The objective of this study was to compare the efficacy of corticotropin (ACTH) versus vigabatrin in treating infantile spasms and to determine which medication has a more favorable long-term outcome in terms of cognitive function, evolution of epilepsy, and incidence of autism. Patients with infantile spasms were included in the study if they were 3 to 16 months old, had hypsarrhythmia, and had no previous treatment with vigabatrin or corticosteroids. Patient evaluation included electroencephalographic and psychometric measures before and after treatment. Patients were stratified based on etiology (idiopathic or symptomatic) and sex and then randomized between the ACTH and vigabatrin treatment groups. Each of the treatment groups received either ACTH or vigabatrin for 2 weeks. At the end of 2 weeks of treatment, patients were considered responders if spasms and hypsarrhythmia resolved. Nonresponders were crossed over and treated with the alternate drug. Nine patients were included in the study. Three patients received ACTH, one of whom was a responder. Six patients received vigabatrin, three of whom were responders. The five nonresponders received both therapies. All patients had some degree of developmental plateau or regression before the initiation of treatment. Four patients with idiopathic infantile spasms showed improved cognitive function following treatment. The remaining five patients remained significantly delayed. Five patients with symptomatic infantile spasms had epilepsy following treatment; three of them were in the autistic spectrum. The small number of infants in this pilot study is insufficient to determine which of the two drugs is more effective. However, the following trends were identified: vigabatrin may be more effective for patients with symptomatic infantile spasms; patients with idiopathic infantile spasms tend to have a better cognitive outcome; and patients with symptomatic infantile spasms tend to develop both epilepsy and autism.