OBJECTIVE: To demonstrate directly that highly reactive hydroxyl radicals (OH.) can be generated in patients with rheumatoid arthritis and contribute to joint damage, and to examine the ability of blood to cause OH. generation. METHODS: The sensitive and specific technique of hydroxylation of aromatic compounds (salicylate and phenylalanine) was used to measure OH.. Synovial fluid and blood from patients with active rheumatoid arthritis were aspirated and immediately added to tubes containing salicylate and phenylalanine as detectors of OH., or to tubes containing saline as a control. Levels of specific products of attack of OH. upon salicylate (2,3- and 2,5-dihydroxybenzoates) and phenylalanine (ortho- and meta-tyrosines) were measured by high performance liquid chromatography. RESULTS: Synovial fluid samples aspirated into saline never contained ortho- or meta-tyrosines or 2,3-dihydroxybenzoate. Of 53 patients examined, synovial fluid and blood from 36 caused formation of ortho- and meta-tyrosines when aspirated into solutions containing phenylalanine. Repeated sampling from three "positive" patients showed consistent evidence of these hydroxylation products. Similarly, of 22 patients examined, synovial fluid and blood from 18 caused formation of 2,3- and 2,5-dihydroxybenzoates when aspirated into salicylate solutions. Further evidence for the role of OH. was provided by inhibition of the hydroxylation by the specific OH. scavengers mannitol and sodium formate. CONCLUSIONS: Aspirated knee joint fluids and blood from rheumatoid arthritis patients can generate OH., consistent with current views on the importance of this radical as a cytotoxic agent in rheumatoid disease. The ability of body fluids to cause OH. formation is not correlated with simple laboratory indices of disease activity, but is reproducible on sequential sampling from the same patients. The mechanism and significance of the phenomenon in rheumatoid arthritis pathology remain to be established.
