standardized or a free diet does not influence the results of impedance-pH testing, although the former seems of little help to obtain a high possibility of symptom-reflux association. Our findings demonstrate that the use of US and BF normal values in Swiss/Italian patients with GERD underestimate the presence of abnormal GER. Although partially explained by different cultural and dietary factors, the reasons for the observed differences need to be further explored. Tu1806 A Burning Issue: Defining GERD in Non-Erosive Disease Abraham Khan, Sam Serouya, Michael A. Poles, Morris Traube, Vani Murthy Halahalli Srinivasa, Chien Ting Chen, Liying Yang, Zhiheng Pei, Fritz Francois BACKGROUND: GERD is a prevalent condition that encompasses bothersome symptoms or mucosal injury. It remains challenging to diagnose the condition in individuals with the most common presentation of non-erosive reflux disease. The current diagnostic tools include symptom assessment, endoscopy, and pH-metry. Few studies have evaluated the utility of combining these factors with histology to diagnose GERD in the general population. This study aimed to assess the diagnostic yield of a model with predefined criteria in diagnosing GERD among those with non-erosive disease. METHODS: Individuals presenting for upper endoscopy were prospectively enrolled, and the presence of upper GI symptoms was assessed via a standardized questionnaire. All subjects underwent EGD with lower esophageal biopsies 2 cm above the squamocolumnar junction. A blinded GI pathologist reviewed all biopsies. Esophageal acid exposure and symptom association were assessed off acid suppression therapy via 48-hour pH-metry. Subjects were categorized as GERD if at least two of the following criteria in this model were present: 1) positive symptoms reported on presentation 2) any endoscopically identified LA grade esophagitis 3) histologic pathology 4) esophageal pH ,4 for more than 4.5% of the time. RESULTS: Among the 292 subjects enrolled 26% were women and 74% were men, while the majority were either Caucasian (43%) or African American (27%). Mean age (59 ± 9 years) did not differ according to gender or race/ethnicity. The prevalence of GERD symptoms was 62% and erosive esophagitis was 20%. Data for all diagnostic criteria was complete for 113 subjects and as expected, the prevalence of positive pH studies (pH+) was highest for those with endoscopic esophagitis compared to those without endoscopic esophagitis (65% vs. 41%, p=0.03). However the sensitivity of EGD alone for pH+ was 29% with a specificity of 87%. Regardless of endoscopic findings, no difference in pH+ studies was found based on symptoms or histology alone. Among the 90 subjects with a normal endoscopy who still met the predefined criteria the prevalence of pH+ was significantly higher compared to those who did not meet the criteria (71% vs. 15%, p,0.001). The positive likelihood ratio for this diagnostic approach among individuals with non-erosive disease was 3. In multivariate logistic regression analysis, the model remained significant for GERD with pH+ after controlling for age, gender, BMI, tobacco and alcohol (OR 11, 95% CI 3.0-39; p ,0.001). CONCLUSIONS: Among individuals presenting for general endoscopic evaluation the prevalence of erosive disease is low. A predictive model that includes histology allows for a clinically useful approach in diagnosing GERD in non-erosive disease and can help predict and potentially obviate the need for pH testing. Further assessment should be conducted in diverse populations. Tu1807 Chasing Your Tail: An in-Depth Analysis of Factors Contributing to Prolonged pH Depressions Erik B. Person, Heather Barton, Donald O. Castell Background: Impedance-pH testing accurately determines when a reflux bolus enters and exits the esophageal lumen as well and time of related decrease in esophageal pH. Many patients have long pH "tails" in which esophageal pH persists ,4 after the acid bolus has cleared as noted by impedance. These tails contribute to DeMeester score and should be carefully evaluated. Aim: Define the characteristics and factors influencing pH tails. Methods: This study was a single institution, retrospective analysis. 12 healthy volunteers, 16 patients with total esophageal acid exposure (EAE) , 5%, and 16 patients with total EAE ≥ 5% were studied. Patients with achalasia or previous esophageal surgery were excluded. All individuals were off acid suppression. Tails were measured on impedance-pH testing in seconds (s) from time of reflux bolus clearance in the distal impedance lead and associated fall of esophageal intraluminal pH to ,4 to time of consistent return of pH ≥4. Esophageal bolus entry/exit times and number of swallows performed during time pH ,4 were noted as well as body position (upright (U) or recumbent (R)). Tails with .1 contributing reflux episode occurring before recovery to pH ≥4 were considered "multi-reflux (MR)". EAE was measured both with and without tails contributing. Impedance-manometry studies analyzed % complete impedance bolus clearance for liquid and viscous swallows and lower esophageal sphincter (LES) pressures. Results: 484 U tails and 53 R tails were identified in the 44 individuals. U tails occurred in 52.5% of U reflux events while R tails occurred in 28.3% of R reflux events overall (p ,.001). U tails were more frequent than R tails (mean 1.05 vs. 0.18 tails/hr. (p,.0001)). R tail times were longer than U tail times (mean 629 vs. 161s, S-851 AGA Abstracts p,.01). 447 single-reflux (SR) U tails, 37 MR-U tails, 43 SR-R tails, and 10 MR-R tails were identified. SR-R tails were not significantly longer than SR-U tails. MR-R tails were longer than MR-U tails (mean 2621 vs. 459 s (p ,.01)). MR-U tails were longer than SR-U tails (mean 459 vs. 229s (p ,.01) and MR-R were longer than SR-R (2621 vs. 229s (p ,.05)). Manometric characteristics of complete bolus transit, LES pressures, and time to clear reflux bolus were not different between groups and did not affect tail duration. Average time and # of tails correlated strongly with total EAE, R=0.7 (attached figure). Tail characteristics by population are tabulated below. Conclusions: Average number of tails and tail times is linearly related to EAE. Tails are a major factor in esophageal acid exposure, removing them removes all significance between groups. Recumbent pH tails occur less frequently than upright pH tails but have longer duration. Number of reflux episodes contributes to longer tails. Average duration of pH tails does not appear to be related to motility factors. pH Tail Characteristics By Patient Population Figures 1A and B display lack of correlation of mean pH tail times with LES pressure and % complete bolus transit as noted on impedance-manometry. Figure 1C displays positive correlation of mean pH tail times with esophageal acid exposure. Tu1808 The Evaluation of the Efficacy of Weight Loss in Controlling Symptoms in Patients With Gastroesophageal Symptoms Nicola de Bortoli, Giada Guidi, Edoardo Savarino, Manuele Furnari, Lorenzo Bertani, Riccardo Franchi, Salvatore Russo, Irene Martinucci, Dario Gambaccini, Maria T. Urbano, Massimo Bellini, Santino Marchi BACKGROUND: An association between obesity and gastroesophageal reflux (GERD) symp- toms has been reported in a large number of studies. Weight loss is commonly recommended as part of first-line management of GERD despite the paucity of published clinical trials. AIM: The aim of the study was to evaluate the effect of weight loss in controlling symptoms in a group of patients with GERD. The secondary aim was to evaluate the progressive AGA Abstracts