Date: February 2nd, 2017

Reference: Bodkin et al. Effectiveness of glucagon in relieving esophageal foreign body impaction: a multicenter study. AJEM June of 2016

Guest Skeptic: Meghan Groth (@EMpharmGirl) is an emergency medicine pharmacist. She has contributed to the Academic Life in Emergency Medicine (AliEM) and Emergency Medicine PharmD blogs, and is a member of the ALiEMU Capsules team.

Case: It’s the day after St. Patrick’s Day, and you have a 28-year-old man that presents to the emergency department with difficulty swallowing. He was enjoying a feast of corned beef and cabbage several hours prior to arrival when he said it felt like “something got stuck” in his throat. He does have a history of esophageal strictures, and this has happened to him before but he thought if he washed the food down with enough Guinness he wouldn’t have a problem. 

Background: Esophageal foreign body impaction (EFBI) is the most common foreign body impaction seen in emergency departments in the United States and accounts for about 75% of cases. Although estimates vary, the majority of foreign body impactions are meat boluses. In addition, the clinical presentation of these cases is complicated by the presence of baseline esophageal pathology, ranging anywhere from 30 to 90% (Weant and Weant, Eisen et al, Sodeman et al, Al-Haddad et al and Tibbling et al).

Current guidelines for the management of foreign body impaction include endoscopic treatment options like food extraction and advancement of the bolus into the stomach, pharmacologic interventions such as glucagon, and plain old observation or expectant management (Eisen et al).

Guidelines on the management of ingested foreign bodies state that glucagon may be used in the setting of food bolus impaction while endoscopic therapy is being coordinated due to the fact that it is relatively safe but its use shouldn’t delay definitive endoscopic removal (Eisen et al).

Glucagon exerts its effect by relaxing smooth muscle and lowering esophageal sphincter resting pressure (Christiansen et al). Glucagon monotherapy is frequently used to manage food bolus impaction despite having variable success rates (ranging from 9% to 38%) (Al-Haddad et al) and a financial cost.

Glucagon administration also has the potential to cause adverse effects such as hypersensitivity reactions, hypotension, nausea, vomiting, and dizziness (Eli Lilly and Company). 

Clinical Question: Is glucagon safe and effective for the management of esophageal foreign body impaction?

Reference: Bodkin et al. Effectiveness of glucagon in relieving esophageal foreign body impaction: a multicenter study. AJEM June of 2016

  • Population: Any adult or pediatric patient with EFBI
  • Intervention: Glucagon administration
  • Comparison: Patients who did not receive glucagon for EFBI
  • Outcome: Efficacy was defined as resolution of symptoms within 60 minutes after administration of glucagon. Patients who vomited within 60 minutes of receiving glucagon were considered not successful.

Authors’ Conclusions: Glucagon-related resolution occurred in 14.2% of patients and was not significantly different compared with those that did not receive glucagon (10.3%). Concomitant medication administration was associated with lower success. Overall, glucagon had a low success rate, was related to adverse effects, and does not offer advantages for treatment.

checklistQuality Checklist for Observational Studies:

  1. Did the study address a clearly focused issue? Yes
  2. Did the authors use an appropriate method to answer their question? No. A better way to evaluate the efficacy and safety of glucagon would be to do a randomized control trial.
  3. Was the cohort recruited in an acceptable way? Yes. They included all adult and pediatric patients presenting to the emergency department that received glucagon for an EFBI.
  4. Was the exposure accurately measured to minimize bias? Yes. Patients were captured by retrospective review of automated dispensing cabinets and ICD-9 codes.
  5. Was the outcome accurately measured to minimize bias? Unsure. Only one abstractor who was not blinded to the hypothesis and no inter-rater reliability performed.
  6. Have the authors identified all important confounding factors? No. There could have been other confounders not captured (ex. Type of food stuck and previous EFBI).
  7. Was the follow up of subjects complete enough? Yes.
  8. How precise are the results? Unsure. They did not report 95% confidence intervals around their point estimates.
  9. Do you believe the results? Yes.
  10. Can the results be applied to the local population? Yes.
  11. Do the results of this study fit with other available evidence? Yes. Other studies have reported a low success rate in using glucagon for EFBI. See the table below created by Dr. Salim Rezaie from REBEL EM.


