354 Cumulative Increase in the Prevalence of Esophageal Cancer in Patients With Gastroesophageal Reflux Disease (GERD) and Multiple Risk Factors

Abstract

INTRODUCTION: The 2015 ACG clinical guidelines on Barrett’s esophagus (BE) recommend endoscopic screening for men with chronic and/or frequent GERD and two or more risk factors for BE or esophageal adenocarcinoma (EAC). However, the additive effect of these risk factors is not known. Our objective was to determine both the individual and combined effect of multiple risk factors on the prevalence of BE associated neoplasia. METHODS: Data were obtained from a commercial de-identified patient database (Explorys, IBM, Inc.) that integrates electronic health records from 26 major U.S. hospital systems from 1999 to May 2019. We identified adult patients (≥18 years) with chronic GERD for ≥5 years (having a GERD diagnosis from May 2014 to May 2019) using Systematized Nomenclature of Medicine - Clinical Terms (SNOMED-CT). Controls were defined as any patient in the Explorys database with chronic GERD, whereas cases were defined as those with chronic GERD and one or more additional BE/EAC associated risk factors per ACG guidelines such as male sex, smoking, age > 50 years, and Caucasian race. Odds ratios (OR) and confidence intervals (CI) for the ORs were calculated for the risk of association of each individual risk factor and with multiple risk factors combined. RESULTS: A total of 4,424,810 patients had GERD for ≥5 years. When compared to GERD controls, cases with GERD and obesity (BMI >30) had a 1.4 (95% CI 1.24 – 1.55) times higher odds of a primary neoplasm of the lower esophagus. Addition of each individual risk factor as described in Table 1 shows the incremental effect of these risk factors on development of primary neoplasm of the lower esophagus. With any three out of four risk factors (chronic GERD and age >50 + any three out of four associated risk factors), the odds of having a primary neoplasm of the lower esophagus increased to 6.26 (5.77 – 6.79) when compared to GERD controls. CONCLUSION: The incremental and strong association of multiple risk factors with primary neoplasm of the lower esophagus suggests a role for screening endoscopies when clinicians encounter GERD patients with multiple associated risk factors. Although this was a retrospective analysis that relied on diagnosis coding, and despite other limitations of Explorys such as lack of histology reports, and unavailability of family history of BE/EAC, our study further validates the utility of the 2015 ACG Barrett’s guidelines in everyday clinical practice.