Assessment of the predictors of response to glucagon in the setting of acute esophageal food bolus impaction.

Esophageal food impactions are frequently seen in endoscopic practice. Glucagon is known to relax the lower esophageal sphincter and has been used with variable success to treat food impactions. We retrieved clinical information of all patients with acute food impactions who attended the emergency room from 1975 to 2000 from the Mayo diagnostic database. Data were abstracted on age, sex, body mass index, relevant prior medical history, food type ingested (meat, bread, vegetable, or other), duration of symptoms at presentation, dosage (in mg) of glucagon, outcome including success of glucagon or spontaneous passage, and endoscopic findings. A total of 222 cases of food impaction were identified, of whom 106 patients (48%) received glucagon, average 1 mg. In glucagon responders, meat was less likely to be the offending food type, accounting for 70% (glucagon responders) vs. 90% (in nonresponders) ( p = 0.03), while responders were less likely to have esophageal rings/strictures detected on subsequent EGD compared with nonresponders, 0% (glucagon responders) vs. 31% (nonresponders) ( p = 0.05). In the patients that did not receive glucagon, spontaneous resolvers had a shorter duration of symptoms at presentation, 3.3 h vs. 12.4 h ( p = 0.07) and were less likely to have an organic esophageal obstruction detected on EGD, 0% vs. 21%. There were no significant differences between the resolvers and nonresolvers in terms of age, gender, BMI, and prior medical history. Conservative management of acute food bolus obstruction, either with or without glucagon, is most successful in the absence of a fixed esophageal obstruction. An impacted meat bolus is more likely to require intervention for removal than other food types. These clinical predictors should be considered before administration of glucagon.