The aims of our study were to: determine the effect of metoclopramide parenterally and orally on delayed gastric emptying of a radionuclide test meal in symptomatic patients with diabetic gastroparesis not explained by ulceration or other mechanical problems; and evaluate in a double-blind crossover fashion the efficacy of metoclopramide in relieving the symptoms of diabetic gastroparesis. Thirteen patients with subjective evidence of gastric stasis had delayed gastric emptying of an isotope-labeled semisolid meal which was significantly accelerated (p less than 0.05) after 10 mg of metoclopramide parenterally. Patients then received metoclopramide 10 mg and placebo before meals and prior to retiring for 3 weeks in a randomized double-blind crossover design. During metoclopramide therapy nausea, vomiting, anorexia, fullness, and bloating were significantly (p less than 0.05) ameliorated compared to placebo with an overall mean symptom reduction of 52.6%. Gastric emptying studies after completion of the trial is seven patients, subjectively improved and receiving open-labeled metoclopramide, showed significantly less gastric retention. Individual improvements in gastric emptying after parenteral or oral metoclopramide, however, could not be correlated with symptom change during the treatment trial. We conclude that metoclopramide is an important therapeutic adjunct in the management of diabetic gastroparesis and its therapeutic effects are mediated through its prokinetic properties as well as centrally mediated antiemetic actions.