Steroids are commonly used for the treatment of intractable epilepsy. Deflazacort has shown similar effects to prednisone, but with a less worrying adverse-effect profile. In this study, we first compared the efficacy, safety, and seizure relapse rate of deflazacort versus hydrocortisone in children affected by drug-resistant epilepsies. This was an open, non-blinded, randomized clinical study of 35 children affected by drug-resistant epilepsies. The study lasted 12 months. Group 1 (16 patients) received hydrocortisone for 6 months; group 2 (19 patients) was treated with deflazacort for the entire study period. Drug efficacy and tolerability were evaluated after 6 months of therapy. Seizure relapse rates were evaluated 12 months after the start of the study. After 6 months of therapy, hydrocortisone was effective in 44% of patients (responders, with a decrease in seizure frequency of >50%). Deflazacort was effective in 47% of patients (P=0.9). Adverse events occurred in 37% of patients using hydrocortisone and in none of those using deflazacort (P=0.002). At the end of the study, seizure relapse rate resulted significantly higher in group 1 than in group 2 (P=0.04). Hydrocortisone may be useful in the treatment of severely drug-resistant childhood epilepsies. However, its effects may be transient. Deflazacort should be considered in the therapeutic armamentarium for epileptic encephalopathies. The drug is as effective as hydrocortisone and may be used in therapy for a long period, with a less worrying adverse-effect profile.