Background and aim: Upper gastrointestinal endoscopy is generally accepted as the gold standard for the clinical evaluation of gastric cancer (GC). However, the efficacy of endoscopic screening for asymptomatic GC remains controversial. The present study is designed to clarify the efficacy of endoscopic screening for the detection of early GC by investigating the clinicopathological features.
Methods: A total of 17 522 patients who had underwent endoscopic screening as a part of their annual health checkup at the Seirei Center for Health Promotion and Preventive Medicine between April 2002 and March 2006 were enrolled in this study. We investigated the clinicopathological findings of GC detected by endoscopy. Furthermore, in accordance with the screening interval at our center, patients with GC were categorized into two groups: group A, patients with repeated endoscopic screening within the last 2 years, and group B, patients without endoscopic screening within the last 2 years.
Results: Thirty-nine GC (mean age of patients: 62.2 +/- 8.0 years, 36 males and three females) were detected in total (0.22%). The proportion of early GC was 87.2%. Notable differences between groups A and B were not found in the rate of early GC (P = 0.6342). However, eight of 27 cases (29.6%) in group A were treated by endoscopic resection, but none in group B (P = 0.0344). In six of 26 cases (23.1%) in group A, the recorded images from the previous endoscopic examination indicated some macroscopic abnormalities at the same location, suggesting GC or premalignant lesions.
Conclusion: Endoscopic screening is useful for detecting GC at the early stages, and repeated examinations at short-time intervals contribute to the detection of resectable lesions by endoscopy. Further studies are needed to decrease the false negative rate of endoscopic screening.