Key Results: This study identified 127 patients who received IV or IM glucagon in the study period, accounting for 133 doses of glucagon.

Most patients were male (67%) with a median age of 35 years (range of 2 to 89 years). Underlying esophageal abnormalities were present in 17% of patients. Food was the most common EFBI and 84% of patients required endoscopy.

The control group was made up of 29 patients who did not receive glucagon for their EFBI. These patients were about 20 years older (median age 35.5 glucagon vs. 55 control) and received less concomitant medication (56% glucagon vs. 7% control).

No difference in resolution of symptoms within 60 minutes

  • Primary Outcome: No difference 14.2% (glucagon) vs. 10.3% (control) p=0.59
  • Adverse Events: Vomiting occurred in 16/127 (12.6%) of patients who received glucagon.

Of the patients who did not experience resolution of their symptoms after glucagon, a large majority (n=92 or 84.4%) required endoscopy; 89.7% of control patients required endoscopy. When endoscopy was performed, no major adverse events were reported.

Screen Shot 2015-04-25 at 3.11.12 PM

1) Low Success Rate: The success rate in this trial for glucagon was only about 14%, compared to several other trials where the success rate for both glucagon and placebo arms hovers around 30%. The authors explain the potential reason for this difference as lying in their definition for efficacy as resolution of symptoms within 60 minutes after administration of drug, where other studies have defined success as resolution of symptoms at any point during the emergency department visit.

This seems like an appropriate method to be able to attribute the success to the glucagon, but if other medications were co-administered (as they were in more than 50% of the glucagon patients), it might be difficult to say that the resolution of symptoms was definitely due to the glucagon versus one of the other interventions or just spontaneous resolution.

2) Confounding Variables: Some confounding variables were collected (age, sex, type of foreign body, esophageal abnormality and concomitant medication) while others were not. They only described the type of EFBI in terms, of food, coins, etc.

There is literature to suggest different types of food may matter. For example, meat causing an EFBI is less likely to respond to glucagon, probably because of its rigid structure (Sodeman et al).

The authors also failed to describe whether or not patients had experienced an EFBI in the past. I think they tried to get at this by describing underlying esophageal abnormalities, but again it seems like a previous visit for an EFBI is predictive of success/failure (Sodeman et al).

3) Vomiting: Patients who vomited within 60 minutes of glucagon administration were not deemed successful. This was because it is not the proposed mechanism of action of glucagon for the relief of EFBI. However, vomiting is a known side effect of glucagon and they reported 16 patients (13%) receiving glucagon vomited. How many of those had resolution of the EFBI?

Also, even though the authors commented that the intervention (glucagon) had the potential to result in adverse effects, they didn’t describe this in the control group. One might assume that this meant that 16 patients (or 13%) in the glucagon group vomited compared with zero in the control group, but this wasn’t described. Therefore, it’s not clear that they can actually draw this conclusion.

4) Small Numbers and Unknown Precision: The number of patients in the control group was small, only 29. Even though the authors didn’t find a difference between the groups with regard to symptom resolution, there was no discussion of how large of a sample size they would have needed to detect a certain difference between groups (e.g. power discussion). However, they did try to increases their numbers by performing this at two sites.

The other point is the precision of the point estimate. They provided the percentage of patients who had resolution of their EFBI with glucagon and in the control group but they did not give a 95% confidence interval. This means we do not know the precision of the point estimate.

5) Retrospective Chart Review (Observational Study): This is the biggest limitation of this study. We can only conclude associations. What we would like to see is a multicenter, blinded, randomized control trial, with a standardized protocol, properly powered with a patient oriented outcome and an intention to treat analysis to determine the efficacy and safely of glucagon in treating EFBI.

