Background MRI sensitively detects inflammation, but the clinical relevance of MRI-detected inflammation is undetermined in early arthritis. Therefore, this cross-sectional study investigated the association between MRI-detected inflammation of hands and feet and functional disability in early arthritis.
Methods 514 early arthritis patients, consecutively included in the Leiden Early Arthritis Clinic, were studied. At baseline, unilateral 1.5T MRI of the wrist, metacarpophalangeal (MCP) and metatarsophalangeal-joints was made and functional disability was measured using the Health Assessment Questionnaire (HAQ). MRIs were scored for bone marrow oedema (BME), tenosynovitis and synovitis by two readers. The sum of these three features yielded the total MRI-inflammation score. Linear and non-linear regression analyses were performed with HAQ as dependent variable.
Results The total MRI-inflammation score was associated with the HAQ-score (β=0.014, p<0.001), as were BME (β=0.015, p<0.001), tenosynovitis (β=0.046, p<0.001) and synovitis (β=0.039, p<0.001) separately. Analysing these three MRI-features in one multivariable model, revealed that only tenosynovitis was independently associated with HAQ-score (β=0.039, p<0.001). Also when correcting for age, gender, joint counts, acute phase reactants and auto-antibodies this association remained (β=0.034, p<0.001). MRI-detected inflammation at wrists or MCP-joints associated significantly with impairments in hand functioning (e.g. difficulties with opening milk cartons or jars). Exploring the relation between MRI-detected inflammation and HAQ-scores non-linear, suggested the presence of a ceiling effect, because after a certain inflammation-level, more inflammation was not associated with higher HAQ-scores.
Conclusions MRI-detected inflammation, and tenosynovitis in particular, is associated with functional disability. This demonstrates the functional relevance of MRI-detected inflammation in early arthritis patients.
Acknowledgement This work was supported by a Vidi-grant of the Netherlands Organisation for Scientific Research and the Dutch Arthritis Foundation
Disclosure of Interest None declared