Colorectal Neoplasia in Veterans Is Associated with Barrett's Esophagus but Not with Use of Proton-Pump Inhibitors (PPIs) or Aspirin/NSAIDs

Abstract

Purpose: An association between Barrett's esophagus (BE) and colorectal neoplasia has been suggested, but is still controversial. We investigated whether BE was associated with an increased risk of colorectal neoplasia, and, if so, whether BE was an independent risk factor or dependent on the use of PPIs or aspirin/NSAIDs. Methods: We performed a case-control study and matched 268 veterans with histologically-proven BE (mean length: 2.0 cm, range: 0.5-10 cm, all intestinal metaplasia) with 268 controls without BE. Controls had undergone upper GI endoscopy close (+/- 14 days) to the time of the corresponding case. Colonoscopy was performed within 6 months of the upper GI endoscopy. Charts were reviewed and scored for patient characteristics (age, gender, ethnicity, and body mass index (BMI)), tobacco or alcohol use, and use of PPIs and aspirin/NSAIDs for at least 6 months prior to upper GI endoscopy, and findings at colonoscopy. Statistical analysis was done by t-tests, chi-square tests and logistic regression analysis. Results: Mean age was 66 +/- 11 years in the BE group and 64 +/- 12 years in the control group. The majority were men (BE group 99% vs. controls: 97%) and white (74% vs. 72%). Mean BMI was similar (28 +/- 6 vs. 29 +/- 6) as was smoking (40% vs. 34%), but alcohol use was more common in the BE group (66% vs. 46%, p < 0.001). BE patients used more often PPIs (54% vs. 25% (p < 0.001), but less often aspirin/NSAIDs (50% vs. 61%, p=0.007). Colorectal adenomas or carcinoma were present in 160 /268 (60%) patients of the BE group and in 105/268 (39%) controls (p<0.001). The total number of polyps (BE group vs. controls: 1.8 vs 1.9) and the mean size of the polyps (0.60 vs. 0.55 mm) were similar in those with polyps. Increasing age (OR 1.23 per decade, 95% CI (1.02, 1.49)), use of alcohol (OR 1.76: 95% CI (1.19, 2.60)), and BE (OR 1.93: 95% CI (1.29, 2.91)) were associated with an increased risk of developing colorectal neoplasia, when statistically adjusted for each other and other potential risk factors in multivariate analysis. Conclusion: These results suggest that veterans with BE are at an increased risk of developing colorectal neoplasia. This association is independent from the use of PPIs or aspirin/NSAIDs.