THE RELATION OF GASTROESOPHAGEAL REFLUX DISEASE AND ITS TREATMENT TO ADENOCARCINOMAS OF THE ESOPHAGUS AND GASTRIC CARDIA

Objective.-To examine the relationship of gastrointestinal disorders and their treatment to the risk of adenocarcinomas of the esophagus and gastric cardia (AEC). Design.-A medical record-based case-control study, with data collected on a standardized form by a trained abstractor, blind to the case-control status. Setting.-A large prepaid health plan. Subjects.-Case patients were plan members newly diagnosed with histologically confirmed AEC from 1986 to 1992. For each of the 196 eligible case patients, one control was selected who matched for membership at time of diagnosis, sex, year of birth, and duration of membership. Main Outcome Measures.-Association between AEC and history of gastroesophageal conditions and their treatment. Conditional logistic regression procedures were used for calculation of odds ratios (ORs) and corresponding 95% confidence intervals (Cls), with adjustment for race, smoking status, and body mass index. Medications were grouped into H-2 antagonists (cimetidine, ranitidine, famotidine, and nizatidine) and anticholinergics (propantheline bromide, dicyclomine hydrochloride, Donnatal [combination of atropine sulfate, hyoscyamine sulfate, phenobarbital, and scopolamine hydrobromide], and Librax [combination of chlordiazepoxide hydrochloride and clidinium bromide]). Results.-Significant twofold or greater risks of AEC were associated with a history of esophageal reflux, hiatal hernia, esophagitis/esophageal ulcer, and difficulty swallowing. The ORs increased with increasing number of these conditions. Although a fourfold risk was linked to four or more prescriptions for H-2 antagonists, the risk was reduced to 1.5 (95% CI, 0.4 to 5.4) after adjusting for the predisposing conditions. Further analysis revealed that the excess risk was restricted to persons with a history of gastroesophageal reflux and related conditions. No association was observed for overall use of anticholinergics. However, after adjustment for predisposing conditions, ORs decreased with increasing number of prescriptions for anticholinergics (P for trend=.08). Conclusions.-This study provides reassuring findings that use of H-2 antagonists and anticholinergics does not increase AEC risk. It also quantifies the elevated risk of AEC associated with gastroesophageal reflux disease. Further research into reflux disease and the production of premalignant epithelial changes may help elucidate carcinogenic mechanisms and measures aimed at early detection and prevention of AEC.