The use of corticosteroids to treat optic neuritis is controversial. At 15 clinical centers, we randomly assigned 457 patients with acute optic neuritis to receive oral prednisone (1 mg per kilogram of body weight per day) for 14 days; intravenous methylprednisolone (1 g per day) for 3 days, followed by oral prednisone (1 mg per kilogram per day) for 11 days; or oral placebo for 14 days. Visual function was assessed over a six-month follow-up period.Visual function recovered faster in the group receiving intravenous methylprednisolone than in the placebo group; this was particularly true for the reversal of visual-field defects (P = 0.0001). Although the differences between the groups decreased with time, at six months the group that received intravenous methylprednisolone still had slightly better visual fields (P = 0.054), contrast sensitivity (P = 0.026), and color vision (P = 0.033) but not better visual acuity (P = 0.66). The outcome in the oral-prednisone group did not differ from that in the placebo group. In addition, the rate of new episodes of optic neuritis in either eye was higher in the group receiving oral prednisone, but not the group receiving intravenous methylprednisolone, than in the placebo group (relative risk for oral prednisone vs. placebo, 1.79; 95 percent confidence interval, 1.08 to 2.95).Intravenous methylprednisolone followed by oral prednisone speeds the recovery of visual loss due to optic neuritis and results in slightly better vision at six months. Oral prednisone alone, as prescribed in this study, is an ineffective treatment and increases the risk of new episodes of optic neuritis.