The aim of this study was to evaluate whether the serum concentration of interleukin-6 (IL-6) reflects disease activity in ankylosing spondylitis (AS). A group of 271 AS patients were enrolled in the study, 261 of whom completed the entire protocol (201 males, 60 females, median age of 53 years). Serum IL-6 was measured three times (I, baseline; II, after 10-12 days; III, after 17-24 days) during a 3- or 4-week treatment at the health resort. At the same times, the variables for mobility were measured, and the patients were asked to assess their complaints (score) in a self-styled questionnaire. The serum concentration of IL-6 correlated with the measurements of occiput-to-wall distance, cervical rotation, finger-floor distance and Schober sign, and with morning pain at all three evaluations. Comparisons between changes in IL-6 and changes in the variables (measures of mobility, scores of the questionnaires) did not reveal significant correlations. Present data would suggest that in AS the serum concentration of IL-6 indicates the degree of mobility restriction resulting from previous disease progression, but is not a reliable marker of current disease activity.