Classification criteria of early rheumatoid arthritis and validation of its performance in a multi-centre cohort

Abstract

Objectives: To develop classification criteria for early rheumatoid arthritis (ERA) based on a large cohort of early inflammatory arthritis patients and to evaluate the performance of these criteria.

Methods: The study population comprised a cohort of early inflammatory arthritis patients with symptom duration less than one year. Classification criteria of ERA were developed by incorporating the most sensitive or specific variables. Performance of the ERA criteria, 1987 ACR and 2010 ACR/EULAR criteria were evaluated.

Results: A total of 803 patients were enrolled in this study. By the end of the one year follow-up, 514 patients were diagnosed with RA, 251 with other rheumatic diseases, and 38 patients with undifferentiated arthritis. The ERA criteria are as follows: 1) morning stiffness ≥30 minutes; 2) arthritis of 3 or more joint areas; 3) arthritis of hand joints; 4) positive RF; 5) positive anti-CCP antibody. Rheumatoid arthritis is defined by the presence of 3 or more of the criteria. The sensitivity (84.4%) of the ERA classification criteria was much higher than the 1987 ACR criteria (58.0%). In a validation cohort of early inflammatory arthritis patients, the area under the ROC curves (AUC) showed a better performance for the ERA criteria (0.906, 95%CI 0.866 to 0.945) than the 1987 ACR criteria (0.786, 95%CI 0.725 to 0.848) and the 2010 ACR/EULAR criteria (0.745, 95%CI 0.677 to 0.814).

Conclusions: A set of ERA classification criteria has been developed with good performance for early RA. It is applicable in clinical practice and research.