Contrast-enhanced magnetic resonance imaging (MRI) is considered the diagnostic standard for identifying involvement of the temporomandibular joint by juvenile idiopathic arthritis. Early or active arthritis is shown as bone marrow oedema, joint effusion, synovial thickening and increased joint enhancement. Subsequent joint damage includes characteristic deformity of the mandibular condyle, bone erosion, disk abnormalities and short mandibular ramus due to impaired growth. In this pictorial essay, we illustrate normal MRI findings and growth-related changes of the temporomandibular joint in children. The rationale and practical application of semiquantitative MRI assessment of joint inflammation and damage are discussed and presented. This atlas can serve as a reference for grading temporomandibular joint arthritis according to the scoring systems proposed by working groups of OMERACT (Outcome Measures in Rheumatology and Clinical Trials) and the EuroTMjoint research network. Systematic assessment of the level of inflammation, degree of osteochondral deformation, and growth of the mandibular ramus by MRI may aid in monitoring the course of temporomandibular joint arthritis and evaluating treatment options.Electronic supplementary materialThe online version of this article (10.1007/s00247-017-4000-0) contains supplementary material, which is available to authorized users.

A novel TMJ MRI scoring system was developed by consensus. Further iterative refinements and reliability testing are warranted in upcoming studies.

Objective The purpose of this study was 1) to assess the inter-reader reliability in detecting and scoring the inflammatory bone lesions in pediatric patients with chronic nonbacterial osteomyelitis (CNO) by using WB-MRI, and 2) to evaluate the responsiveness of the MRIdetected CNO lesions to pamidronate therapy. Methods Eighty-eight WB-MRI examinations were independently reviewed and scored by two radiologists blinded to clinical details in 32 retrospectively enrolled pediatric patients with CNO. Inflammatory bone lesions, soft tissue abnormality, and bony structural changes were scored before and after pamidronate therapy. Lesion responsiveness was calculated by using standardized response mean and inter-reader reliability was assessed by kappa statistics. Results There was good to excellent inter-reader agreement for the detection and quantification of bone lesions. After the first cycle of pamidronate in all 32 patients, 96 of the 279 lesions (34%) (after excluding 108 lesions of hand and feet) resolved, while in a subset of 11 patients with two or more cycles, 76% of lesions resolved after the second cycle. Twenty-one (7.5%) lesions worsened and 46 (16.4%) new lesions developed after one cycle in all 32 patients. In these 11 patients, the number of worsened lesions reduced to 2 (2%) and new lesions to 14 (14.9%) after the second cycle as detected on MRI. Vertebral lesions had the highest response to treatment. Conclusion WB-MRI is a reliable tool for objective quantification and assessment of response to treatment of pediatric CNO bone lesions and could be used to monitor disease activity for clinical and research purposes.

No abstract