Abstract

Objective To explore the feasibility of high-resolution micro-endoscopy for diagnosis of gastric cancer. Methods A retrospective analysis of HRME pictures of gross specimen of gastric carcinoma of 20 cases of gastric cancer and normal gastric mucosa was made. Picture characteristics were studied and the HRME diagnostic criteria for normal gastric mucosa and gastric cancer tissues were developed. Based on the diagnostic criteria, a prospective study on gastric biopsy specimens of 64 cases of suspected gastric carcinoma was conducted and the diagnostic value of HRME for gastric carcinoma was evaluated by comparing HRME pre-diagnostic results with pathological findings. Results In the superficial layer of the fundic mucosa, numerous regularly branched arranged glands as well as oval or elongated openings of the gastric pits and linear peripheral cracks were visible; the nuclei were arranged regularly. In the superficial layer of the antral mucosa, irregular or tubular openings of the gastric pits and cracked glandular cavities were visible, with the cells surrounding the gastric pits regularly arranged and the nuclei small and densely distributed. In the gastric carcinoma, nuclei sizes were different, arranged messy. Gland sizes were various, with unclear structure or glandular structures that could not be observed. Structure of gastric pits was damaged and normal gastric pits disappeared. After HRME imaging was performed on 64 cases of gastric biopsy specimens, the sensitivity, specificity, accuracy, negative predictive value and positive predictive value of HRME diagnosis of gastric carcinoma were 96.4%(53/55), 88.9%(8/9), 95.3%(61/64), 98.1%(53/54) and 80.0%(8/10) respectively. Conclusion HRME, a new and instant pathological imaging tool with low price and simple operation, can distinguish normal gastric mucosa and gastric cancer tissue clearly, with a high diagnosis value for gastric cancer. Key words: High-resolution micro-endoscopy; Stomach neoplasms; Diagnosis

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call