In a prospective randomized trial we compared the antiemetic efficacy of metoclopramide (MCP) and dexamethasone (DXM). A total of 172 patients receiving chemotherapy based on cisplatin 100 mg/m2/day were randomly allocated to four groups. CP was administered in 5 doses of 1.5 mg/kg (group A), 2 mg/kg (group C), or combined with 4 doses of DXM 4 mg/6 h (MCP: 1.5 mg/kg in group B; 2 mg/kg in group D). Vomiting was decreased in the DXM groups compared to groups A and C (p < 0.002), and the duratoin of nausea was greater in group A than in group D (P < 0.004). A greater disturbance in appetite was observed in group A compared to groups B (p < 0.028), D(p < 0.001) and C (p < 0.045). Activity problems were greater in group A than in C (p < 0.003) and D(p < 0.005). We noticed that a small increase in MCP (0.5 mg/kg) did not influence the antiemetic effect. The addition of DXM did not significantly alter the antiemetic results at the lower MCP dose, but improved them when MCP was slightly increased. We conclude that DXM slightly improves the antiemetic effect of MCP and the effects are related to the MCP dose.