OBJECTIVE: Palindromic rheumatism is characterized by attacks of acute arthritis of short duration. In the long term, a substantial proportion of patients will develop rheumatoid arthritis (RA) or other connective tissue diseases, but the determinants of subsequent chronic disease have not been adequately established. We identify clinical prognostic factors for the development of RA and other connective tissue diseases in patients with palindromic rheumatism in a retrospective cohort study. METHODS: The medical records of 4900 patients with arthritis referred from 1986 to 1996 to 3 rheumatologists at an academic center were reviewed. One hundred sixty patients were diagnosed as having palindromic rheumatism. After review, 127 complied with diagnostic criteria for palindromic rheumatism. Disease duration was estimated as time of first attack until the last consultation, or the development of RA or other connective tissue disease. Survival analysis including Cox regression was used to identify clinical variables associated with the risk of developing RA or other connective tissue disease, adjusting for varying disease duration. RESULTS: Sixty-five percent of the patients were female. Age at onset was 40+/-12 years. Mean disease duration was 6+/-6 years, and mean followup by the rheumatologists was 40+/-45 months. Joints more frequently affected were wrist, knee, and metacarpophalangeal. Forty-three patients (34%) subsequently developed a connective tissue disease including 36 (28%) RA, 3 (2%) systemic lupus erythematosus, and 4 (3%) other connective tissue diseases. In the final Cox regression model the hazard ratio for development of a connective tissue disease in the presence of a positive rheumatoid factor (RF) was 2.9 (p = 0.002), for proximal interphalangeal (PIP) joint involvement 2.4 (p = 0.02), for wrist involvement 2.5 (p = 0.05), for female sex 2.2 (p = 0.05), and for age at onset 1.03 (per year) (p = 0.001). Female patients with positive RF and involvement of the hands had an 8-fold risk of developing disease, compared with patients with one or fewer of these features. CONCLUSION: Positive RF and early involvement of the wrist and PIP joints predict the subsequent development of RA or other connective tissue disease in patients with palindromic rheumatism, and identify a group of patients at increased risk.
OBJECTIVE:Palindromic rheumatism is characterized by attacks of acute arthritis of short duration. In the long term, a substantial proportion of patients will develop rheumatoid arthritis (RA) or other connective tissue diseases, but the determinants of subsequent chronic disease have not been adequately established. We identify clinical prognostic factors for the development of RA and other connective tissue diseases in patients with palindromic rheumatism in a retrospective cohort study. METHODS: The medical records of 4900 patients with arthritis referred from 1986 to 1996 to 3 rheumatologists at an academic center were reviewed. One hundred sixty patients were diagnosed as having palindromic rheumatism. After review, 127 complied with diagnostic criteria for palindromic rheumatism. Disease duration was estimated as time of first attack until the last consultation, or the development of RA or other connective tissue disease. Survival analysis including Cox regression was used to identify clinical variables associated with the risk of developing RA or other connective tissue disease, adjusting for varying disease duration. RESULTS: Sixty-five percent of the patients were female. Age at onset was 40+/-12 years. Mean disease duration was 6+/-6 years, and mean followup by the rheumatologists was 40+/-45 months. Joints more frequently affected were wrist, knee, and metacarpophalangeal. Forty-three patients (34%) subsequently developed a connective tissue disease including 36 (28%) RA, 3 (2%) systemic lupus erythematosus, and 4 (3%) other connective tissue diseases. In the final Cox regression model the hazard ratio for development of a connective tissue disease in the presence of a positive rheumatoid factor (RF) was 2.9 (p = 0.002), for proximal interphalangeal (PIP) joint involvement 2.4 (p = 0.02), for wrist involvement 2.5 (p = 0.05), for female sex 2.2 (p = 0.05), and for age at onset 1.03 (per year) (p = 0.001). Female patients with positive RF and involvement of the hands had an 8-fold risk of developing disease, compared with patients with one or fewer of these features. CONCLUSION: Positive RF and early involvement of the wrist and PIP joints predict the subsequent development of RA or other connective tissue disease in patients with palindromic rheumatism, and identify a group of patients at increased risk.