Esophageal food bolus obstruction: Natural history, management, and outcomes.

Abstract

Background and Aim: Esophageal food bolus obstruction (EFBO) is a common presenting problem to emergency departments worldwide. There is a paucity of data relating to efficacy of pre-endoscopic and endoscopic management, resulting in a wide variety of clinical practice. Consensus guidelines do not advocate strongly for pre-endoscopic medications, but they remain commonly used. The aim of this project was to evaluate the natural history, management, and outcomes of patients presenting with EFBO and to establish the efficacy of pre-endoscopic and endoscopic treatment options. Method(s): This project was a retrospective audit of patients who were admitted with a food bolus impaction at two large tertiary centers from 2010 to 2021. International Classification of Diseases (ICD-10) codes were used to identify adult patients aged >= 18 years admitted with EFBO. Medical records were reviewed to determine baseline characteristics, medical and endoscopic treatments, and outcomes. A subgroup analysis compared patients who passed the EFBO without endoscopic intervention with patients who required endoscopy. A multivariate logistical regression analysis was performed. Result(s): A total of 675 patients (median age, 55 years; 67% male) were included. Of these, 222 patients (33%) had a prior history of EFBO, 119 (17%) had gastroesophageal reflux disease, and 90 (13%) had asthma. No differences in baseline characteristics were noted between patients who spontaneously cleared the obstruction and those who required endoscopy. Median length of stay was 2 days (IQR, 1-2). There were 432 patients (64%) who received a mean of two pre-endoscopic medications. Glucagon (46%), glyceryl trinitrate (36%), and Coca Cola (37%) were most commonly administered. No medication significantly reduced the need for endoscopic intervention. Multivariable analysis showed that tolerating saliva (odds ratio [OR], 1.95; P = 0.025) and liquids (OR, 2.28; P = 0.001) significantly predicted spontaneous resolution of the EFBO without endoscopy. A total of 461 patients (68%) required endoscopy, with a median time to endoscopy of 12.7 h, and 37% were intubated. The EFBO was most often located in the lower esophagus (32%). Endoscopic complications were rare, with 20 patients sustaining partial esophageal tears and five having aspiration pneumonia. The most common endoscopic findings were esophagitis (24%) and esophageal strictures (12%), with two esophageal tumors identified. Esophageal biopsy samples were taken in 158 cases (34%), and 36% showed eosinophilic esophagitis. Conclusion(s): Medical management is commonly trialed in patients with EFBO, but efficacy is poor. Patients' clinical symptoms, in particular the ability to tolerate saliva or liquids, are the best predictor for spontaneous EFBO clearance. Most patients with EFBO require endoscopy, which is highly effective and has a low complication rate.