AGA Abstracts 665 Persistent Epithelial Changes in Inactive Eosinophilic Esophagitis: Is Inactive Really Inactive? Bridget Godwin, Kelly Whelan, Alain Benitez, Andres Klein-Szanto, Jonathan Spergel, Gary W. Falk, Amanda B. Muir, Hiroshi Nakagawa Background and Aims: Eosinophilic esophagitis (EoE) is a chronic, antigen and immune- mediated esophageal disease characterized by symptoms such as dysphagia and food impac- tion. Histologic changes of EoE include an eosinophil count of ≥15 eosinophils per high- power field (eos/hpf) and basal cell hyperplasia (BCH) in the esophageal mucosa. EoE is considered to have transitioned from active to inactive when the eosinophil count drops below 15/hpf. We aimed to determine whether BCH is normalized at <15/hpf peak eosinophil count, and to correlate prevalence of eosinophils with persistent epithelial changes and symptoms in inactive EoE at the time of biopsy. Methods: Histology reports were reviewed from an IRB-approved database containing clinical, endoscopic, and pathology information regarding 761 pediatric and 139 adult esophageal biopsies. Patient ages ranged from 1- 67 yrs old, with 72% percent of patients being male. The pediatric database included 257 active EoE, 216 inactive EoE, and 288 non-EoE. The adult database included 71 active EoE, 36 inactive EoE, and 32 non-EoE. The database included peak eosinophil counts, the frequency of BCH noted on pathology report, and symptoms. A subset of these biopsies was sent to an independent pathologist for scoring of BCH as further validated by immunohistochemistry (IHC) for p63, a basal cell marker. Results: BCH scoring by an independent pathologist revealed 30% (n=10) of the pediatric and 56% (n=9) of the adult biopsies with inactive EoE had BCH as corroborated by a higher number of p63-positive basal cells. The average BCH score was significantly higher (p=0.03) in patients with inactive EoE (n=19) than in non-EoE (n=14). Out of the 216 inactive EoE in our pediatric database, 17% had BCH based upon the original pathology report. Eosinophil count in pediatric inactive EoE was significantly higher (p=<0.0001) with persistent BCH (n=36, average eos/hpf = 7.3 +/- 0.6) than without (n=180, average eos/hpf = 2.1 +/-0.25). Of the inactive pediatric EoE patients showing BCH, 50% (n=36) had symptoms such as heartburn, dysphagia, regurgitation and abdominal pain in contrast to 11% (n=180) of inactive patients without BCH (p=<0.0001). Conclusions: Although our current EoE criteria define inactive disease as <15 eos/hpf present on esophageal biopsies, there continues to be a high prevalence of BCH in inactive patients. These changes are significantly correlated with the number of eosinophils that continue to be present on biopsy, suggesting that a lower eosinophil count than 15 may be necessary to truly define inactive disease. Further studies regarding basal cell hyperplasia and it's role in inflammation and barrier defect will further characterize the importance of re-defining inactive disease. S-136 AGA Abstracts 666 Mucosal Impedance Measurements at Index Endoscopy Can Diagnose EoE Without the Need for Histology Tina Higginbotham, Pooja Lal, James C. Slaughter, Michael Vaezi INTRODUCTION: Eosinophilic esophagitis (EoE) is a prevalent condition among young patients presenting with dysphagia and has emerged as a leading cause of esophageal morbidity among children and adults over the last 20 years. Diagnosis is based on symptom of dysphagia and presence of both distal and proximal esophageal eosinophilia [ ≥15 eosino- phils (eos)/hpf]. Esophageal biopsy is the current gold standard in the diagnosis of EoE; however, histologic variability due to patchy infiltration of eos may result in diagnostic errors. Mucosal Impedance (MI) is a minimally invasive innovative test which detects changes in the conductivity of esophageal epithelium and recent studies suggest a unique MI pattern in EoE compared to GERD. The aim of this study was to assess the predictive value of MI