PRACTICAL ASPECTS OF CYTOKINE-DETERMINATI ON IN THE SYNOVIAL-FLUID OFPATIENTS WITH OSTEOARTHRITIS, RHEUMATOID-ARTHRITIS AND OTHER JOINT-DISEASES

Objective To determine whether the activity of cartilage-degrading enz ymes in the synovial fluid (SF) of patients with rheumatoid arthritis and other joint diseases is correlated with the concentration of cytok ines in the SF. Methods Cytokines and cartilage-degrading enzymes were determined in the SF of 97 patients with various disorders involving the knee joints (rheumatoid arthritis (RA) n 44; osteoarthritis (OA) n 35; meniscal trauma (Men) n 10; reactive arthritides (ReA) n 8). In t hese samples we measured the concentrations of interleukin-1 alpha and beta, IL-1-receptor antagonist (IL-1ra), IL-6, IL-8, tumor necrosis f actor alpha (TNF alpha; all by ELISA), collagenase-activity and casein ase-activity (by substrate assays). Results With the exception of IL-1 alpha and IL-6, cytokine-concentrations were significantly higher in RA than in OA SF-samples (p < 0.05; ANOVA on ranks). IL-1ra, IL-6, and IL-1 beta were correlated best with the collagenase-activity in the S F (r = 0.63; 0.57; 0.55; Spearman's rank correlation), while IL-1 beta (r = 0.53) and IL-1ra (r = 0.52) were best correlated with the casein ase-activity in the samples. The SF-concentration of IL-1ra was well c orrelated with the levels of IL-6, IL-1 beta, II-8, and TNF alpha (r f rom 0.73 to 0.66; all p < 0.005), but not with IL1 alpha. The molar ra tio of IL-1 to IL-1ra in the SF was neither correlated with the activi ty of collagenase nor caseinase. IL-1 beta and IL-1ra in the SF were p ositively correlated with the erythrocyte sedimentation rate (ESR). Co nclusions The determination of IL-1 beta and IL-1ra in the SF of patie nts with joint disorders as examined in this study seems to allow to a certain extent a prediction of the collagenase- and caseinase-activit y contained in the diseased joint. We would favor the measurement of I L-1ra, since it was usually found in a more than 100-fold molar excess over IL-1 beta. The molar ratio of IL-1 to IL-1ra had no predictive v alue for the activity of cartilage-degrading enzymes in the sample, or for the patient's ESR. The hypothesis that in rheumatic diseases the IL-1/IL-1ra - ratio in the SF may reflect the inflammatory activity of the respective joint, is not supported by our data.