Twenty patients with chronically progressive multiple sclerosis (MS) were randomised in a double-blind controlled study to assess the efficacy of plasma exchange therapy. All patients were immunosuppressed with prednisone and azathioprine and underwent either plasma exchange or sham apheresis. The 10 patients in each group were similar in age, sex, duration of disease and degree of disability. Clinical and laboratory responses were assessed immediately following the course of exchange or sham therapy, and 3 to 6 months later, by individuals blinded to the type of therapy administered. Although modest improvement was suggested on clinical examination in 7 of 10 patients exchanged and 3 of the 10 sham treated group, this was transient and was not accompanied by any change in disability status scores. No differences in abnormal laboratory investigations were demonstrable between the two patient groups following therapy. We conclude that plasma exchange therapy using this protocol is unlikely to be of clinical benefit as an adjunct in the management of chronically progressive M.S.