BACKGROUND It is unclear whether smoking affects the course of Graves ophthalmopathy and therapeutic outcomes. OBJECTIVE To observe smoking behavior in a randomized study of the effect of radioiodine therapy on ophthalmopathy and in a case series of patients with Graves ophthalmopathy receiving orbital radiation therapy and glucocorticoids. DESIGN Randomized, single-blind study of smoking and mild ophthalmopathy after radioiodine therapy (study 1) and a retrospective cohort study of the association between smoking and response of severe ophthalmopathy to treatment (study 2). SETTING University medical center. PATIENTS 300 patients with mild ophthalmopathy (study 1) and 150 patients with severe ophthalmopathy (study 2). INTERVENTION In study 1, patients received radioiodine alone or radioiodine and a 3-month course of oral prednisone (initial dosage, 0.4 to 0.5 mg/kg of body weight per day). In study 2, patients received high-dose oral prednisone for 6 months (initial dosage, 80 to 100 mg/d) and underwent orbital radiation therapy by linear accelerator (cumulative dose, 20 Gy per eye over 2 weeks). MEASUREMENTS Degree of ophthalmopathy was assessed by overall evaluation (inflammatory changes, proptosis, extraocular muscle dysfunction, corneal involvement, and optic neuropathy). RESULTS In study 1, ophthalmopathy progressed in 4 of 68 nonsmokers (5.9% [95% CI, 3% to 9%]) and 19 of 82 smokers (23.2% [CI, 13% to 33%]) who received radioiodine alone (P = 0.007). Ophthalmopathy was alleviated in 37 of 58 nonsmokers (63.8% [CI, 51% to 78%]) and 13 of 87 smokers (14.9% [CI, 10% to 26%]) who received radioiodine plus prednisone (P < 0.001). In study 2, 61 of 65 nonsmokers (93.8% [CI, 90% to 98%]) and 58 of 85 smokers (68.2% [CI, 57% to 78%]) responded to treatment (P < 0.001). CONCLUSIONS Cigarette smoking increases the risk for progression of ophthalmopathy after radioiodine therapy and decreases the efficacy of orbital radiation therapy and glucocorticoid therapy.