56 female patients with breast carcinoma and treated with polychemotherapy: adriamycin, vindesine, ciclophosphamide and 5-fluorouracil were studied, divided in three groups with different antiemetic schedule: group A 40 mg/iv of metoclopramide; group B 40 mg iv of metoclopramide and 125 mg/iv of methylprednisolone before beginning chemotherapy; and group C 2 mg/oral of lorazepam, 125 mg/iv methyl-prednisolone and 1 mg/kg/oral of metoclopramide previously to begin chemotherapy; at two and four hours of the first dose, metoclopramide was repeated the same doses (really) three doses each two hours. The combination of methylprednisolone-metoclopramide (B), decreases significantly the intensity, duration and frequency of nausea and vomiting, achieving total protection (no vomiting) in 4.34% of cycles of chemotherapy in group A, 32.2% in group B and 30.2% in group C (p less than 0.005 group A versus group B and A versus group C). Also we observed, without statistical analysis, better subjective tolerance to chemotherapy in patients receiving lorazepam associated to schedule B (schedule C). This makes it the recommended schedule.