Associated Factors of Erosive Arthritis in Rheumatoid Arthritis and Other Connective Tissue Diseases: A Retrospective Study from Southern Chinese Population

Abstract

Background Erosive arthritis (EA) was common in Rheumatoid Arthritis (RA) and connective tissue diseases (CTDs), which indicated poor outcomes. It would be clinically useful to identify a subset of patients at an early stage of the disease process who have a high risk of developing EA. Most of the previous studies focused on EA in RA, less about other CTDs. T

Objectives To investigate the putative predictors of EA in patients with RA and other CTDs, thus improve appropriate medical intervention and a better prognosis.

Methods A retrospective review of the medical records between 2010 and 2013 was performed at the First Affiliated Hospital of Sun Yat-Sen University. Patients' baseline information including demographic variables, clinical features, radiological characters and serological parameters was collected.

Results One thousand and sixty seven patients [338 with RA, 408 with systemic lupus erythematosus (SLE), 69 with primary Sjögren's syndrome (pSS), 33 with systemic sclerosis (SSc), 35 with adult onset Still's Disease (AOSD), 94 with vasculitis, 59 with undifferentiated connective tissue disease (UCTD), 8 with multiple connective tissue disease (MCTD), 19 with inflammatory myopathy, 2 with polymyalgia rheumatica and 2 with erythema nodosum] were selected. EA was noted in 60.4% (204/338) of patients with RA, 1.5% (6/408) of patients with SLE, 5.8% (4/69) of patients with pSS, and 9.1% (3/33) of patients with UA. The multivariate logistic regression analysis indicated that rheumatoid nodules (OR=3.04, 95%CI 1.26–7.34, P=0.01), anemia (OR=3.31, 95%CI1.21–9.02, P=0.02), and positive anti-cyclic citrullinated peptide antibodies (ACPA) (OR=6.49, 95%CI 3.34–12.63, P<0.01) were strongly associated factors for the occurrence of EA in RA patients. When compared with patients without EA, high level and prominently higher positive rate of ACPA was found in SLE and pSS. Statistically significant association was found between ACPA (P=0.03) as well as IgM-rheumatoid factors (RF) positivity (P=0.01) and the presence of EA in SSc.

Conclusions EA occurred early in RA and indicated poor outcomes. Anemia, rheumatoid nodules and ACPA were associated with EA development in RA. SLE overlapping RA, pSS progressing to RA-associated secondary disease and UA evolving to RA also could induce EA. Anti-CCP antibodies also could be detected in these diseases and was closely associated with EA. This may be a useful biomarker in identifying a subset of CTDs patients with high risk for development of EA.