RESUMEN
OBJECTIVE: To evaluate the sensitivity, specificity and accuracy of the double-contrast upper gastrointestinal series (UGI series) in the diagnosis of gastric cancer. MATERIAL AND METHOD: The authors retrospectively reviewed the radiographs of the double-contrast UGI series for diagnosis of gastric cancer of 84 patients who had pathological confirmation by gastric biopsies and/or operations. RESULTS: The lesions were true positive in 45 patients, true negative in 28 patients, false positive in 8 patients and false nagative in 3 patients. The authors found pathologically proved gastric cancers in 48 patients. They were 42 adenocarcinomas (87.5%), 3 gastric mucosa-associated lymphoid tissue lymphomas (6.2%), one gastrointestinal stromal tumor (2.1%), one squamous cell carcinoma at the distal esophagus involving the cardia (2.1%) and one cancer of the pancreatic head with invading gastric antrum (2.1%). The sensitivity, specificity and accuracy of the present study were 93 per cent, 77 per cent and 86 per cent, respectively. They showed small differences compared to the previous studies. CONCLUSION: The present findings indicate that the double-contrast UGI series has high sensitivity, specificity and accuracy. It should be considered a sensitive technique and initial examination in the diagnosis of gastric cancer.