A case-control study of endoscopy and mortality from adenocarcinoma of the esophagus or gastric cardia in persons with GERD

Background. This study assessed whether EGD reduces mortality from adenocarcinoma of the esophagus or gastric cardia for patients with gastroesophageal reflux. Methods: A case-control study was performed. A total of 245 incident cases of death caused by adenocarcinoma of the esophagus or gastric cardia (1995-1999) in which reflux was present were identified using Veterans Health Administration databases. A total of 980 controls with reflux but no death from adenocarcinoma were frequency matched for age, gender, and race. The occurrences of EGD from 1990 onward were compared for cases and controls. Logistic regression analysis with adjustment for potential confounding factors was performed. Results: All the subjects were men. Cases were significantly less likely to have had an EGD in the time period of interest as compared with controls (adjusted odds ratio 0.66: 95% Cl [0.45, 0.96], p = 0.03). This negative association was as strong for any EGD performed within 1 to 8 years before diagnosis as for a more recent EGD. However, there were no controls that included esophagectomy and no controls with a nonfatal diagnosis of adenocarcinoma, raising the question of whether EGD and reduced mortality are causally linked. The risk of dying from adenocarcinoma was significantly lower for men with a diagnosis of GERD as an inpatient relative to men in whom the diagnosis was made as an outpatient (adjusted odds ratio 0.21: 95% Cl [0.15, 0.31], p < 0.01). Conclusions: For patients with GERD, performing an EGD is associated with reduced mortality from adenocarcinoma of the esophagus or gastric cardia, but whether this is a causative association remains unclear.