Comparison of serum and synovial fluid concentrations of beta 2-microglobulin and C reactive protein in relation to clinical disease activity and synovial inflammation in rheumatoid arthritis.

beta 2-Microglobulin and C reactive protein (CPR) were measured in 33 and 57 matched pairs of serum and synovial fluid (SF) respectively, from patients with active rheumatoid arthritis (RA). Serum beta 2-microglobulin concentrations were higher than in normal controls and the SF concentration was higher than the serum concentration on 25 of 33 occasions (76%), suggesting a local production of beta 2-microglobulin within the synovial membrane. There was a correlation between serum and SF concentrations of beta 2-microglobulin (r = 0.50). Patients' serum CRP concentrations in 57 samples were higher than in normal controls and were greater than in the matched SFs on 49 of the 57 paired samples (86%). In 18 samples CRP was absent in the SF, suggesting a local consumption or binding within the synovial membrane. Twenty four patients with RA given either sodium aurothiomalate or D-penicillamine for six months showed highly significant clinical improvements accompanied by reductions in serum and SF immunoglobulin concentrations and knee joint suprapatellar pouch synovial membrane T lymphocyte infiltrates. In this group of patients serum CRP, but not beta 2-microglobulin, fell significantly, but there were no significant changes in SF beta 2-microglobulin or CRP. These data suggest that serum and SF beta 2-microglobulin concentrations are not a useful index for determining the therapeutic response to sodium aurothiomalate and D-penicillamine and that serum rather than SF CRP concentrations are more helpful. The persistent raised serum and SF concentrations of beta 2-microglobulin probably reflect synovial inflammatory infiltrates, which are still considerable despite apparent clinical remission.