S1004 Predictors of Low Esophageal Biopsy Rates During Endoscopy for Food Impaction and Resultant Delayed Diagnosis and Missed Cases of Eosinophilic Esophagitis in a National Veteran Population

Introduction: Best practice guidelines recommend obtaining esophageal biopsies during an index food impaction to evaluate for eosinophilic esophagitis (EoE). We performed a retrospective study using a national sample to evaluate rates of esophageal biopsy during an index foreign body/food impaction and assess outcomes related to EoE diagnoses. Methods: We utilized national Veterans Affairs (VA) data to identify adults ≥18 years of age diagnosed with a first esophageal foreign body (FB) (International Classification of Diseases [ICD]-9 and -10 codes) or food impaction (FI) (ICD-10 codes) between January 1, 2008 (first ICD codes for EoE) and December 31, 2023. Diagnoses of esophageal cancer were excluded. Descriptive statistics characterizing biopsy rates and time to EoE diagnosis were performed. Multivariable regression analyses were used to evaluate factors associated with obtaining biopsies. Results: Four thousand five hundred forty-seven unique patients in the Veterans Health Administration (median age 68, 96% men, 81% White) underwent an esophagogastroduodenoscopy (EGD) for an index esophageal FB/FI. Esophageal biopsies were collected in 28.5% (2,153/7,547) of these patients. Ten point three percent (778/7,547) of patients had biopsies documented from multiple esophageal levels. Biopsy frequency was significantly higher following the EoE guidelines in 2018 (31.1%) compared to previous years (26.7%) (P < 0.0001). Ten percent (755/7,547) of the cohort had a diagnosis of EoE: 38% diagnosed at the time of index FB/FI, and 62% after presentation. Median delay in EoE diagnosis was 103 days from index FB/FI (Q1-Q3: 39-473 days). 43.1% (3,256/7,547) of the total cohort never had documented esophageal biopsies collected during their lifetime in the VA system. In multivariable analyses, younger age, performing an “overnight” vs “daytime” endoscopy, male vs female sex, and lower VA facility complexity were associated with obtaining biopsies. Race, ethnicity, and use of propofol during the procedure did not significantly impact biopsy rates. Conclusion: In a national sample, less than 1 in 3 patients presenting with an index FB/FI and who underwent an EGD had esophageal biopsies obtained during the procedure. Older patients and women were at higher risk for not having biopsies collected. More than half of EoE cases were diagnosed after this index presentation, with a median delay in diagnosis of 103 days. Almost 1 in 2 FB/FI patients never had esophageal biopsies collected, suggesting hundreds of EoE cases may have been missed. Adherence to the best practice recommendation of obtaining esophageal biopsies during index FI may improve EoE diagnosis timing and initiation of treatment.