Outcome of patients with early arthritis without rheumatoid factor and ACPA and predictors of rheumatoid arthritis in the ESPOIR cohort

Abstract

Objective: To describe the disease course of patients with early arthritis without rheumatoid factor (RF) and anti-citrullinated protein auto-antibodies (ACPA) in an inception cohort. To determine baseline predictors of fulfilling 2010 ACR/EULAR criteria for rheumatoid arthritis (RA) for these patients within 3 years. Method: Patients included in the multicenter ESPOIR cohort were compared at baseline and 3 years by whether they were negative for RF and ACPA ("seronegative") or positive for RF and/or ACPA ("seropositive"). Univariate analysis was used to determine the association between baseline variables in seronegative patients and RA classification. Stepwise multiple logistic regression was used to identify predictors of RA classification within 3 years, estimating odds ratios (ORs). Results: Among 354 seronegative patients, 224/340 with available data (65.9%) fulfilled RA classification at baseline and 189/233 (81.1%) at 3 years. As compared with seropositive patients, seronegative patients had lower DAS28 (p = 0.002) and lower modified total Sharp score (mTSS; p = 0.026) at baseline; DAS28 remission was similar (p = 0.634), but radiographic progression rate was lower in seronegative patients (p < 0.001) at 3 years. In seronegative patients, factors predicting RA classification within 3 years were additive (OR = 3.61), bilateral (OR = 2.59) and hand, wrist or forefeet involvement (OR = 3.87); presence of a trigger event (OR = 3.57); pain at rest (OR = 2.76); morning stiffness (OR = 2.62); number of tender joints (OR = 23.73); and mTSS (OR = 2.56). Conclusion: "Seronegative" patients have less active disease at baseline and less radiographic progression during follow-up than "seropositive" patients. With inflammatory pain, symmetric involvement of numerous small joints and erosive disease, a classification of RA is likely.