OBJECTIVE: To determine the current best practice for treatment of infantile spasms in children. METHODS: Database searches of MEDLINE from 1966 and EMBASE from 1980 and searches of reference lists of retrieved articles were performed. Inclusion criteria were the documented presence of infantile spasms and hypsarrhythmia. Outcome measures included complete cessation of spasms, resolution of hypsarrhythmia, relapse rate, developmental outcome, and presence or absence of epilepsy or an epileptiform EEG. One hundred fifty-nine articles were selected for detailed review. Recommendations were based on a four-tiered classification scheme. RESULTS: Adrenocorticotropic hormone (ACTH) is probably effective for the short-term treatment of infantile spasms, but there is insufficient evidence to recommend the optimum dosage and duration of treatment. There is insufficient evidence to determine whether oral corticosteroids are effective. Vigabatrin is possibly effective for the short-term treatment of infantile spasm and is possibly also effective for children with tuberous sclerosis. Concerns about retinal toxicity suggest that serial ophthalmologic screening is required in patients on vigabatrin; however, the data are insufficient to make recommendations regarding the frequency or type of screening. There is insufficient evidence to recommend any other treatment of infantile spasms. There is insufficient evidence to conclude that successful treatment of infantile spasms improves the long-term prognosis. CONCLUSIONS: ACTH is probably an effective agent in the short-term treatment of infantile spasms. Vigabatrin is possibly effective.