Synovitis (inflamed joint synovial lining) in rheumatoid arthritis (RA) can be assessed by clinical examination (CE) or ultrasound (US). To investigate the added value of US, compared with CE alone, in RA synovitis in terms of clinical effectiveness and cost-effectiveness. Electronic databases including MEDLINE, EMBASE and the Cochrane databases were searched from inception to October 2015. A systematic review sought RA studies that compared additional US with CE. Heterogeneity of the studies with regard to interventions, comparators and outcomes precluded meta-analyses. Systematic searches for studies of cost-effectiveness and US and treatment-tapering studies (not necessarily including US) were undertaken. A model was constructed that estimated, for patients in whom drug tapering was considered, the reduction in costs of disease-modifying anti-rheumatic drugs (DMARDs) and serious infections at which the addition of US had a cost per quality-adjusted life-year (QALY) gained of £20,000 and £30,000. Furthermore, the reduction in the costs of DMARDs at which US becomes cost neutral was also estimated. For patients in whom dose escalation was being considered, the reduction in number of patients escalating treatment and in serious infections at which the addition of US had a cost per QALY gained of £20,000 and £30,000 was estimated. The reduction in number of patients escalating treatment for US to become cost neutral was also estimated. Fifty-eight studies were included. Two randomised controlled trials compared adding US to a Disease Activity Score (DAS)-based treat-to-target strategy for early RA patients. The addition of power Doppler ultrasound (PDUS) to a Disease Activity Score 28 joints-based treat-to-target strategy in the Targeting Synovitis in Early Rheumatoid Arthritis (TaSER) trial resulted in no significant between-group difference for change in Disease Activity Score 44 joints (DAS44). This study found that significantly more patients in the PDUS group attained DAS44 remission (p = 0.03). The Aiming for Remission in Rheumatoid Arthritis (ARCTIC) trial found that the addition of PDUS and grey-scale ultrasound (GSUS) to a DAS-based strategy did not produce a significant between-group difference in the primary end point: composite DAS of < 1.6, no swollen joints and no progression in van der Heijde-modified total Sharp score (vdHSS). The ARCTIC trial did find that the erosion score of the vdHS had a significant advantage for the US group (p = 0.04). In the TaSER trial there was no significant group difference for erosion. Other studies suggested that PDUS was significantly associated with radiographic progression and that US had added value for wrist and hand joints rather than foot and ankle joints. Heterogeneity between trials made conclusions uncertain. No studies were identified that reported the cost-effectiveness of US in monitoring synovitis. The model estimated that an average reduction of 2.5% in the costs of biological DMARDs would be sufficient to offset the costs of 3-monthly US. The money could not be recouped if oral methotrexate was the only drug used. Heterogeneity of the trials precluded meta-analysis. Therefore, no summary estimates of effect were available. Additional costs and health-related quality of life decrements, relating to a flare following tapering or disease progression, have not been included. The feasibility of increased US monitoring has not been assessed. Limited evidence suggests that US monitoring of synovitis could provide a cost-effective approach to selecting RA patients for treatment tapering or escalation avoidance. Considerable uncertainty exists for all conclusions. Future research priorities include evaluating US monitoring of RA synovitis in longitudinal clinical studies. This study is registered as PROSPERO CRD42015017216. The National Institute for Health Research Health Technology Assessment programme.