MRI predicts 5-year joint bleeding and development of arthropathy on radiographs in hemophilia

Abstract

Magnetic resonance imaging (MRI) is considered as the reference standard to assess early joint changes in hemophilia. However, the clinical relevance of MRI findings is still unknown. The aim of this prospective study was to assess the predictive value of MRI for 5-year joint bleeding and progression of arthropathy in patients with hemophilia. Both knees and ankles of patients with hemophilia and absent or limited arthropathy on radiographs were assessed by using MRI and radiographs. MRI scans were scored according to the International Prophylaxis Study Group MRI score for hemophilic arthropathy. Patients were followed up for 5 years, including assessment of joint bleeding and repeated radiographic assessment. Associations between baseline MRI findings with 5-year bleeding and progression of arthropathy were expressed as odds ratios (OR), adjusted for severity of disease and joint bleeding history. Baseline assessment included 104 joints of 26 patients with hemophilia (median age, 21 years). Four ankles with severe joint changes were excluded. Follow-up was available for 96 (92%) of 104 joints. During 5 years of follow-up, bleeding was reported for 36% of joints. Five-year bleeding was significantly increased in joints with synovial hypertrophy at 80% vs 27% in joints without synovial hypertrophy (OR, 10.1; 95% confidence interval, 3.4-31.3). In joints with normal baseline radiographs, any osteochondral or synovial changes on MRI were associated with radiographic changes 5 years later (positive predictive value, 75%; negative predictive value, 98%). Joints with synovial hypertrophy on MRI had a significantly higher chance of 5-year bleeding. All MRI changes, except effusion, were strong predictors for development of arthropathy on radiographs.