as a diagnostic modality for the evaluation of EoE at index endoscopy without the need for esophageal biopsy. METHODS: Patients with dysphagia undergoing routine esophageal endoscopy were enrolled in this prospective blinded study. Distal and proximal esophageal biopsies were obtained and MI values [ohms ( Ω)] were collected at 2, 5 and 10cm from squamocolumnar junction. EoE was predicted (pre-esophageal biopsy) based on low MI values (<2000 Ω) and pattern of MI along the esophageal axis (<2000 Ω along the esophagus). Non-EoE MI pattern was defined by either normal (>2000 Ω) or low MI values (<2000 Ω) in distal esophagus but normal values proximally along esophageal axis. A member of the research team (TH), blinded to endoscopic view and final diagnosis, predicted EoE or non- EoE patient status. Predictions were later compared to surgical pathology reports to determine level of concordance between the blinded prediction and histopathologic results. EoE diagno- sis was established based on current guidelines of ≥ 15 eos counts/hpf from both distal and proximal esophagus. RESULTS: Twenty patients with dysphagia (10 with EoE and 10 non- EoE) constituted the study population [mean (SD) age=33 (28-41) (IQR) BMI= 26 (23-30); 65% male, 100% White]. Demographics were similar between the groups except the non- EoE patients were older [age=44 (32-58)]. Figure 1 demonstrates the MI values and pattern in patients predicted and confirmed to have EoE compared to those without EoE. MI predicted EoE in 10/10 (100%) of patients histologically confirmed to have EoE and predicted non-EoE in 9/10 (90%) of patients. MI had excellent test characteristics (sensitivity-100%, specificity-90%, PPV-91%and NPV-100%) in predicting EoE without the need for histology. The positive likelihood ratio for EoE diagnosis based on MI values and pattern was 10.00 (1.56-64.20). CONCLUSION: MI during endoscopy accurately predicts EoE status in patients with dysphagia. MI in EoE may increase diagnostic efficiency by providing the diagnosis at index endoscopy thus eliminating the need for histology. 667 Moderate or High Dose Proton Pump Inhibition Trial is Associated with Better Delineation of PPI-REE Diagnosis than Standard Dose Alison Goldin, Wai-Kit Lo, Matthew Hamilton, Karen S. Hsu Blatman, Jason Hornick, Walter W. Chan Background: PPI-responsive esophageal eosinophilia (PPI-REE) is the subset of esophageal eosinophilia with clinical and histologic response to proton pump inhibitor (PPI) therapy. However, PPI-REE may be clinically and endoscopically indistinguishable from eosinophilic esophagitis (EoE), and the mechanism of PPI responsiveness in PPI-REE remains unclear. While current consensus guidelines suggest a PPI trial in all patients with esophageal eosinophilia to distinguish between EoE and PPI-REE, the optimal dose of PPI is unknown. Aim: To compare the prevalence of PPI-REE diagnosis after a trial of standard, moderate, and high dose PPI therapy. Methods: This was a retrospective cohort study of patients diagnosed with esophageal eosinophilia (>15 eos/hpf) on mucosal biopsies from upper endoscopy (EGD) at a tertiary care center in 5/2006-7/2015. After initial EGD, all included patients underwent PPI trial for ≥8 weeks in one of three dosing arms: standard (total daily dose: omeprazole 20 mg or equivalent), moderate (total daily dose: omeprazole 40 mg or equivalent) or high (total daily dose: omeprazole 80 mg daily or equivalent). All patients underwent repeat EGD after PPI trial and were classified based on histologic response on biopsies into PPI-REE (<15 eos/hpf) vs EoE (>15 eos/hpf). Patient demographics, clinical symptoms, allergy history, and endoscopic findings were recorded. Fisher-exact test for binary variables and student t-test for continuous variables were used to assess differences between cohorts. Multivariate analysis was performed using forward stepwise logistic regres- sion. Results: 122 patients (mean age: 42, 56% male) with esophageal eosinophilia were