Influence of Central Obesity and Reflux on Esophageal Injury: A Prospective Study: Presidential Poster

Abstract

Purpose: Both central obesity and gastroesophageal reflux (GER) have been implicated in the esophageal injury, inflammation and neoplasia pathway. We have previously shown an association of visceral abdominal fat area with esophageal inflammation and neoplasia. Central obesity may influence esophageal injury via mechanical (promoting reflux) and non-mechanical (systemic inflammatory) means. In this study, we aimed to assess contributions of: 1.) BMI and central obesity to acidic (AR) and non-acidic reflux (NAR), and 2.) Obesity and GER to esophageal injury and metaplasia. Methods: Patients undergoing ambulatory pH testing and upper endoscopy were prospectively recruited. Patients had anthropometric measurements (height, weight, waist circumference (WC), waist to hip ratio (WHR)) followed by biopsies from the GEJ junction (columnar) and esophageal (squamous) mucosa. Data on AR and NAR were prospectively collected. The biopsies were analyzed for tissue PGE2 (marker of tissue injury) and CDX2 (marker of intestinal metaplasia). Univariate and multivariate linear regression analyses were performed to assess associations. Results: One hundred patients were recruited. Mean age (SD) was 50.89 (16.1) years and 67% were females. Thirty-five percent patients underwent wireless (Bravo) pH monitoring and 59% conventional pH impedance. Six percent had endoscopic findings of esophagitis, thereby eliminating need for a pH study. Thirty-eight percent patients were off PPIs, and 77% consented to biopsies. Both central obesity and BMI were independently associated with AR (Table 1). Only central obesity was a predictor of NAR. BMI (p=0.0011) and central obesity (WC, trend, p=0.059) appeared to influence supine and NAR. Obesity and reflux do not independently predict tissue injury, as measured by PGE2 in the squamous and GEJ epithelium (Table 2). ASA or PPI use was not associated with PGE2. CDX1 and CDX2 were not detected in the squamous mucosa. At GEJ, CDX1 was positive in 7% of patients and CDX2 in 51%. There was no correlation with reflux or obesity to CDX2.Table 1: Multivariate linear regression analyses of the predictors of refluxTable 2: Multiple variable linear regression analyses of tissue biomarkersConclusion: Central obesity, as measured by WC or WHR, is an independent predictor of AR and NAR in non erosive reflux disease. However, neither reflux nor central obesity was associated with esophageal injury (measured by tissue PGE2) or molecular markers of intestinal metaplasia (CDX2). Specific measures such as abdominal visceral fat area or volume may need to be investigated as predictors of esophageal tissue injury and metaplasia.