Objectives: To compare the short-term prognosis of patients with severe acute rheumatic carditis when treated with an intravenous pulse of methylprednisolone in comparison with conventional treatment using oral prednisone. Methods: We designed a randomized clinical trial in the setting of a university general hospital in Brazil. We randomly allocated 18 patients with the diagnosis of severe acute rheumatic carditis and congestive heart failure to receive an intravenous pulse as opposed to oral prednisolone. Methylprednisolone was administered in a dose of 1 g intravenously for 3 consecutive days in the first and second weeks, for two days in the third, and one day in the fourth week. Prednisone was administered in a dose of 1.5 mg/kg/day over the period of 4 weeks. Results: The mean age of the patients was 11.1 ± 3.7 years, with a median of 12 years. Patients on oral treatment showed a more pronounced decrease in the heart rate, sedimentation rate, and in the titres of C-reactive protein than those receiving intravenous therapy. At the end of treatment, a mild decrease in the left ventricular end-systolic dimension was found in those having oral treatment, compared to an increase in the group having intravenous treatment (p = 0.036). The ejection fraction showed a median increase of 5% in those undergoing oral treatment, and a median decrease of 6% in the group with intravenous therapy (p = 0.009). There were 5 therapeutic failures in those receiving intravenous therapy (56%), including 1 death. Therapeutic failures were not observed in those treated orally (p = 0.03). Conclusion: Intravenous treatment of methylprednisolone, as a single anti-inflammatory agent, was inferior to conventional treatment with oral prednisone in the control of severe rheumatic carditis.