Background Up to 20–30% of patients enrolled into RA cohorts and clinical trials are seronegative. However, in studies examining predictors, prognosis, and response to treatment, seropositive and seronegative groups of patients seem to behave differently. Only a few studies have focused on the long term follow up of seronegative arthritis.
Objectives To investigate long-term outcomes of patients with seronegative arthritis during a 10-year follow-up period including clinical outcomes and reclassification of diagnosis when applicable.
Methods A total of 1046 patients were classified as early RA in 1997–2005 at single rheumatology center and scheduled for a ten year follow-up, including 434 seronegative patients who are subjects of the present analysis. Follow-up examinations were carried out for at least 2 years and then at 5 and 10 years by the treating specialist including complete clinical examination and patient reported outcomes. In addition, case –reviews were performed, with re-classification of the cases when applicable
Results Among the 434 seronegative patients (69,4% women, mean age 58), 271 subjects were seen for the 10 year visit with the mean disease activity DAS28 of 2,3 (SD 1,02) and the mean HAQ of 0,71 (SD 0,72). Out of the remaining 146 patients, 88 had died and 53 did not attend the 10-year visit due to altered diagnosis, refusal or comorbidity. Five patients had dropped out and files of 17 patients were missing. During the follow-up of 10 years, 12/434 (2,7%) patients could be classified as seropositive or erosive RA: 4 turned seropositive (2 for ACPA and 2 for RF [>2x normal level]) and 8 developed erosions typical for RA. Reclassification revealed 70 (16,1%) cases of polymyalgia rheumatica, 47 (10,8%) cases of osteoarthrosis without evidence of inflammatory disease, 47 (10,8%) cases of psoriatic arthritis, 39 (9,0%) cases of spondylarthritis and 16 (3,7%) cases of plausible reactive arthritis. Few cases were reclassified as gout (11 cases (2,5%)) and pseudogout (3 cases (0,7%)). Also paraneoplastic arthritis (6 cases (1,4%)), juvenile arthritis (5 cases (1,2%)), hemochromatosis (2 cases (0,5%)), ankylosing spondylitis (2 cases (0,5%)) and temporal arteritis (2 cases (0,5%)) were revealed during follow up. One case of each reflex sympathetic dystrophy, trauma-induced arthritis, meniscal injury, optional Nasu Hakola disease, microscopic polyangiitis (MPA), granulomatous polyangiitis (GPA), antisynthetase syndrome and colitis ulcerosa were also found. The remaining 147 patient (33,8%) could not be reclassified in any clear cut diagnosis. A total of 44 of these undifferentiated cases had transient arthritis, 43 cases had features of seronegative spondylarthritis and 57 cases remained totally unspecified, while three patients had features of inflammatory connective tissue disease (SLE and Sjögren's syndrome), but they did not meet available classification criteria. In addition, files of 17 (3,9%) patients were missing from the analyses.
Conclusions Over a 10-year period, 97% of seronegative patients remained seronegative and did not develop RA-like erosions. Reclassification revealed significant heterogeneity in the diagnosis of seronegative RA. Therefore, seronegative arthritis should not be studied as a homogenous disease entity.
Padyukov, Arthritis Rheum 2004, 50: 3085–3092.
van Dongen, Arthritis Rheum 2007, 56: 1424–1432.
Jantti, Clin Rheumatol 2002, 21: 353–356.
Disclosure of Interest None declared