A randomized double-blind study was conducted to compare the anti-emetic efficacy of 1 mg/kg vs 2 mg/kg metoclopramide, administered 5 times over a period of 8.5 h in 50 cancer patients treated with platinum-containing regimens (cisplatin, ethylenediamine platinum II malonate or spiroplatin). Twenty-six patients were treated with combination chemotherapy, mainly consisting of cisplatin and 5-fluorouracil. No statistically significant differences were observed with respect to the protection against nausea and vomiting between the two dose levels of metoclopramide. However, in the group of patients who received a high dosage of cisplatin (70-100 mg/m2), or ethylenediamine platinum II malonate (800-900 mg/m2), there was a significant difference in nausea and vomiting between patients who had and those who had not received prior chemotherapy, most probably due to anticipation. No difference in the severity of the side effects was observed with the 2 metoclopramide dose levels. Considerable interpatient variation was present in serum metoclopramide levels. No clear correlation was observed between serum metoclopramide levels and prevention of platinum-induced emesis. We conclude that 1 mg/kg of metoclopramide is just as effective as 2 mg/kg as an anti-emetic agent in patients receiving platinum-based chemotherapy, and that metoclopramide serum levels are not related to the anti-emetic effect.