A multicenter, double-blind, randomized, placebo-controlled clinical trial was conducted to determine the safety and efficacy of thalidomide for treating esophageal aphthous ulceration in persons infected with human immunodeficiency virus (HIV). Twenty-four HIV-infected patients with biopsy- confirmed aphthous ulceration of the esophagus were randomly assigned to receive either oral thalidomide, 200 mg/day, or oral placebo daily for 4 weeks. Eight (73%) of 11 patients randomized to receive thalidomide had complete healing of aphthous ulcers at the 4-week endoscopic evaluation, compared with 3 (23%) of 13 placebo-randomized patients (odds ratio, 13.82; 95% confidence interval, 1.16-823.75; P = .033). Odynophagia and impaired eating ability caused by esophageal aphthae were improved markedly by thalidomide treatment. Adverse events among patients receiving thalidomide included somnolence (4 patients), rash (2 patients), and peripheral sensory neuropathy (3 patients). Thalidomide is effective in healing aphthous ulceration of the esophagus in patients infected with HIV.