Twenty patients with classical rheumatoid arthritis were enrolled in a study to determine the effects of piroxicam therapy on neutrophil function as defined by chemotaxis and superoxide anion (O2-) production. T-lymphocyte chemotaxis was also evaluated in these patients. Leukocytes were obtained from these subjects initially, after two weeks of placebo treatment, and subsequently after four and 10 weeks of piroxicam therapy (20 mg, once daily). Responses were compared to simultaneously tested normal controls and to the patient's own cells obtained at the different time points. Studies showed that four and 10 weeks of piroxicam therapy resulted in significantly suppressed neutrophil O2- production in response to phorbol myristate acetate (PMA) and formyl methionyl leucyl phenylalanine (FMLP). O2- production in response to opsonized zymosan was not significantly affected after four weeks of therapy, but was significantly reduced after 10 weeks of therapy when compared to the patient's own cell response after two weeks of placebo treatment. Unlike O2- production, PMN random migration and chemotaxis in response to C5a or FMLP did not differ significantly from normal or untreated patient controls. Analysis of T-lymphocyte migration showed that T-cell random migration or migration to the chemokinetic agent, casein, was not significantly altered by piroxicam therapy. However, when T lymphocytes were tested for chemotaxis in response to lymphocyte-derived chemotactic factor for T cells (LCF), T cell migration was significantly suppressed after 10 weeks of therapy. Furthermore, when T cells from these subjects were cultured for 24 h, random migration was significantly reduced after four and 10 weeks of piroxicam therapy when compared to the patient prior to therapy, and migration in response to LCF was suppressed after four weeks of therapy when compared to normal controls. These data indicate that in patients with rheumatoid arthritis, treatment with piroxicam will significantly suppress PMN O2- production and may also alter the locomotor capacity of T lymphocytes. These actions may contribute to the antiinflammatory effects of piroxicam.