ObjectiveTo clarify the usefulness of 18F-FDG PET/CT for detecting primary gastric cancer via gastric distention using a mixture of milk and Diatrizoate Meglumine. Materials and methodsA total of 68 patients (male: 47, female: 21; age: 41–87 years) suspected of gastric carcinoma underwent 18F-FDG PET/CT imaging. After whole-body PET/CT imaging in a fasting state, the patients drank a measured amount of milk with Diatrizoate Meglumine. Local gastric district PET/CT imaging was performed 30min later. The imaging was analyzed by semi-quantitative analysis, standardized uptake value (SUV) of the primary tumor was measured in a region of interest. The diagnosis results were confirmed by gastroscopy, pathology, and follow-up results. ResultsOf the 68 patients, 56 malignant gastric neoplasm patients (male: 37, female: 19) were conformed. The sensitivity, specificity, positive predictive value and negative predictive value of fasting whole-body PET/CT imaging for a primary malignant tumor were 92.9%, 75.0%, 94.5%, and 69.0%, respectively. The values for distension with a mixture of milk and Diatrizoate Meglumine were 91.1%, 91.7%, 98.1%, and 68.8%, respectively. The area under the curve was 0.919±0.033 and 0.883±0.066 for the diagnosis of gastric cancer with SUVmax in a fasting state and after intake of mixture respectively, the differences were not statistically significant (P=0.359). Using gastric distension with a mixture of milk and Diatrizoate Meglumine, the mean ratio of the lesion's SUVmax to the adjacent gastric wall SUVmax increased significantly from 3.30±3.05 to 13.50±15.05, which was statistically significant (P<0.001). Conclusions18F-FDG PET/CT imaging is highly accurate for the diagnosis of primary gastric carcinoma. Gastric distention can display the lesions more clearly, however, it cannot significantly improve diagnostic accuracy.