Proton Pump Inhibitor And Nonsteroidal Anti-Inflammatory Use And The Development Of Neoplasia In Barrett'S Esophagus

Purpose: The role of chemoprevention in Barrett's esophagus (BE) is unclear. Evidence suggests a protective role for proton pump inhibitors (PPIs), non-steroidal anti-inflammatories (NSAIDs), and possibly statins. However, human data are limited. Methods: This is a retrospective study of a well characterized large cohort of patients with documented BE diagnosed between 1985 and 2005. Prescription information was collected from pharmacy records before and after BE diagnosis. Patients were followed until the development of dysplasia, adenocarcinoma, death, or 12/2005. Cox regression analyses were performed to examine the association between NSAID, PPI, or statin prescription and the risk of developing dysplasia or cancer. Results: We examined 408 patients with BE with a mean age of 61 at the time of BE diagnosis; Caucasian 91.2%, men 94.4%. The mean duration of follow-up was 6.6 years (SD 4.9). During 2690 patient-year follow-up, 125 developed dysplasia (20 high grade) yielding an incidence of 4.65 per 100 PY and 29 patients developed adenocarcinoma (1.08 per 100 PY). Approximately 38.4% were prescribed NSAIDs for a mean duration of 12.9 months, 66.4% were prescribed a PPI for a mean duration of 31.5 months, and 26.2% were prescribed a statin for a mean duration of 10.5 months. In unadjusted analyses, only patients with BE segment > 3 cm, and more recent time of BE diagnosis were associated with increased risk of dysplasia or cancer, whereas PPI prescription was associated with reduction in that risk. This persisted in multivariable analysis (Table), and were exaggerated in analysis limited to those developing dysplasia or cancer after the first year of diagnosis; for example, PPI use (0.23, 95% CI: 0.10–0.61). No consistent associations were observed for NSAID or statin use where neither any prescription nor prescriptions > 12 months was associated with the risk of dysplasia or cancer. Conclusion: PPI use seems to reduce the risk of neoplastic changes in patients with BE. NSAID or statin use is not associated with the risk of neoplasia.Table: Multivariable COX Ph model predicting the risk of dysplasia