The advent of TNF antagonists has had a revolutionary impact on the treatment of ankylosing spondylitis (AS). However, whether the benefits recorded in controlled trials are replicated in everyday practice has not been extensively evaluated.

To evaluate the effectiveness of infliximab in patients with axial spondyloarthropathies, to identify factors associated with the treatment response, and to assess fulfillment of modified New York criteria and compliance with guidelines about using TNF antagonists.

Retrospective review of patients given infliximab for axial spondyloarthropathy between 2001 and 2003. Disease activity, motion limitation, laboratory tests, and the 6-week response rate were recorded.

Of the 86 included patients (48 women and 38 men, mean age, 44±11years), 37% responded to infliximab therapy. Uveitis and C-reactive protein elevation at baseline predicted a response. Only 53% of patients met modified New York criteria and only 23% met ASAS criteria for starting TNF antagonist therapy.

Infliximab was less effective in our patients with axial spondyloarthropathy than expected based on the results of controlled trials. However, many patients did not meet New York criteria for AS and/or ASAS criteria for TNF antagonist therapy. Therefore, our results do not challenge the usefulness of TNF antagonists in axial AS. Our patients were treated before the development of guidelines for TNF antagonist therapy and before the introduction of magnetic resonance imaging as an evaluation tool in AS. Strict criteria should be used to decide when TNF antagonist therapy is appropriate in patients with spondyloarthropathies.