Objectives: Medications that may increase gastroesophageal reflux could be risk factors for esophageal adenocarcinoma; however, epidemiologic studies present conflicting results. We evaluated patients with a high-risk condition, Barrett's esophagus, to identify risk factors that may act early in the carcinogenic process.
Methods: We conducted a nested case-control study within a large integrated health-services organization. Electronic databases were used to identify incident diagnoses of Barrett's esophagus (cases); two controls were matched to each case. Electronic databases provided information on the use of medications that may induce reflux (nitrates, calcium channel blockers, xanthines, benzodiazepines, and beta agonists) and potential confounders. A supplemental mailed questionnaire evaluated additional potential confounders.
Results: We identified 421 cases and selected 842 controls. The association between any medication use and a Barrett's esophagus diagnosis was modified by age; an increased risk was observed only among subjects <70 yr of age (adjusted odds ratio [OR] = 2.6; 95% confidence interval [CI] 1.5-4.6). A Barrett's esophagus diagnosis was associated with asthma medication use (OR 5.8; 95% CI 2.2, 14.9), but not with the other medications studied. Subgroup analyses suggested that medication use was not independently associated with reflux symptoms and that adjustment for asthma symptoms substantially reduced the association between medication use and a Barrett's esophagus diagnosis.
Conclusion: The use of medications that may induce reflux was associated with a Barrett's esophagus diagnosis among younger persons. This association was only observed with asthma medications; the analyses suggested the possibility of confounding by indication, whereby reflux may cause both asthma and Barrett's esophagus.