In a study of the antiemetic effectiveness of high-dose oral metoclopramide, 66 previously untreated patients receiving 60 mg/m2 cisplatin were entered into a double-blind randomized trial. Patients were stratified according to age and tumor type, then randomized to receive either oral or intravenous (IV) high-dose metoclopramide. Patients were evaluated for antiemetic protection, toxicity, affect (anxiety, hostility, and depression), and autonomic arousal (pulse rate and blood pressure) at three-hour intervals on the day of their chemotherapy. Metoclopramide serum levels were measured by high-performance liquid chromatography. Results indicated no significant differences between the oral and IV groups on any measurement of antiemetic protection, affect, or autonomic arousal. There were also no significant differences in side effects except for frequency of stools; patients who received oral metoclopramide had significantly more stools than patients who received IV metoclopramide. The mean (+/- SD) serum metoclopramide level at four hours achieved orally was 1,171 +/- 660 ng/mL; the mean (+/- SD) level achieved IV was 1,030 +/- 392 ng/mL (P = .498). We conclude that high-dose oral and IV regimens of metoclopramide as administered in this study have equivalent antiemetic efficacy in previously untreated patients receiving 60 mg/m2 cisplatin.