The preliminary core set for endpoints in disease controlling antirheumatic therapy includes acute phase reactants. The objective of this clinically oriented literature review was to examine and compare the validity of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) in ankylosing spondylitis (AS) clinical trials. A MEDLINE search was performed covering the years 1967 through April 1998. AS studies were identified and selected if they included ESR and/or CRP and either presented data about their relation with disease activity or were designed as longitudinal clinical trials. Additional studies were identified by scrutinizing references cited in the retrieved studies. The selected studies were examined for truth (association with disease activity), discriminative power (sensitivity to change and discrimination between active and inactive treatment in longitudinal clinical trials), and feasibility (e.g., applicability and costs) of ESR and CRP in AS. We identified 12 articles on the association of ESR and/or CRP with disease activity and 13 longitudinal clinical trials reporting ESR and/or CRP data. Although the applied definitions or disease activity proved very inhomogenous, there was some evidence that both acute phase reactants are correlated with disease activity. In terms of discriminative capacity the available data are inconclusive. Relevant feasibility aspects are general availability, technically simple measurement, and an advantage in the cost of ESR and central laboratory facilities for CRP. Acute phase reactants do not comprehensively represent the disease process in AS. Their worth in AS clinical trials is limited. Based on the currently existing data neither measure is clearly superior in terms of validity. When selecting an acute phase reactant, feasibility aspects may be most relevant in choice of measure.