor pollicis brevis, extensor carpi radialis longus, extensor pollicis longus, extensor carpi radialis brevis, extensor digiti minimi, extensor carpi ulnaris, flexor carpi ulnaris, palmaris longus, flexor pollicis longus, and the flexor carpi radialis), except the flexor digitorum or extensor digitorum complexes. These tendon groups were each counted as a single tendon. The total number of tendon compartments for the wrist was 13. For the ankle, each tendon was counted individually (tibialis anterior, extensor hallucis longus, extensor digitorum longus, flexor hallucis longus, flexor digitorum longus, and tibialis posterior), except the peroneus complex, which was counted as a single tendon. The total number of tendon compartments for the ankle was seven. Fig. 3 8-year-old girl with oligoarticular juvenile idiopathic arthritis and isolated tenosynovitis of ankle without active joint inflammation., STIR axial image shows fluid within peroneus complex (arrow) as well as tibialis posterior (not shown) and flexor digitorum longus tendon sheaths (not shown)., STIR sagittal image shows absence of underlying active joint inflammation in ankle. Results Forty-five patients (32 female and 13 male patients; mean age, 10 years; age range, 1 19 years) met the study criteria and underwent 72 MRI examinations. Tenosynovitis was present in more than half of the patients (56% [25/45]) and in 50% of the studies (36/72) (Table 1). Tenosynovitis was not documented in the clinical notes of any patient before MRI. Patients were evaluated by a rheumatologist before MRI in 88% (63/72) of the studies, with an average interval of 36 days between the visit and MRI examination (range, days). In 46% (29/63) of the studies, patients were treated before MRI with both diseasemodifying antirheumatic drugs and nonsteroidal antiinflammatory drugs; in 29% (18/63) of studies, patients received only nonsteroidal antiinflammatory drugs; in 13% (8/63) of studies, patients received only disease-modifying antirheumatic drugs; and in 13% (8/63) of studies, patients did not receive any medical therapy between the rheumatology visit and MRI. In total, patients did not receive any treatment before MRI in 23.6% (17/72) of studies. When tenosynovitis was present (36 studies total), an average of 3.5 tendons per examination were involved (range, 1 12 tendons) (Table 1). For the wrist, 59% (23/39) had tenosynovitis. The most commonly involved tendons were the extensor digitorum (Fig. 2) (23% [9/39]) and extensor digiti minimi (23% [9/39]). For the ankle, 39% (13/33) had tenosynovitis. The most commonly involved tendons were the tibialis posterior (39% [11/33]) and the peroneus complex (Fig. 3) (18% [6/33]). ctive joint inflammation was present in 87% (39/45) of patients or 80.6% (58/72) of studies. For the wrist, 89.7% (35/39) of joints had active inflammation, with an average of 3.1 joints involved (range, 0 6 joints). The most commonly involved were the intercarpal (69% [27/39]), radiocarpal (64% [25/39]), and distal radioulnar (56% [22/39]) joints. For the Fig. 4 4-year-old boy with polyarticular juvenile idiopathic arthritis, active joint inflammation, and tenosynovitis of ankle., T1-weighted fat-saturated coronal contrast-enhanced image shows thick tendon sheath enhancement (arrows) affecting peroneus complex and ankle plantar flexor tendons. Thick pannus formation (arrowhead) is present in tibiotalar joint., STIR sagittal image shows synovial thickening of tibiotalar joint (arrowhead) and small subtalar effusion. JR:202, February

4 Javadi et al. Fig. 5 7-year-old-girl with oligoarticular juvenile idiopathic arthritis and active joint inflammation of ankle without tenosynovitis. STIR sagittal image shows tibiotalar (arrow) and talonavicular (arrowhead) joint effusions. There is also juxtaarticular bone marrow edema in talar head, navicular, and cuneiform bones. ankle, 69.7% (23/33) had active inflammation with an average of 2.4 joints involved (range, 0 5 joints). The most commonly involved were the tibiotalar (52% [17/33]) (Fig. 4), talonavicular (36% [12/33]) (Fig. 5), and subtalar (27% [9/33]) joints. Subset analysis of the number of joints and tendons involved in oligoarthritis, polyarthritis, and systemic juvenile idiopathic arthritis is detailed in Table 2. TLE 2: Joint and Tendon Involvement in Patients With Oligoarticular, Polyarticular, and Systemic Juvenile Idiopathic rthritis Type of Juvenile Idiopathic rthritis Patients Examinations ffected Joints per Examination ffected Tendons per Examination Oligoarticular Polyarticular Systemic Discussion MRI can reliably evaluate the extent of inflammation in patients with juvenile idiopathic arthritis, including the detection of synovitis, tenosynovitis, and active hypervascular pannus [6 10], and is also able to diagnose subclinical synovitis [11]. In our study, more than half of the patients (56%) had tenosynovitis according to MRI, which was not documented in rheumatologic clinical notes, and the majority had more than one inflamed joint compartment. MRI in children with established juvenile idiopathic arthritis is important for precise localization of inflammation rather than the clinical examination alone if targeted steroid injection for local control of disease is considered. The goal of local control of symptomatic sites with targeted steroid injections is to control inflammation to prevent early cartilage loss and loss of joint function [10]. Tendon involvement is common in patients with juvenile idiopathic arthritis [12]. Differentiating tenosynovitis and underlying arthritis may be challenging on the basis of clinical examination alone because these structures are in close proximity and both may cause diffuse swelling and