To evaluate the efficacy of steroid therapy in epileptic encephalopathies and refractory epilepsies other than West syndrome. Retrospective analysis of treatment and outcome data of patients treated with steroids for epileptic encephalopathies. Outcome was analyzed in terms of the patient's epileptic syndrome, age at seizure onset, duration of seizure disorder, age at treatment, abnormalities noted on neurological examination, developmental status and type of steroid. Fifty-eight consecutive patients from six pediatric neurology outpatient clinics were included in the study. The epileptic syndrome and types were as follows: myoclonic epilepsy (n = 16), epilepsy with continuous spike waves during slow-wave sleep (n = 12), partial epilepsy (n = 10), Landau-Kleffner syndrome (n = 6), generalized epilepsy with multiple epileptic seizure types (n=5), Rasmussen's encephalitis (n = 3), Lennox-Gastaut syndrome (n = 2), atypical absence (n = 2) and spasms (n = 2). Sixteen patients (27%) had received adrenocorticotrophic hormone (ACTH) and 42 (7%) had received prednisone (22 of the latter patients were given initial pulse methylprednisolone). Complete cessation of seizures as well as disappearance of epileptic discharges in electroencephalogram were achieved in 16 patients (28%); 15 other patients (26%) had improvement of >50%. Of the 31 patients with >50% improvement and follow-up of at least 1 year, 16 (52%) subsequently relapsed. ACTH was found to be significantly more efficacious than prednisone in controlling seizures and improving function. Initiation of treatment with high-dose methylprednisolone did not increase efficacy. ACTH may be more efficacious than prednisone in treating epileptic encephalopathies and refractory epilepsies other than West syndrome.