BackgroundIn the present study, we explored the risk factors for relapse after discontinuation of biologic disease-modifying antirheumatic drug (bDMARD) therapy in patients with rheumatoid arthritis (RA) whose ultrasound power Doppler (PD) synovitis activity and clinical disease activity were well controlled.MethodsIn this observational study in clinical practice, the inclusion criteria were based on ultrasound disease activity and clinical disease activity, set as low or remission (Disease Activity Score in 28 joints based on erythrocyte sedimentation rate <3.2). Ultrasound was performed in 22 joints of bilateral hands at discontinuation for evaluating synovitis severity and presence of bone erosion. Patients with a maximum PD score ≤1 in each joint were enrolled. Forty patients with RA were consecutively recruited (November 2010–March 2015) and discontinued bDMARD therapy. Variables at the initiation and discontinuation of bDMARD therapy that were predictive of relapse during the 12 months after discontinuation were assessed.ResultsThe median patient age was 54.5 years, and the median disease duration was 3.5 years. Nineteen (47.5%) patients relapsed during the 12 months after the discontinuation of bDMARD therapy. Logistic regression analysis revealed that only the presence of bone erosion detected by ultrasound at discontinuation was predictive of relapse (OR 8.35, 95% CI 1.78–53.2, p = 0.006). No clinical characteristics or serologic biomarkers were significantly different between the relapse and nonrelapse patients. The ultrasound synovitis scores did not differ significantly between the groups.ConclusionsOur findings are the first evidence that ultrasound bone erosion may be a relapse risk factor after the discontinuation of bDMARD therapy in patients with RA whose PD synovitis activity and clinical disease activity are well controlled.