decreased function. Rooney et al. [12] sonographically evaluated 49 swollen ankles in patients with established juvenile idiopathic arthritis and found that 71% had tenosynovitis. Pascoli et al. [13] evaluated 61 ankles in 42 patients with juvenile idiopathic arthritis and found poor agreement between clinical and ultrasound findings in compartmentalization of ankle inflammation by using a linear weighted kappa analysis. They found that clinical examination for the detection of joint and tendon involvement has a low positive predictive value (0.6 for tibiotalar joint and posterior tibialis tendon and 0.4 for peroneal tendon involvement) [13]. These prior studies and our results illustrate that imaging is necessary to accurately determine the extent of tendon sheath involvement in patients with juvenile idiopathic arthritis with active inflammation. Our results affirm that juvenile idiopathic arthritis is a spectrum disorder and that the current classification as set by the International League of ssociations for Rheumatology [14] may not correctly group certain subsets of patients, especially those with oligoarticular disease with primarily wrist or ankle involvement. Furthermore, future revisions of the juvenile idiopathic arthritis classification could include MRI confirmation of disease extent and include tendon involvement. MRI provides more objective, detailed, and reproducible assessment of disease status compared with the clinical examination alone in the ankle and wrist. This retrospective study has several limitations. First, there is referral bias in our numbers because we evaluated only patients who were referred for MRI. It is beyond the scope of this study to assess patients with mild or medically controlled disease, who usually do not undergo MRI. Second, there was heterogeneity of patients with juvenile idiopathic arthritis because oligoarticular, polyarticular, and systemic juvenile idiopathic arthritis subtypes were assessed together in our study. Larger studies with data obtained prospectively in patients with only oligoarticular or polyarticular arthritis would be helpful to further refine the utility of MRI in the subcategorization of patients with juvenile idiopathic arthritis. Third, the majority of patients began systemic therapy for juvenile idiopathic arthritis before MRI and, therefore, our data likely underreport the true incidence of multicompartmental joint and tendon sheath involvement. In summary, tenosynovitis and multicompartmental arthritis are common findings in the ankle and wrist in patients with juvenile idiopathic arthritis referred for MRI. The International League of ssociations for Rheumatology classification of juvenile idiopathic arthritis should also include image-based criteria for joint inflammation and the presence or absence of tenosynovitis. Wrist or ankle MRI may be helpful in patients before targeted steroid injections because they frequently show additional sites of involvement. History and physical examination findings alone may be insufficient for determining which specific joint compartments or tendons should undergo targeted steroid injections. References 1. Ravelli, Martini. Juvenile idiopathic arthritis. Lancet 2007; 369: Ruth NM, Passo MH. Juvenile idiopathic arthritis: management and therapeutic options. Ther dv Musculoskelet Dis 2012; 4: Marti P, Molinari L, olt I, Seger R, Saurenmann RK. Factors influencing the efficacy of intra-articular steroid injections in patients with juvenile idiopathic arthritis. Eur J Pediatr 2008; 167: Cleary G, Murphy HD, Davidson JE. Intra-articular corticosteroid injections in juvenile idiopathic arthritis. rch Dis Child 2003; 88: Damasio M, Malattia C, Tanturri de Horatio L, et al. MRI of the wrist in juvenile idiopathic arthritis: proposal of a pediatric synovitis score by a consensus of an international working group re- 416 JR:202, February 2014

5 Wrist and nkle MRI of Patients With Juvenile Idiopathic rthritis sults of a multicentre reliability study. Pediatr Radiol 2012; 42: Huppertz HI, Tschammler, Horwitz E, Schwab KO. Intraarticular corticosteroids for chronic arthritis in children: efficacy and effects on cartilage and growth. J Pediatr 1995; 127: Lamer S, Sebag GH. MRI and ultrasound in children with juvenile chronic arthritis. Eur J Radiol 2000; 33: Johnson K. Imaging of juvenile idiopathic arthritis. Pediatr Radiol 2006; 36: Hervé-Somma CM, Sebag GH, Prieur M, onnerot V, Lallemand DP. Juvenile rheumatoid arthritis of the knee: MR evaluation with Gd-DOT. Radiology 1992; 182: Kan JH, Graham T. Combined pre-injection wrist and ankle MRI protocol and steroid joint injections in juvenile idiopathic arthritis. Pediatr Radiol 2011; 41: Gardner-Medwin JM, Killeen OG, Ryder C, radshaw K, Johnson K. Magnetic resonance imaging identifies features in clinically unaffected knees predicting extension of arthritis in children with monoarthritis. J Rheumatol 2006; 33: Rooney ME, Mcllister C, urns JF. nkle dis- FOR YOUR INFORMTION Mark your calendar for the following RRS annual meetings: May 4 9, 2014 Manchester Grand Hyatt San Diego, San Diego, C pril 19 24, 2015 Toronto Convention Centre, Toronto, ON, Canada pril 17 22, 2016 Los ngeles Convention Center, Los ngeles, C ease in juvenile idiopathic arthritis: ultrasound findings in clinically swollen ankles. J Rheumatol 2009; 36: Pascoli L, Wright S, Mcllister C, Rooney M. Prospective evaluation of clinical and ultrasound findings in ankle disease in juvenile idiopathic arthritis: importance of ankle ultrasound. J Rheumatol 2010; 37: Petty RE, Taunton RS, Manners P, et al. International League of ssociations for Rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, J Rheumatol 2004; 31: JR:202, February