OBJECTIVE: Our objective was (1) to determine if serum creatine kinase (CK) activity is reduced in rheumatoid arthritis (RA) compared with that of noninflammatory rheumatic diseases, (2) to examine the recently described association of low CK activity and disease variables in our RA population, and (3) to examine the influence of steroid on serum CK activity in patients with RA. METHODS: Cross sectional and longitudinal retrospective analyses of clinical and biochemical data of consecutive patients with RA and noninflammatory arthropathies. In all subjects we evaulated age, sex, weight, and, only for patients with RA, history of use of corticosteroids and Ritchie index. C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), hemoglobin, and platelet count were simultaneously determined as variables of disease activity. CK activity was determined by automated biochemical analyzer (Hitachi 747, Japan). RESULTS: Serum CK activity was significantly reduced in RA (mean+SD: 45.7 +24.2 IU/L) compared to controls (81.3+33.9 IU/L) (p < 0.001). Ritchie index, CRP, and platelet count correlated inversely with CK values (correlation coefficient: 0.31, p < 0.01; 0. 45, p < 0.001; 0.42, p < 0.001, respectively). Patients taking steroids had lower CK activity than those without steroid, but not statistically significant.