367 Epidemiology of Barrett’s Esophagus in Cirrhosis in the United States: A Population-Based Study

Abstract

INTRODUCTION: The risk of progression of Barrett's Esophagus (BE) increases with worsening liver function. Cirrhosis has high rates of GERD, but data are limited on the prevalence of BE in cirrhosis. We aimed to study the epidemiology of BE in cirrhosis and the underlying associations METHODS: We queried a commercial database (Explorys Inc, Cleveland,OH) that contains an aggregate of electronic health record data from 26 major integrated US healthcare systems. We identified patient cohort between 2014-2019 with a Systematized Nomenclature of Medicine-Clinical Terms (SNOMED-CT). Patients with BE in the absence of cirrhosis were the control group. Patients with a new diagnosis of BE after at least 1 year of cirrhosis were the study group. We performed analyses to describe age, gender and race-based distributions and potential associations RESULTS: Of the 36,893,970 patients in the database between 2014-2019, 208,630 (0.57%) patients with cirrhosis and 179,610 (0.49%) patients with BE were identified. One year after the diagnosis of cirrhosis, 1,500 (0.72%) patients were diagnosed with BE for the first time. When compared to patients without cirrhosis, those with cirrhosis were 1.5 times more likely to develop BE [Odds Ratio (OR) 1.5, 95% confidence interval (CI) 1.46-1.61; P < 0.0001]. The odds of developing BE in patients with cirrhosis was higher in males vs females [OR 1.4, 95% CI 1.26-1.55; P < 0.0001], African American vs Caucasians [OR 1.36, 95% CI 1.08-1.70; P < 0.0082] and young adults (age < 65years) vs elderly (age > 65years) [OR 1.43; 95% CI 1.29-1.58; P < 0.0001] as compared to patients without cirrhosis. When compared to patients without cirrhosis, patients with cirrhosis and BE were more likely to have a history of smoking, Helicobacter pylori infection, GERD, non-steroidal anti-inflammatory use, obstructive sleep apnea and be in the 6th decade of life. Body mass index carried an increased risk of developing BE in cirrhosis [ORs in Table]. Prevalence rates of BE in cirrhosis over the past 5 years is noted in Figure 1. CONCLUSION: Limitations-Endoscopic and histology data are unavailable. Logistic regression analysis was not performed, thus the risk of confounding bias remains. Conclusion-This is the largest study to evaluate BE in cirrhosis. We found a rising trend over the past 5 years in the prevalence of BE in cirrhosis with 1.5 times increased occurrence. This may be a true association or attributed to increased health utilization by patients with cirrhosis or increased endoscopic evaluation by physicians