Anti-Proteus antibodies and Proteus organisms in rheumatoid arthritis: a clinical study.

We have studied anti-Proteus antibodies (APA), isolation of Proteus, and their relation to various measures of RA disease activity. Seventy RA patients with a CRP > 10 mg/l had higher APA titres than 17 RA patients with CRP < or = 10 mg/l (P = 0.006), and 36 non-RA controls (P = 0.003). However, in a cross-sectional study of the RA group, there was no correlation between APA and a number of clinical and laboratory measures of disease activity, including the CRP and Stoke RA activity index. A longitudinal study showed no correlation between changes in these measures of disease activity and change in APA titre. We were unable to isolate Proteus in the urine or faeces of RA patients more frequently than controls, and the isolation of Proteus did not correlate with serum APA titres. Urinary APA was present in equal frequencies in RA and non-RA patients. NSAIDs, DMARDs and steroids did not appear to influence APA titres in the RA group. These results suggest that APA may act as an acute phase protein, distinct from CRP, but not correlating with RA disease activity in its broadest context. The fact that the antibody we are measuring binds to Proteus may be irrelevant, and the study does not support a role for Proteus in RA.