One of the author, Dr. Kyle Weant (EM Pharmacist), was not available to join us in recording the SGEM episode but kindly sent some comments.

  • “So the challenge with this study was that everyone gets glucagon, that’s why the control group was so small. We would have loved more but current practice prohibits this unfortunately. The time frame was chosen largely based on the change in CPOE systems and so we were limited by that in acquiring data. That being said, it still ended up being one of the largest trials of this therapy. There was no discussion of power because this was a retrospective trial whose patient numbers were dictated by time, not enrollment. Also, it would be challenging to do a power calculation that would be accurate because it relies on existing data on the incidence of success of this therapy, which we don’t really have a good handle on. This is definitely a limitation and it could be that we didn’t have enough patients. That being said, if we don’t find a difference over a multiyear, multisite investigation, one could suggest that any difference that does exist isn’t clinically relevant.”

Comment on authors conclusion compared to SGEM Conclusion: We generally agree the authors’ conclusions.

SGEM Bottom Line: Glucagon has a low success rate for EFBI, does not seem to offer much benefit over observation alone and is associated with adverse events like vomiting.

Case Resolution: You discuss with the patient that glucagon is sometimes used to treat esophageal foreign body impactions, but it doesn’t work much better than just observation alone and could result in nausea or vomiting. You call the gastroenterologist consultant to arrange for an emergent endoscopy for your patient if the impaction doesn’t resolve spontaneously.

Meghan Groth

Meghan Groth

Clinical Application: Glucagon doesn’t appear to offer additional benefit over expectant management for EFBI, may potentially result in adverse effects, and also has costs associated with it. The data makes sense in that it generally fits with what else is out there, but it doesn’t seem like it is strong enough to change practice on its own. Endoscopy for either management or diagnosis of underlying abnormal esophageal anatomy should be performed.

What do I tell my patient? You probably have food stuck in your throat. Glucagon is a medicine that has been used to get the food unstuck. However, there is not good evidence that it works any better than just watching you for a bit. Glucagon also has side effects like vomiting. Our plan is to watch you and if the food is still stuck we will call a specialist. They will come and talk about putting a tube down your throat with a camera on the end the tube to see what is wrong and fix the problem.

Keener Kontest: We had lots of correct answers to the keener contest last time. The first person to correctly respond was Dr. Elaine Blau from Tobermory, Ontario. She new Hippocrates (ca. 460-370 BC) first introduced the term sepsis. It is derived from the Greek word sipsi (“make rotten”).

Listen to the podcast for this weeks’ question. If you know the answer then send an email to TheSGEM@gmail.com with “keener” in the subject line. The 1st correct answer will receive a cool skeptical prize.

Other FOAMed Resources:


  • Weant KA, Weant MP. Safety and efficacy of glucagon for the relief of acute esophageal food impaction. Am J Health Syst Pharm. 2012 Apr 1;69(7):573-7).
  • Eisen GM, Baron TH, Dominitz JA, et al. American Society for Gastrointestinal Endoscopy. Management of ingested foreign bodies and food impactions. Gastrointest Endosc. 2011 73(6): 1085-91.
  • Sodeman TC, Harewood GC, Baron TH. Assessment of the predictors of response to glucagon in the setting of acute esophageal food bolus impaction. Dysphagia. 2004 Winter;19(1):18-21.
  • Al-Haddad M, Ward EM, Scolapio JS, et al. Glucagon for the relief of esophageal food impaction does it really work? Dig Dis Sci. 2006 Nov;51(11):1930-3.
  •  Tibbling L, Bjorkhoel A, Jansson E, et al. Effect of spasmolytic drugs on esophageal foreign bodies. Dysphagia. 1995 Spring;10(2):126-7
  • Christiansen J, Borgeskov S. The effect of glucagon and the combined effect of glucagon and secretin on lower esophageal sphincter pressure in man. Scand J Gastroenterol. 1974;9(7):615-8
  • Eli Lilly and Company. Glucagon (rDNA origin) for injection prescribing information. Indianapolis, IN; 1999 Feb

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.

EMA 2017 courses