Abstract
We defined cases of gastric cancer detected in patients annually undergoing screening as false-negative cases, and evaluated the following: 1) clinicopathological findings of cases of gastric cancer detected in patients annually undergoing screening, 2) contents of false-negative cases by re-evaluation of images in the previous year, and 3) comparison of the screening conditions and years of gastroscopy experience of physicians between the previous screening and the screening at the time of detection. All cases of gastric cancer in patients annually undergoing screening were in the early stage, and about 60% of them received gastroscopic treatment. Cancer was often overlooked in the lesser curvature of the M and L areas for the differentiated type and in the greater curvature of the M region for the undifferentiated type. About 50% could have been detected by more careful observation. The &ldqo;quality of examination&rdqo; in terms of the time required for examination, imaging/recording, and examination contents was better, and physicians more experienced in gastroscopy performed examination in the screening at the time of detection. When all cases of gastric cancer detected in patients annually undergoing screening were defined as false-negative cases, the false-negative rate was 33.9%. When only cases in which the lesion could be confirmed by the re-evaluation of images were defined as false-negative cases, the false-negative rate was 15.7%. High quality examination should be performed with consideration given to the considerable number of false-negative cases by gastroscopy.
