New York University Langone Medical Center New York, NY
Melissa Magrath, MD, Margaret Vallely, BS, Abraham Khan, MD, Rita M. Knotts, MD, MS; New York University Langone Medical Center, New York, NY
Introduction: Esophageal food impaction (EFI) is a common initial presentation of eosinophilic esophagitis (EoE). Patients presenting with EFI requiring endoscopic intervention present an opportunity to obtain esophageal biopsies to evaluate for EoE and optimize diagnostic yield before initiation of empiric treatment. We aimed to evaluate practices of esophageal biopsy at time of EFI at our institution and identify missed opportunities to diagnose EoE.
Methods: We performed a single center retrospective chart review on a subset of adult patients from 10/2015 -1/2020 who presented to NYU Langone Health with EFI. Patients who underwent upper endoscopy (EGD) and were found to have a retained esophageal food bolus were included. Those with prior diagnosis of EoE were excluded. Proportions were compared using Chi-square or Fisher’s exact test, and rank sum tests were used to compare continuous variables. Logistic regression was used to assess factors associated with subsequent need for diagnostic EGD. Results: 123 patients with EFI were reviewed, 50 (40.7%) were biopsied at the time of EGD (Table 1). Among those biopsied, a new diagnosis of EoE was found in 52%. Of the patients who did not undergo biopsy at index EGD, 23% underwent repeat EGD at our institution and were found to have a new EoE diagnosis. Biopsies performed during EGD for EFI did appear to decrease the need for repeat procedure (OR 0.63, 95% CI 0.25-1.63), although not statistically significant. Patients with furrows were more likely to be biopsied however the report of other classic endoscopic features of EoE (such as exudates, ring, stricture, edema), was not significantly associated with the rate of biopsy (Table 2). Time of procedure and history of prior EFI also did not appear to influence rate of biopsy. Discussion: Biopsy at the time of EFI is important to obtain a timely diagnosis of EoE. At our center, less than half of patients were biopsied at time of initial EGD for EFI, and over half of those biopsied were diagnosed with EoE. The diagnosis of EoE was missed in 23% patients who were not biopsied at time of EFI, and on subsequent EGD with biopsy were found to have EoE. This delay in biopsy likely leads to unnecessary EGDs and more patients who are unaware of their diagnosis. Future educational initiatives aimed at GI providers are needed to improve rates of biopsies during initial EGD for EFI to reduce the need for additional diagnostic procedures.
Disclosures: Melissa Magrath indicated no relevant financial relationships. Margaret Vallely indicated no relevant financial relationships. Abraham Khan: Medtronic – Consultant. Rita Knotts indicated no relevant financial relationships.