Objective: The classification of rheumatoid arthritis (RA) is increasingly important as new therapies can halt the disease in its early stages. Antibodies to cyclic citrullinated peptides (anti-CCP) are widely used for RA diagnosis, but are not in the 1987 American College of Rheumatology (ACR) Criteria for RA Classification. We developed and tested the performance characteristics of new criteria for RA classification, incorporating anti-CCP.
Methods: We identified all subjects seen in our arthritis centre with rheumatoid factor (RF) and anti-CCP tested simultaneously between 1 January and 30 June 2004 and reviewed their medical records for the ACR criteria, rheumatologists' diagnoses, RF and anti-CCP. We revised the ACR criteria in two ways: (a) adding anti-CCP, and (b) replacing rheumatoid nodules and erosions with anti-CCP (CCP 6 criteria). We compared sensitivity and specificity of all criteria, in all subjects and in subjects with arthritis symptoms </=6 months.
Results: Medical records of 292 subjects were analysed: mean age was 54 years, 82% were women, and mean symptom duration was 4.1 years. 17% were RF positive and 14% were anti-CCP positive at initial testing. 78 (27%) had definite RA per treating rheumatologist at latest follow-up. The CCP 6 criteria increased sensitivity for RA classification for all subjects regardless of symptom duration: 74% vs 51% for ACR criteria with a loss in specificity (81% vs 91%). Sensitivity was greatly improved in subjects with symptoms < or =6 months: 25% vs 63% for ACR criteria with a decrease in specificity.
Conclusions: The CCP 6 criteria improved upon the sensitivity of the ACR criteria, most remarkably for subjects with symptoms < or =6 months and could be used for the classification of subjects for RA in clinical studies.