The known antiphlogistic in vitro effect of vitamin E was tested in a double-blind randomized study in patients with osteoarthritis. Fifty-three in-patients with osteoarthritis of the hip (n = 34) or the knee (n = 19) were treated for 3 weeks with 400 mg vitamin E (d-alpha-tocopherolacetate, V, n = 26) or 50 mg Diclofenac (D, n = 27) three times daily. A standardized therapeutic exercise program was performed; local therapy was not permitted. There were no significant differences in the efficacy of the two drugs, although one patient of the V-group refused further treatment after 8 days because of inefficacy. V reduced or abolished the pain at rest in 77% (D in 85%), the pain on pressure in 67% (D in 50%), and the pain on movement in 62% (D in 63%). Both treatments appeared to be equally effective in reducing the circumference of the knee joints (p = 0.001) and the walking time (p less than 0.001) and in increasing the joint mobility (p less than 0.002). Patients (n = 11) with a plasma-alpha-tocopherol increase higher than two standard deviations of the mean value at onset (greater than 25.2 mg/l) seemed to have a more pronounced reduction of pain (eight out of 11 patients) compared with four out of 11 patients with a moderate increase of vitamin E. Side effects occurred in two out of 26 patients with V (7.7%), and in 25.9% during D-treatment. One patient with D therefore stopped the therapy after 9 days.(ABSTRACT TRUNCATED AT 250 WORDS)