Plasmapheresis therapy in rheumatoid arthritis. A controlled, double-blind, crossover trial.

Twenty-six patients with rheumatoid arthritis (average age, 57 years; average duration of disease, 11 years) who were unresponsive to antiinflammatory and slow-acting antirheumatic drug therapy were entered into a controlled, double-blind, crossover study to assess the efficacy of plasmapheresis therapy. All patients received 10 true and 10 sham aphereses as outpatients and continued to take their usual drugs. Twenty patients completed the study, and six were withdrawn--three because of poor venous access. Standard clinical and laboratory measures were assessed by personnel blinded to the therapy administered. Paired t-test analysis of seven clinical measures failed to show significant differences between the true and sham procedures (P = 0.36 to 0.96), although transient, mild improvement did occur during both cycles of apheresis, probably because of a placebo response. Significant reductions in the erythrocyte-sedimentation rate, rheumatoid factor titer, and levels of hemoglobin, IgM, and C3 occurred only with true therapy (P = 0.001, 0.01, 0.03, 0.045, and 0.005, respectively). We conclude that plasmapheresis does not have clinical benefit in chronic rheumatoid arthritis, in spite of impressive laboratory changes.