Objectives: The provisional ACR/European League Against Rheumatism (EULAR) definition of remission in RA requires a score of ≤1 on the patient global assessment (PGA, 0-10 scale). We explored the relation between the PGA criterion and the patient's clinical disease state in an observational dataset.
Methods: Data of 512 newly diagnosed RA patients of the Dutch Rheumatoid Arthritis Monitoring (DREAM) remission induction cohort were analysed. Both 28-joint counts and more comprehensive joint counts (tender joint count-53, swollen joint count-44) were used.
Results: ACR/EULAR remission was present in 20.1% of the patients when using 28-joint counts and in 17.4% of the patients when applying more comprehensive joint counts. In 108 patients, the PGA score was >1 despite fulfilment of the remaining criteria (TJC28, SJC28 and CRP in mg/dl ≤1). Residual disease activity was observed in 31.5% (34/108) and median (interquartile range) scores on PGA, pain and fatigue were 2.4 (1.8-4.0), 2.0 (1.1-3.0) and 2.7 (1.3-5.0), respectively. Applying more comprehensive joint counts showed comparable results. In 19.5% (100/512) of patients, disease activity was absent (TJC53 = 0, SJC44 = 0, and CRP ≤1). In 41% (n = 41) of these patients, the PGA score was >1. Receiver operating characteristic analysis showed moderate accuracy of the PGA to discriminate between fulfilment and no fulfilment of all remaining criteria.
Conclusion: Frequently, patients did not meet the PGA criterion despite a good clinical disease state. Apparently the PGA is not solely influenced by RA disease activity. In patients with marked divergence between the PGA and objective clinical measurements, caution should be taken when applying the provisional ACR/EULAR definition of remission.